PreprintPDF Available

IPBES Global Assessment on Biodiversity and Ecosystem Services Chapter 4. Plausible futures of nature, its contributions to people and their good quality of life

  • Institute of Research for Development, Univ Montpellier
Preprints and early-stage research may not have been peer reviewed yet.


This document contains the draft of Chapter 4 of the IPBES Global Assessment on Biodiversity and Ecosystem Services Chapter 4 focuses on scenarios and models that explore the impacts of a wide range of plausible future changes in social, economic and institutional drivers on Nature, Nature's Contributions to People and Good Quality of Life.
Chapter 4.
Copyright © 2019, Intergovernmental Science-Policy Platform on Biodiversity and Ecosystem Services (IPBES)
ISBN No: 978-3-947851-20-1
Yunne-Jai Shin (France), Almut Arneth (Germany), Rinku
Roy Chowdhury (United States of America), Guy F. Midgley
(South Africa)
Elena Bukvareva (Russian Federation), Andreas Heinimann
(Switzerland), Andra Ioana Horcea-Milcu (Romania),
Melanie Kolb (Mexico), Paul Leadley (France), Thierry Oberdorff
(France), Ramon Pichs Madruga (Cuba), Carlo Rondinini
(Italy/IUCN), Osamu Saito (Japan), Jyothis Sathyapalan
(India), Yaw Agyeman Boafo (Ghana), Pavel Kindlmann
(Czech Republic), Tianxiang Yue (China), Zdenka Krenova
(Czech Republic), Philip Osano (Kenya)
Ignacio Palomo (Spain), Zeenatul Basher (Bangladesh/
Michigan State University), Patricio Pliscoff (Chile)
Jesús Alcalá-Reygosa (Mexico), Rob Alkemade
(Netherlands), Peter Anthoni (Germany), Mrittika Basu
(United Nations University), Celine Bellard (France),
Erin Bohensky (Australia), Laurent Bopp (France),
Andrea Buchholz (Canada), James Butler (Australia), Jarrett
Byrnes (United States of America), Tim Daw (Sweden),
Emmett Duffy (United States of America), Mariana Fuentes
(United States of America), Patricia Glibert (United States
of America), Chun Sheng Goh (Japan), Burak Güneralp
(United States of America), Paula Harrison (United Kingdom
of Great Britain and Northern Ireland), Elliott Hazen (United
States of America), Andrew Hendry (Canada), Robert M.
Hughes (United States of America), María José Ibarrola
(Mexico), David Iles (United States of America), Stéphanie
Jenouvrier (France), Jed Kaplan (Switzerland), HyeJin
Kim (Germany), Andreas Krause (Germany), Heike Lotze
(Canada), Isabel Maria Rosa (Germany), Ines Martins
(Germany), Alicia Mastretta-Yanes (Mexico), Zia Mehrabi
(Canada), David Mouillot (France), Elvira Poloczanska
(Australia/IPCC), Thomas Pugh (United Kingdom of Great
Britain and Northern Ireland), Benjamin Quesada
(Germany/Colombia), Laura Sauls (United States of
America), Verena Seufert (Germany), Andrew Sweetman
(United Kingdom of Great Britain and Northern Ireland),
Zachary Tessler (United States of America), Britta Tietjen
(Germany), Derek Tittensor (Canada/UNEP-WCMC),
Boris Worm (Canada)
Rainer Krug (Germany)
Milan Chytrý (Czech Republic)
Shin, Y. J., Arneth, A., Roy Chowdhury, R., Midgley, G.F.,
Leadley, P., Agyeman Boafo, Y., Basher, Z., Bukvareva,E.,
Heinimann, A., Horcea-Milcu, A. I., Kindlmann, P., Kolb, M.,
Krenova, Z., Oberdorff, T., Osano, P., Palomo, I., Pichs
Madruga, R., Pliscoff, P., Rondinini, C., Saito, O.,
Sathyapalan, J. and Yue, T. 2019. Chapter 4: Plausible
futures of nature, its contributions to people and their
good quality of life In: Global assessment report of the
Intergovernmental Science-Policy Platform on Biodiversity
and Ecosystem Services. Brondízio, E. S., Settele, J., Díaz, S.,
Ngo, H. T. (eds). IPBES secretariat, Bonn, Germany.
168pages. Doi: 10.5281/zenodo.3832074
P. 599-600: iStock_Andrea Izzotti
The designations employed and the presentation of material on the maps used in the present report do not imply the
expression of any opinion whatsoever on the part of the Intergovernmental Science-Policy Platform on Biodiversity and
Ecosystem Services concerning the legal status of any country, territory, city or area or of its authorities, or concerning
the delimitation of its frontiers or boundaries. These maps have been prepared for the sole purpose of facilitating the
assessment of the broad biogeographical areas represented therein.
Table of
EXECUTIVE SUMMARY ...................................................604
4.1 INTRODUCTION .....................................................611
4.1.1 Context and objectives of the chapter ................................611
4.1.2 Exploratory scenarios .............................................612
4.1.3 Archetype scenarios ..............................................613
4.1.4 Projected indirect and direct drivers of change in scenarios ...............617 Indirect Drivers (including consideration of diverse values) in scenarios .......................617 Direct Drivers ...................................................................620
4.1.5 Considering Indigenous Peoples and Local Communities (IPLCs)
and indigenous and local knowledge (ILK) in scenarios ..................623
4.2 PLAUSIBLE FUTURES FOR NATURE. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 624
4.2.1 Impacts of future global changes on biodiversity: feedbacks
and adaptation capacity ...........................................624 Projected negative changes at all levels of biodiversity ...................................624 Future biodiversity adaptation and reorganisation .......................................626 The importance of feedbacks between hierarchical levels of biodiversity ......................629
4.2.2 Marine ecosystems ...............................................629 Global state and function of marine ecosystems and future drivers of change .................629 Future climate change impacts on marine biodiversity and ecosystem functioning ..............633 Climate change impacts in open ocean ecosystems .....................................633 Climate change impacts in shelf ecosystems ..........................................637 Climate change impacts in deep seas ................................................640 Climate change impacts in polar seas ................................................641 Future impacts of fisheries exploitation on marine ecosystems .............................642 Future impacts of pollution on marine ecosystems ......................................645 Persistent organic pollutants and plastics: another ‘Silent Spring’? ..........................645 Nutrient loads and eutrophication ...................................................646 Future impacts of coastal development on marine ecosystems ............................647
4.2.3 Freshwater ecosystems ...........................................648 Freshwater biodiversity and current threats ............................................648 Future climate change impacts on freshwater biodiversity and ecosystem functioning ...........649 Future land-use change impacts on freshwater biodiversity and ecosystem functioning ..........651 Future impacts of habitat fragmentation on freshwater biodiversity and ecosystem functioning . . . .652 Future impacts of non-native species on freshwater biodiversity and functioning ...............653 Future impacts of harvest on freshwater biodiversity and functioning ........................654 Future impacts on peatlands .......................................................654
4.2.4 Terrestrial ecosystems ............................................655 Future climate change and atmospheric CO2 impacts on habitats, biodiversity,
and ecosystem state and functioning ................................................655 Climate change impacts on vegetation cover ..........................................655 Climate change impacts on species diversity ..........................................655 The combined impact of atmospheric CO2 concentration and climate change
on projected vegetation cover ......................................................655 Projected changes in ecosystem state and function .....................................656 Future land-use and land-cover change impacts on habitats, biodiversity,
and ecosystem state and functioning ................................................656 Future global ecosystem functioning and biodiversity in strong climate change
mitigation scenarios ..............................................................662 Invasive alien species ............................................................663
603 Pollution impacts on terrestrial ecosystems: Ozone (O3) and Nitrogen ........................663
4.2.5 Challenges in linking biodiversity and ecosystem functioning
at the global level ................................................664
4.3.1 Nature’s contributions to people across scenario archetypes .............665
4.3.2 Changes in nature’s contributions to people . . . . . . . . . . . . . . . . . . . . . . . . . . .666 Nature’s contribution to people – regulating contributions .................................666 Nature’s contributions to people – changes in material contributions ........................669 Nature contributions to people – changes in non-material contributions .....................670
4.3.3 How changes in nature’s contributions to people will manifest in different
regions, including teleconnections across regions ......................674
4.4 PLAUSIBLE FUTURES FOR GOOD QUALITY OF LIFE ......................677
4.4.1 Linking good quality of Life to nature and nature’s contributions to people . . .677 Key Dimensions of good quality of life and their links to nature and nature’s
contributions to people ...........................................................677 Material dimension of good quality of life ..............................................677 Non-material dimensions of good quality of life .........................................681 Good quality of life across worldviews and knowledge systems ............................683
4.4.2 Linking good quality of life to nature and nature’s contributions
to people across future scenarios ...................................683 Mediating factors of future GQL and NCP ............................................683 Future scenarios of GQL and NCP ..................................................685
4.5.1 Analysis of interactions from the Systematic Literature Review ............690
4.5.2 Feedbacks .....................................................691
4.5.3 Trade-offs ......................................................692
4.5.4 Co-benefits .....................................................694
4.5.5 Regime Shifts, Tipping Points and Planetary Boundaries .................695
NATURE’S CONTRIBUTIONS TO PEOPLE ...............................696
4.6.1 How good will we be at reaching international biodiversity and
sustainability targets beyond 2020? ..................................696
4.6.2 How can the evidence from scenarios contribute to the development
of future biodiversity targets and the 2050 vision? ......................702 Habitat loss and degradation (Target 5) ...............................................702 Sustainable fisheries (Target 6). . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . .703 Sustainable agriculture (Target 7) ....................................................704 Vulnerable ecosystems (Coral Reefs) (Target 10) ........................................705 Protected areas and other Effective Area-based Measures (Target 11) .......................706 Preventing Extinctions and Improving Species Conservation Status (Target 12) ................706 Ecosystem Restoration and Resilience (Target 15) ......................................707
DECISION-MAKING ..................................................708
4.7.1 Scenarios and models help prepare decision makers for uncertainty
and long-term thinking ............................................708
4.7.2 Dealing with uncertainty when using scenarios and models to support
decision-making .................................................709
4.7.3 The challenge of spatial and temporal scales in using scenarios and
models to support decision-making ..................................711
4.7.4 Improving communication and building capacity to enhance the use
of scenarios and models in decision-making ...........................713
REFERENCES ...........................................................714
Chapter 4 focuses on scenarios and models that
explore the impacts of a wide range of plausible
future changes in social, economic and institutional
drivers on nature, nature’s contributions to people
(NCP) and good quality of life. The chapter’s assessment
concentrates on studies published since 2008 that cover
large regional to global spatial scales and time periods
from the present to 2050, and up to 2100. This framing of
the assessment means that this chapter is best suited to
help setting the agendas for decision-making at national
to international levels by identifying future challenges
and providing a compelling case for action. Chapter 4
provides new insights compared to previous assessments
by including the most recent scenarios and models, by
examining a broad range of global change drivers and their
interactions, and by highlighting the impacts on a wide range
of indicators of nature, nature’s contributions to people
and good quality of life. Where possible, results are also
interpreted in view of their implications for achieving the Aichi
Biodiversity Targets and the Sustainable Development Goals.
This chapter endeavours to provide a balanced perspective
on drivers of change and their impacts, but the strong
bias in the scenario literature towards climate change
impacts on nature limits the scope to which the chapter
can provide a comprehensive vision of plausible futures
to decision makers. Climate change has been studied
far more extensively than other drivers (such as land use
change, pollution, use and extraction of natural resources,
and invasive alien species), and studies of interactions
between drivers, especially more than two drivers, are
relatively rare (well established) {4.2.1, 4.2.2, 4.2.3, 4.2.4}.
Terrestrial systems are studied more extensively than marine
systems, with a paucity of studies of freshwater systems
(well established) {}. Impacts on biodiversity and
ecosystem function have been the focus of much more
attention than nature’s contributions or good quality of life
(78%, 16% and 5% of literature reviewed, respectively;
(well established) {A1.1}. Among nature’s contributions to
people, material (such as food production) and regulating
contributions (such as carbon dioxide removal from the
atmosphere into ecosystems) are more studied than
non-material contributions in relation to scenarios (well
established) {4.3.1}.
The large majority of the studies covered in this chapter is
based on scenarios developed in support of climate change
assessments (93% of literature reviewed; {4.1.3}, the most
recent of which are the Representative greenhouse gas
Concentration Pathways (RCPs) and their associated Shared
Socio-economic Pathways (SSPs). This has the benefit of
providing strong coherence with climate assessments but
results in biases in terms of drivers of change and socio-
economic processes included in the scenarios. For example,
only few of the scenarios assessed in this chapter explore
mechanisms leading to social or ecological regime shifts
{4.5}. In addition, most scenarios do not explicitly take
into account different worldviews and values associated
with many non-material nature’s contributions to people
and, in general, were not designed to address a wide
range of Sustainable Development Goals {4.5, Chapter
5}. Nonetheless, this chapter recognizes that the different
scenario archetypes hold inherently different worldviews and
values that ultimately drive the scenario outcomes {4.1}.
Participatory scenarios are one means of including a richer
range of processes and values explored, but it is difficult
to extrapolate from the local scale of most participatory
scenarios to the large regional and global spatial scales that
are the focus of this chapter {4.4.2, 4.7}.
1Significant changes at all biodiversity levels –
from genetic diversity to biomes – are expected to
continue under future global changes. Despite
projections of some local increases in species
richness and ecosystem productivity, the overall
effect of global changes on biodiversity is projected to
be negative (well established). Interactions within and
between biodiversity levels can significantly influence
future biodiversity responses to global changes
(established but incomplete). A substantial fraction of
wild species is simulated to be at risk of extinction during
the 21st century due to climate change, land use, natural
resource extraction and impact of other direct drivers (well
established) {4.2.1, 4.2.2, 4.2.3, 4.2.4}. Loss in intraspecific
genetic diversity is expected due to the projected decrease
in species population sizes and spatial range shifts. Genetic
loss should be recognized as a serious threat to future
potential for adapting to global change (established but
incomplete) {,}. Expected species range
shifts, local species extinctions, changes in species
abundances will lead to disruptions of species relations
including disturbance of trophic webs, plant-pollinator and
other mutualistic relations (well established) {4.2.2, 4.2.3,
4.2.4}, that can cascade through the entire ecosystem.
Novel (no-analogue) communities, where species will
co-occur in historically unknown combinations, are
expected to emerge (established but incomplete) {,}. As a consequence, new approaches to
conservation are warranted that are designed to adapt to
rapid changes in species composition and ensuing
conservation challenges. Intraspecific diversity and
interactions between different biodiversity levels need to be
represented in global models and scenarios to improve
future projections of nature {,}.
2In marine ecosystems, most scenarios and
models point towards a global decrease in ocean
production and biodiversity, but the level of impact
can vary widely, depending on the drivers, scenarios,
and regions considered (well established). All
anthropogenic greenhouse gas emission scenarios result in
a global increase in sea temperature, ocean acidification,
deoxygenation and sea level rise (well established) {}.
By the end of the century, these environmental changes are
projected to decrease net primary production (by ca. -3.5%
under the low greenhouse gas emissions scenario, RCP2.6
and up to -9% in the very high emissions scenario, RCP8.5),
and secondary production up to fish (by -3% to -23% under
RCP2.6 and RCP8.5, respectively), as well as top predator
biomass (established but incomplete) {}. Fish
populations and catch potential are projected to move
poleward due to ocean warming (well established) with a
mean latitudinal range shift of 15.5 km to 25.6 km per
decade to 2050 (under RCP2.6 and RCP8.5, respectively)
(inconclusive), leading to high extirpation rates of biomass
and local species extinctions in the tropics (well established)
{}. The rapid rate at which sea ice is projected to
retreat in polar seas, and the enhanced ocean acidification,
imply major changes to be expected in the future for
biodiversity and ecosystem function in the Arctic and
Southern oceans (well established) {}. All
components of the food webs will potentially be impacted,
from phytoplankton to top predators, and from pelagic to
benthic species (established but incomplete).
3Relative to climate change impacts, published
scenarios project that the choice of fisheries
management and market regulation measures can
have the strongest impacts on the future status of
marine fish populations (well established) {}. In the
face of continuous growth of human population that is
projected to reach 9.8 billion (± ca. 0.4 billion) people in
2050 combined with rising incomes, the demand for food
fish will likely increase (well established). Business-as-usual
fisheries exploitation is foreseen to increase the proportion
of overexploited and collapsed species (well established), as
well as species impacted by bycatch {}. Adaptive
fisheries management that responds to climate induced
changes of fish biomass and spatial distribution could offset
the detrimental impacts of climate change on fish biomass
and catch in most RCPs (but RCP8.5)(inconclusive) {}.
4For marine shelf ecosystems, additional future
threats include extreme climatic events, sea level rise
and coastal development which are foreseen to cause
increased pollution and species overexploitation but also
fragmentation and loss of habitats that directly impact the
dynamics of marine biodiversity (well established) {,}. These impacts could potentially feedback to the
climate as coastal wetlands play a major role in carbon
burial and sequestration globally (well established)
{}. In coastal waters, increasing nutrient loads and
pollution in combination with sea warming are expected to
stimulate eutrophication and increase the extent of oxygen
minimum zones with potential detrimental effects on living
organisms (well established) {}. Coral reefs are
projected to undergo more frequent extreme warming
events, with less recovery time in between, declining by a
further 70-90% at global warming of 1.5°C, and by more
than 99% at 2°C causing massive bleaching episodes with
high mortality rates (well established) {}.
5Concerns about rapidly increasing plastic
pollution now match or exceed those for other
persistent organic pollutants. If current production and
waste management trends continue, about 12,000 Mt
(million tons) of plastic waste will accumulate in the
environment by 2050, especially in the ocean which acts as
a sink (established but incomplete). The harmful effects of
plastics have been evidenced at all levels of marine food
webs from plankton to top predators but are not yet
projected into the future {}.
6In freshwater ecosystems, all scenarios and
models point towards a decrease in freshwater
biodiversity and substantial changes in ecosystems
state and functioning, especially in tropical regions
(well established). Freshwater ecosystems cover only
0.8% of the world surface area but host almost 8% of the
world’s species described, making a high contribution to
global biodiversity. Given that all scenarios are based on
continued growth of human population density until 2050,
impacts due to combined anthropogenic drivers on
freshwater biodiversity and ecosystems are projected to
increase worldwide, and to be strongest in tropical regions
where human population growth and biodiversity are
concentrated (well established) {4.2.3}. Increases in land
area used for urbanization, mining, cropland and
intensification of agriculture are projected to boost the risk of
pollution and eutrophication of waters, leading to extirpation
of local populations, changes in community structure and
stability (e.g. algal blooms) (well established) {}, and
establishment and spread of pathogens (established but
incomplete) {}. Under all scenarios, habitat
fragmentation (e.g., damming of rivers) and exploitation are
projected to increase the risk of species extinction with
potential effects on food web dynamics, especially in tropical
regions (well established) {,}. These impacts
on freshwater flows, biodiversity and ecosystems will likely
be exacerbated by climate change, especially under
moderate (RCP4.5) and high emissions (RCP6.0, RCP8.5):
higher temperatures are projected to generate local
population extinctions especially for cold-water adapted
species, and species extinctions in semi-arid and
Mediterranean regions, since the area extent of these
climatic regions will shrink due to projected decrease in
precipitation (increase of estimated extinction rates by ca.
18 times in 2090 under the SRES A2 scenario, compared
with natural extinction rates without human influence)
(inconclusive) {}.
7In terrestrial ecosystems, scenarios and models
point towards a continued decline in global terrestrial
biodiversity and regionally highly variable changes in
ecosystem state and functioning (well established).
Land-use change, and invasive alien species will continue to
cause biodiversity loss across the globe in the future, with
climate change rapidly emerging as an additional driver of
loss that is increasing over the coming decades in relative
importance across all scenarios (well established) {4.2.4}.
Although large uncertainties exist regarding the exact
magnitude of loss, it is well established that increasing
global warming will accelerate species loss {4.2.4}. Already
for relatively minor global warming, biodiversity indices are
projected to decline (established but incomplete) {4.2.4}.
Extinction risks are projected to vary between regions from
5% to nearly 25%, depending on whether a region harbours
endemic species with small ranges or is projected to
experience climate very different from today (inconclusive).
Substantial climate change driven shifts of biome
boundaries, in particular in boreal and sub-arctic regions,
and (semi)arid environments are projected for the next
decades; warmer and drier climate will reduce productivity
(well established) {}. In contrast, rising atmospheric
CO2 concentrations can be beneficial for net primary
productivity of ecosystems, and is expected to enhance
woody vegetation cover especially in semi-arid regions
(established but incomplete) {}. The combined
impacts of CO2 and climate change on biodiversity and
ecosystems remain (unresolved) {}.
8The relative impacts of climate change versus
land-use change on biodiversity and ecosystems are
context-specific and vary between scenarios, regions,
and indicators of biodiversity and ecosystem
functioning (well established) {,}.
Land-use change pressures differ between scenarios, but
managed land area continues to increase, with exception of
some scenarios exploring sustainability trajectories.
Scenarios of large-scale, land-based climate change
mitigation rely on large increases of bioenergy crop area or
large reforestation or afforestation with potentially
detrimental consequences for biodiversity and some
ecosystem functioning (well established) {,,
4.5.2}. Interactions of land-cover change and future climate
change enhance the negative impacts on biodiversity and
affect multiple ecosystem functions (established but
incomplete) {,}. Pressure on biodiversity and
ecosystem function from other drivers such as biological
invasions will likely be accentuated at global scale, as trade
between climatically and environmentally similar regions are
projected to increase, and habitats continue to be disturbed
(established but incomplete). Overall, the small number of
regional to global scale scenario studies that assess
pollution or invasive alien species’ impacts on nature
precludes a robust assessment {,}.
9Many scenarios project increases in material
nature’s contributions to people, which are generally
accompanied by decreases in regulating and non-
material contributions (established but incomplete)
{3.1, 3.2}. The simulated trade-offs between material vs.
regulating and non-material ecosystem services are
especially pronounced in scenarios with strong human
population growth and per capita consumption (established
but incomplete) {4.3.4,, 4.2.4}. Assumptions about
population growth and increase in per capita consumption
are projected to lead to rising demand for material services,
especially food, materials and bioenergy, and are projected
to reduce regulating contributions such as provision of clean
water, pollination, or ecosystem carbon storage (well
established) {4.3.2, 4.3.3, 4.5.3,,, 4.2.3,
4.2.4}. In the long term, substantial decreases in regulating
contributions may have detrimental effects on material
contributions, for example climate change impacts on all
systems will be increased if climate regulation by forests or
oceans is weakened (well established). The future
magnitude of these cascading effects has yet to be
determined (inconclusive). This is because most scenarios
and models do not consider fully the interactions between
multiple drivers and multiple ecosystem impacts, and as a
consequence cannot quantify important feedbacks {4.3.3,
4.3.4, 4.5.1, 4.5.4}.
10 Scenarios examining trends in nature and
nature’s contributions to people show significant
regional variation (well established). The
interconnectedness of the world regions emphasizes
the need for decision-making on ocean, freshwater
and land management to be informed by
considerations of regional trade-offs among nature’s
contributions to people (well established). Future
scenarios show that many regions will experience a general
decrease of biodiversity and many regulating and non-
material ecosystem services, but others will see increases
(well established) {4.2.2, 4.2.4, 4.3.3}. The degree to which
regions differ regarding impacts of global environmental
changes depends on the underlying socio-economic
scenarios, with climate change being an additional driver
(established but incomplete) {4.1, 4.2, 4.3}. Scenarios of a
world with regional political- and trade-barriers (Regional
Competition Scenario) tend to result in the greatest
divergence across regions, scenarios that emphasize liberal
financial markets (economic optimism and reformed market
scenarios) in intermediate levels of disparity, while scenarios
that encapsulate aspects of sustainable development
(Regional Sustainability and Global Sustainability scenarios)
result in more modest differences between regions
(established but incomplete) {4.3.3, 4.2.4}. For example, an
analysis of the impacts of the shared socio-economic
pathway (SSP) scenarios indicates that terrestrial biodiversity
and regulating contributions will be more heavily impacted in
Africa and South America than in other regions of the world,
especially in a regional competition scenario and in an
economic optimism scenario compared to a global
sustainability scenario {4.2.1,}.
Irrespective of the underlying socio-economic assumptions,
spatial telecoupling (socioeconomic and environmental
interactions over distances) implies that increasing future
demand for ecosystem services in certain regions will
affect supply of services in others. Material contributions,
especially food and energy production, play a dominant
role in these telecouplings (well established) {4.2.4, 4.3.3,
4.5.2}. Material contributions tend to be traded between
regions {4.1,,, 4.5.2, 4.6}, but locally
declining biodiversity cannot be replaced by increased
biodiversity in a different location {4.2.2-4.2.4}. If tele-
couplings are not accounted for in future scenarios,
unrealistically overoptimistic responses to a regional political
intervention (e.g., land-based climate mitigation, negative
emission policies, sustainable fisheries management for
local resources and not for imported ones) are assumed,
and measures to reduce detrimental side effects not taken
(established but incomplete) {4.3.3}.
11 Limiting mean global warming to well below 2oC
will have large co-benefits for nature and nature’s
contributions to people in marine, freshwater and
terrestrial ecosystems. Land-based climate change
mitigation efforts offer opportunities for co-benefits,
but if large land areas are required, trade-offs with
biodiversity conservation and food and water security
goals will need to be addressed in terrestrial and
freshwater ecosystems (well established). Climate
warming and ocean acidification associated with increasing
atmospheric CO2 are already causing damage to marine,
freshwater and terrestrial biodiversity (well established)
{4.2.2, 4.2.3, 4.2.4} which confirms the urgency of meeting
the goals of the Paris Climate Agreement. The degree to
which marine and land ecosystems will continue to remove
CO2 from the atmosphere, which at present amounts to
nearly 50% of anthropogenic CO2 emissions, is highly
uncertain {,}. On land, reduction of
deforestation combined with management practices in
cropland, pastures and forests can contribute notably to
greenhouse gas emissions reductions (well established).
Recent cost-effective estimates are between ca. 1.5 and
11Gt CO2eq a-1 over the coming few decades, the
undetermined range depending, amongst others, on which
types of measures are included {4.5.3}. Along coastlines, a
combination of reduced nutrient discharge (mitigating
pollution) and space to allow inland wetland migration
(adapting to sea level rise), is essential to preserve the
capacity of coastal wetlands to sequester carbon
(established but incomplete) {,}.
Regionally, land conversion pressure is large both
in scenarios of high population growth and lack of
sustainability considerations, and in scenarios requiring
land for bioenergy or afforestation and reforestation to
mitigate climate change (established but incomplete)
{4.1,}. Recent projections of an annual carbon
uptake in 2050 projected for bioenergy pathways (with
carbon capture and storage about 0.9-2.2 GtC a-1) and
afforestation/reforestation (0.1-1 GtC a-1) are equivalent
to an additional one third to three quarters of today’s land
carbon sink {}. It remains uncertain whether the
required land area would be available for large bioenergy
plantations or afforestation/reforestation efforts, where
these areas would be located and whether such net
carbon uptake rates can be achieved and maintained
{, 4.5.2}. Likewise, detrimental environmental
and societal side effects have been projected to arise
from strong mitigation scenarios that rely on large area
expansion of managed crop and forested land associated
with intensification of production (established but
incomplete) {,, 4.5.2}.
12 Scenarios repeatedly show that changing food
consumption patterns and reducing waste and losses
in the food system can contribute significantly to
mitigating loss of biodiversity and ecosystem
services. Human population growth over the coming
decades is projected to increase to nearly 9.8 billion (± 0.4
billion) by 2050 and to 11.4 billion (± 1.8 billion) by 2100. As
a consequence of the projected population growth,
continued urbanisation, and changes in many countries’
diets towards increasing per capita animal protein share and
processed food, most scenarios foresee increasing crop
area, and in some cases pasture area as well. These
projected changes in agricultural land area are combined
with intensification of land management and continued
increases in crop yields, that are projected to have
detrimental environmental and biodiversity side effects
associated with agricultural intensification (well established)
{,,, 4.5.2}. An increasing number of
scenarios emphasizes the potential role of consumption as
part of the solutions to overcome these challenges, such as
shifting diets towards a globally equitable supply of nutritious
calories or reducing wastes and losses along the entire
chain from crop production to consumers (well established)
{4.5.4}. Enhancing efficiencies in the food system has large
potential to free up land for other uses such as for
biodiversity conservation. Studies that explore dietary
scenarios of reduced consumption of animal protein
estimate that between ca. 10% and 30% of today’s area
under agriculture may be freed for other purposes, with
possible co-benefits in the form of a globally more equitable
distribution of animal protein intake by humans and
improved health. Reduced greenhouse gas emissions from
the land sector, and reduced irrigation water needs are an
additional benefit, which will also release pressure on
freshwater pollution and biodiversity (established but
incomplete). Nearly one-quarter of total freshwater used
today in food crop production are estimated to be spared if
wastes and losses in the food system were minimized
(inconclusive) {4.3.1, 4.3.2, 4.5.2, 4.5.3}.
13 Societies and individuals within societies value
differently the regulating, material, and non-material
contributions from nature that underpin their quality of
life (well established). In future scenarios governed by
market forces, multiple dimensions of good quality of life are
expected to decline. The decline is particularly pronounced
for indicators related to livelihood and income security
(established but incomplete) {4.4.1, 4.4.2}. Market-based
and regionally-fragmented scenarios, associated with
growth in population and consumption, indicate continuous
deterioration of nature to support economic growth, with
some regions affected more than others. Without
decoupling economic growth from unsustainable extraction
and uses, scenarios show continuous decline in nature’s
contributions to people. Scenarios exploring sustainability or
reformed financial market pathways are projected to result in
improved good quality of life (established but incomplete)
{4.4.1, 4.4.2}. In general, the lack of explicit consideration in
global scenarios of good quality of life explicitly, and its
regionally and socially differentiated nature, impedes robust
projections into the future, in particular for non-material
aspects. Interactions of future changes in nature, its
contributions to people and good quality of life can be better
understood and, therefore, potentially better anticipated and
managed, when they are evaluated at regional scales as well
as the global scale.
Small-scale farming, fishing and other communities, and
Indigenous Peoples around the world that depend directly
on local environments for food production, especially in
low-income countries, are particularly vulnerable to climate-
related food insecurity, which raises important equity and
fairness issues. Similarly, in coastal regions, decreases in
precipitation and fresh water supplies, along with projected
increases in sea level, sea surface temperatures and air
temperatures, and ocean acidification are projected to
have major negative effects on water security for societies.
Nature-based livelihoods may become precarious
with intensifying future trends in environmental change
(established but incomplete) {4.4.1, 4.4.2}. Future threats
to biodiversity and ecosystem services also constitute
imminent challenges to the cultural identity of communities,
particularly when faced with environmental degradation
(unresolved) {4.4.2}.
14 The role of people’s knowledge, values and
traditions, and their potential future changes have
been barely explored in global scenarios of future
socio-economic and environmental change. A
challenge to the assessment of nature’s contribution to
people and good quality of life under different future
scenarios is their socially differentiated nature. People’s
values and traditions are crucial in shaping the future, yet
they are rarely central to scenario exercises (established but
incomplete) {4.4.1}. Novel methods are beginning to be
developed to fully integrate people’s worldviews into
scenario planning, however transcendental values held by
the social groups have so far not been well incorporated.
The process of elaborating scenarios with participatory
approaches is increasingly taking into account value
negotiations around the meaning of good quality of life
(established but incomplete) {4.4.2}. Consequently, ethical
questions emerge regarding how to build scenarios so that
local knowledge, particularly that of Indigenous Peoples and
Local Communities (IPLCs), are not coopted in ways that
may exacerbate processes of their social marginalization.
15 Different social groups experience change in
ecosystem function and services differently so that a
given change scenario usually implies winners and
losers in terms of the projected impacts on good
quality of life (established but incomplete) {4.4.1, 4.4.2,
4.4.3}. People vary in their access to ecosystem services,
exposure to disservices, dependence on ecosystems,
needs and aspirations. These are further mediated by
societal structures and norms as individual characteristics
and power relations {4.4.2, 4.4.3}. Many IPLCs are found in
protected areas, where dimensions of good quality of life
such as food and energy security may trade off with other
dimensions of ecosystem functioning. Indirect drivers of
change such as climate mitigation policy (e.g., REDD+) may
disproportionately impact the possible trajectories towards
achieving good quality of life by IPLCs (unresolved) {4.4.1}.
Thus, decision-making about environmental management
with implications for different bundles of ecosystem
services is an intently political process, with often divergent
stakeholder interests and power dynamics. Evaluating
the implications for the good quality of life of IPLCs under
different scenarios of change can benefit from deliberative
and participatory approaches that consider a wide range
of stakeholder views, and disciplinary perspectives. Such a
diversity of perspectives needs to draw on indigenous and
local knowledge, to take account of the multiple interacting
factors and socially differentiated experiences, vulnerabilities
and preferences (established but incomplete) {4.4.2, 4.4.3}.
A limitation with participatory approaches is the difficulty of
imagining future scenarios of changes in the ‘demand side’
of nature’s contributions. So, a group may discuss how
changes in a resource might be affected by climate change,
but it is often framed in terms of current social conditions.
Likewise, participatory approaches are likely to be more
successful if the scale of scenarios (e.g., local, regional,
global) and stakeholder group perspective can be matched.
16 Most internationally agreed policy goals and
targets for biodiversity are missed by most countries
under business-as-usual scenarios because the
current patterns and future trends of production and
consumption are not environmentally sustainable.
Indeed, trajectories of most biodiversity indicators
under business-as-usual increasingly deviate from
targets over time (well established) {sections 2 and 6}.
The achievement of most biodiversity targets therefore
requires a steer away from the current socio-economic
trajectory and the worldviews and values that underpin it
(well established). Scenarios that assume increased
sustainability show that achieving most SDGs is possible at
some point in the future, but this requires substantive and
immediate action (established but incomplete) {4.6.1}, and
the time horizon of the possible achievement of the SDGs
is undetermined.
Scenarios and models can support the formulation of
future biodiversity targets in terms of concept, phrasing,
quantitative elements, and selection of indicators to monitor
progress (established but incomplete). Scenario and models
are also amenable to exploring interactions among targets
(well established). For example, scenarios have shown that
ambitious protected area expansion plans would conflict
with agricultural production under business-as-usual
assumptions, and that achieving SDGs for both biodiversity
and hunger would require a 50-70% increase in land
productivity (inconclusive) {4.6.1}.
Focusing future quantitative targets for biodiversity on
management outcome rather than effort may improve
policy implementation and related management
decisions. For example, the numeric component of Aichi
BiodiversityTarget11 relates to the global proportion of
protected areas. But the aim of protected areas is to achieve
the long-term conservation of nature, which suggests to
move the focus to the amount of nature that is protected
and the effectiveness of protection rather than proportion of
area under protection. Scenarios and models have shown
that the outcome of a protected area network is determined
by its location, connectivity and management, other than
its size.
17 There is a lack of global-scale impact analyses
that integrate across natures, nature’s contributions
to people and good quality of life. Most scenarios
developed for global environmental assessments have
explored impacts of humans on ecosystems, such as
biodiversity or productivity loss {4.1, 4.2}. The effects of
alternative trajectories of socioeconomic development on
ecosystems and ecosystem services have been assessed
as one-way outcomes, ignoring the possible interactions
between natural and socioeconomic systems. A better
understanding of feedback mechanisms is needed on many
fronts, for instance: in what ways pollution arising from
agricultural intensification does impact pollinators and/or
water quality, which in turn impact land use and
intensification? How do changes in food prices arising from
different land uses feed back to land-use decision-making?
How is overfishing leading to the depletion of large
predatoryfish and development of global markets for
alternative species, often their own prey, leading to further
collapse of marine resources? To what extent climate
change induced sea level rise is decreasing wetland area
and is affecting carbon sequestration? (established but
incomplete) {4.1,, 4.5.1-4.5.3, 4.6.1, 4.7.3}. In
addition, storylines of socio-economic development that
underlie global scenarios consider mostly material aspects
of GQL and do not consider other indicators of GQL
{4.4.1-4.4.3}. There is a knowledge gap in scenario studies
about non-material contributions to people compared to
material contributions and regulating contributions, which
limits our capacity to understand quantitatively how nature,
its contributions to people and good quality of life interact
and change in time.
In particular, human decision-making at multiple levels is
not well integrated in global scenario modelling tools such
as Integrated Assessment Models that focus on economic
objectives (well established) {4.1, 4.2, 4.5.1, 4.5.2, 4.4.1-
4.4.3}. A paradigm shift in scenario design could be
achieved by considering, alongside of economic principles,
provisioning of multiple ecosystem services and GQL as
part of the storyline and human decisions (and subsequent
scenario realisation), rather than as an outcome of socio-
economic drivers {4.6.1}. For a more robust scientific
underpinning of biodiversity and multiple sustainability
targets, these non-material aspects need to be explicitly
addressed in the scenarios (unresolved) {4.6.1}. Such
scenarios would facilitate policy-relevant scientific evidence
through exploration of trade-offs and co-benefits between
targets related to biodiversity and ecosystem services,
including the interconnected nature of drivers across
regions {4.3.4, 4.5.1}. Participatory Scenario Planning, with
stakeholders aligned to the scale of the scenario (e.g., the
CBD for global scenarios) would allow for a differentiated
assessment of good quality of life across stakeholder groups
and highlighting winners and losers across environmental or
policy scenarios (established but incomplete) {4.4.2}.
18 Large uncertainties remain in future scenarios
and related impact studies at the global scale. Careful
analysis and communication of sources of uncertainty in
scenarios and models are vital when using them in support
of decision-making (well established). Global modelling tools
to explore futures of biodiversity and futures of ecosystem
state and function are still mostly disconnected and do not
consider diversity-function links {4.2, 4.7}. Projected future
changes in species ranges, community diversity or
ecosystems may be under- or overestimated by most
studies because they do not explicitly account for impacts
of multiple drivers, adaptive capacity of species and for
feedbacks arising from species interactions {established but
incomplete) {4.2.5, 4.5}. Effectively linking scenarios and
models across spatial and temporal scales is
methodologically difficult and in early stages of development
and use but can make important contributions to decision-
making when achieved (established but incomplete).
However, linking must be done with considerable caution
because it creates additional complexity that can make the
behaviour of scenarios and models difficult to understand
and may introduce important sources of uncertainty {4.5,
4.7}. Substantial efforts are needed to identify uncertainty
related to models and scenarios and improve the treatment
of uncertainty between and within models {4.2, 4.6, 4.7}.
Strong, sustained dialogue between modellers, stakeholders
and policymakers are one of the most important keys to
overcoming many of the significant challenges to dealing
with uncertainty and scales issues when mobilizing
scenarios and models for decision-making.
4.1.1 Context and objectives of
the chapter
Rapid biodiversity loss and its adverse consequences for
nature, nature’s contributions to people and Good quality
of life clearly remain as key challenges for the coming
decades. Economic inequality, societal polarization and
intensifying environmental threats have been identified
by the World Economic Forum’s Global Risks Report
(GRR) 2017 (WEF, 2017) as the top three challenges for
global developments over the next decade or more. For
the first time, all five environmental risks in the report
(extreme weather; failure of climate change mitigation
and adaptation; major biodiversity loss; natural disasters;
human-made environmental disasters) were ranked both
high-risk and high-likelihood (WEF, 2017). These challenges
emphasize the importance of the UN 2030 Agenda and
the Sustainable Development Goals (SDGs) and the 2050
Global Vision for Biodiversity to facilitate a sustainable future
state for the planet, with a recognition of the connections
between humans and ecosystem well-being at their core
(Costanza et al., 2016).
This chapter focuses on the assessment of scenarios and
models that have been used to explore a wide range of
plausible futures of nature, nature’s contributions to people
(NCP) and good quality of life (GQL), focusing on the
current-to-2050 time frame and on continental to global
spatial scales. One objective is to alert decision makers to
potential undesirable impacts of a broad range of plausible
Figure 4 1 1
Scope of Chapter 4 “Plausible futures of nature, its contributions to people and
their good quality of life” of the IPBES Global Assessment, content of sections
and their relationships, and linkages with the other chapters of the Global
Assessment. NCP: nature’s contributions to people; GQL: good quality of life.
Global change scenarios
Future projections of drivers
Future impacts on NATURE
Ecosystem state and function
Future Interactions between
Future impacts on GQL
Future impacts on NCP
Likelihood of achieving a
sustainable future
(SDGs, Aichi Biodiversity Targets)
Spatio-temporal scales
Indirect drivers Direct drivers
Section 1 Section 2
Section 5
Section 4 Section 3
Section 6 Section 7
Chapter 2
Chapter 3
Chapter 2
Chapter 2
Chapter 5
Chapter 6
Policy options
socio-economic development pathways. A second
objective is to highlight development pathways and actions
that can be taken to minimize impacts, as well as restore
nature and enhance its contributions to people. As is clearly
highlighted in Chapters 2 and 3 of this assessment, the
context is that pressures, such as resource exploitation
and climate change, continue to increase, and most
measures of the state of nature and nature’s contributions
to people continue to decline. This chapter is designed to
help understand the conditions under which these trends
might accelerate vs. stabilize or even improve over the
coming decades.
Scenarios are a means of exploring plausible future
trajectories of direct and indirect drivers of environmental
change (IPBES, 2016b). Models provide a means to
estimate qualitatively or quantitatively the impacts of indirect
and direct drivers on nature and nature’s contributions to
people (IPBES, 2016b). Building upon an analysis of drivers
of change presented in chapter 2.1, this chapter starts
with an assessment of the key underlying assumptions
about drivers in scenarios and a synthesis of the projected
trajectories of key direct drivers, such as climate change
and land-use change, and indirect drivers, such as human
population and economic growth, over the next several
decades and places these in the context of current trends
(section 4.1; Figure 4.1.1, see Chapter 2.1).
Sections 4.2 and 4.3 of this chapter focus on the
assessment of a wide range of quantitative models that
have been used to project future dynamics of nature and
its contributions, and these sections also place these
projections in the context of observed trends as well as
the current understanding of the mechanisms underlying
these trends (see Chapter 2). Models can also be used
to evaluate the impacts of changes in nature and its
contributions on quality of life, but this has rarely been
done (IPBES, 2016b). As such, section 4.4 focuses on the
underlying assumptions about quality of life embedded
explicitly or implicitly in models and scenarios, as well as
making qualitative connections with modeled impacts
on nature and its contributions. Projected synergies and
trade-offs between nature, NCP and GQL are explored in
section 4.5.
Finally, comparisons of scenarios and model outcomes
are then made with internationally agreed objectives,
such as the Sustainable Development Goals for 2030 and
the Convention on Biological Diversity’s 2050 Vision, in
order to better understand the types of socio-economic
development pathways that lead to outcomes that are
closest to or furthest from these objectives (section 4.6).
This is then put in the broader context of the use of
scenarios and models in decision-making (section 4.7), with
a focus on the importance of scales and uncertainty in the
use of models and scenarios to inform decisions.
Chapter 5 follows by providing a more in-depth analysis of
“target-seeking” scenarios designed to evaluate sustainable
futures, including evidence regarding sustainable transition
pathways, for which specific policy options are discussed in
Chapter 6.
4.1.2 Exploratory scenarios
Scenarios can be defined as plausible representations of
possible futures for one or more components of a system,
or as alternative policy or management options intended
to alter the future state of these components (IPBES,
2016b). They provide a useful means of dealing with many
distinct possible futures (Cook et al., 2014; Pereira et
al., 2010). Policy and decision-making processes rely on
estimates of anticipated future socio-economic pathways,
and knowledge of the potential outcomes of actions
across distinct geographic regions, sectors and social
groups. The process of scenario development itself can
help to build consensus by integrating the objectives of
different stakeholder groups (Priess & Hauck, 2014). This
is particularly germane in efforts that seek to integrate the
knowledge, perspectives and goals of local stakeholders,
particularly Indigenous Peoples and Local Communities
(IPLCs), who are frequently marginalized from policy and
decision-making processes (IPBES, 2016b; Petheram et
al., 2013).
When assessing future impacts on nature, its contribution to
people and related good quality of life, there is a need to link
the trajectory of direct and indirect drivers to different future
scenarios. Exploratory scenarios can be either qualitative,
in the form of storylines, or quantitative, in the form of
model outputs (van Vliet & Kok, 2015). The main objective
of exploratory scenarios is informing stakeholders of the
potential impacts of different driver combinations, e.g., a
proactive set of actions that may increase the likelihood of
social, economic or political targets versus a “business-
as usual” scenario that involves no major interventions
or paradigm shifts in the organization of functioning of a
system. Exploratory scenarios may provide a plurality of
plausible alternative and contrasting futures.
Exploratory scenarios for global scale environmental studies
and assessments have been developed for a range of
UN related assessments, including scenarios developed
under the IPCC process, such as the so-called SRES
scenarios (Nakicenovic et al., 2000) in the late 1990s, the
Representative Concentration Pathways (RCPs) and the
recent Shared Socio-economic Pathways (SSPs), as well as
scenarios considered for the UNEP Global Environmental
Outlook (GEO) (UNEP, 2012) process, Global Biodiversity
Outlook (GBO) and the Millennium Ecosystem Assessment
(MA, 2005). The Global Scenario Group has also developed
a range of contrasting global scenarios (Raskin et al., 2002).
In addition, organizations such as FAO, OECD, IEA and
UNESCO have developed several scenarios for specific
purposes, such as the OECD Environmental Outlook to
2050 where a trend-based scenario was developed and a
large number of policy alternatives were evaluated (OECD,
2012). Several of these scenarios have been evaluated
by Integrated Assessment Models (IAMs) to specify and
quantify ecological and environmental changes, including
climate change, land-use change, vegetation dynamics and
water (Kok et al., 2018).
An important advance in the last few years has been to link
representative concentration pathways (RCPs) with shared
socio-economic pathways (SSPs) (O’Neill et al., 2014)
in support of the IPCC process, to inform deliberations
under the UN Framework Convention on Climate Change
(UNFCCC). Some of these scenarios imply significant
mitigation efforts in the land-use sector, including large-scale
reforestation and afforestation, or bioenergy crops with
implications for both biodiversity and ecosystem services
(Riahi et al., 2017).
Existing environmentally relevant scenarios include scenarios
that are most often either exploratory (this chapter focus) or
target-seeking (Chapter 5) (IPBES, 2016b). In many cases,
these scenarios may be appropriate for specific temporal
or spatial scales or limited in scope (e.g. relevant to one or
a few sectors). They can also be incomplete with regard to
quantitative information about nature, NCP and GQL, and
thus less useful for the purposes of this IPBES assessment.
This is because integrated assessment models that often
underpin scenarios of future greenhouse gas emissions,
land-use change, or demand for food have a strong
economic perspective and do not consider e.g., monetary
or non-monetary values of ecosystem services. Issues
related to conservation or biodiversity, or feedbacks from
changes in ecosystem services to socio-economic decision-
making, have typically not been well considered in the wide
range of global scenarios that are well established in the
climate change scientific communities. Likewise, scenarios
of the future of biodiversity typically do not seek to quantify
the possible co-benefits related to ecosystem services (Kok
et al., 2017; Pereira et al., 2010; Powell & Lenton, 2013).
Important gaps remain in scenario development, such as
the development of integrated scenarios for areas projected
to experience significant impacts and possible regime
shifts (e.g. Arctic, semi-arid regions and small islands), and
socioeconomic scenarios developed for and in collaboration
with Indigenous Peoples and Local Communities (IPLCs)
and their associated institutions, values and worldviews
(Furgal & Seguin, 2006).
4.1.3 Archetype scenarios
From the many scenarios developed in the last few
decades, it is apparent that groups of scenarios have
many aspects of their underlying storylines in common and
may be considered as “archetype scenarios”. Archetypes
represent synthetic overviews of a set of assumptions
about the configuration and influence of direct and indirect
drivers used in scenarios. They vary mainly in the degree
of dominance of markets, dominance of globalization, and
dominance of policies toward sustainability. Hunt et al.
(2012) and van Vuuren et al. (2012) analysed a large number
of local and global scenarios and came to the similar
Box 4 1 1 Scenario archetypes.
(from Hunt et al., 2012; IPBES, 2016b; van Vuuren et al., 2012; see also section 5.2.2 in IPBES, 2018i): description of
underlying storylines, and links with indirect and direct drivers.
Economic Optimism. Global developments steered by
economic growth result in a strong dominance of international
markets with a low degree of regulation. Economic growth
is assumed to coincide with low population growth due to a
strong drop in fertility levels. Technology development is rapid
and there is a partial convergence of income levels across
the world. Environmental problems are only dealt with when
solutions are of economic interest. The combination of a high
economic growth with low population growth leads to high
demands of commodities and luxury goods. These demands
will however be unequally distributed among regions and within
regions. Consequently, energy use and consumption are high.
In addition, high technological development in combination
with increased global market leads to high yields in agricultural
and wood production on the most productive lands. Therefore,
pollution and climate change will be relatively high, but land use
relatively low. Direct exploitation will continue but also replaced
by cultivation of for example fish and livestock. Global trade will
increase the risks of invasive species.
Reformed Markets. Similar to the economic optimism
scenario family but includes regulation and other policy
assumptions to correct market failures with respect to social
development, poverty alleviation or the environment. Thereby,
relative to the economic optimism archetype, high demands
for goods are expected to be more equally distributed and
pollution will be lower.
Global Sustainable Development. A globalized world with
an increasingly proactive attitude of policymakers and the
public at large towards environmental issues and a high level
of regulation. Important aspects on the road to sustainability
conclusion that four to six scenario archetypes cover the
large range of possible futures (Box 4.1.1).
This chapter makes frequent reference to archetype
scenarios because the use of scenario archetypes was also
adopted in the IPBES regional assessments. This approach
helped to synthesize results across a very broad range of
scenario types. Synthesis across regional assessments is
hampered by the use of different archetype classifications
for each of the regions, which was done in order to match
archetypes to regional contexts.
The IPBES methodological assessment on scenarios and
models (IPBES, 2016b) adopted the “scenario families”,
as described in van Vuuren et al. (2012), which include the
scenario archetypes (Box 4.1.1) distinguished by Hunt et
al. (2012).
The different scenario archetypes describe different visions
of the future (de Vries & Petersen, 2009), reflecting different
values, guiding principles of society, understanding of
good quality of life, approaches to decision-making and
distribution of power (among other aspects). These aspects
are often included in scenarios as implicit assumptions and
have a large impact on the outcomes of the scenarios. For
example, some scenario archetypes may prioritize intrinsic
values of nature, while others may emphasize instrumental
or relational values (Pascual et al., 2017). These differences
ultimately affect the different archetypes in various ways.
Table 4.1.1. shows all these aspects synthesized across
the six scenario archetypes. The most common global scale
scenarios encountered in the literature can be assigned
to these archetypes (Table 4.1.2), with the caveat that
individual scenarios do not match all of the characteristics of
the archetype defined in Table 4.1.1 and Box 4.1.1.
Analysis of the data sourced from the systematic literature
review (Appendix A4.1.1) carried out as part of the
background work for this chapter indicates a skewed
representation of scenarios between and across the three
components nature, NCP and GQL (Table 4.1.3). This
skew reflects to some extent the length of time scenarios
have been available, but also reflects a bias towards
climate change related scenarios. The analysis shows
are technological change, strong multi-level governance,
behavioural change through education, and a relatively healthy
economy. All variations of this archetype are beneficial for
biodiversity. This scenario combines a low population growth
with moderate economic development, and sustainable
production and consumption. Low demands of especially
luxury goods are expected, and a shift in diet towards less meat
can be expected. Energy use will be low to moderate and fossil
fuel use will be reduced, leading to low climate change and low
land-use change. Due to environmental policies and sustainable
production, pollution will be lower and direct harvesting will
partly be replaced by cultivation. The global focus will increase
the risk of invasive species
Regional Sustainability. A regionalized world based on an
increased concern for environmental and social sustainability.
International institutions decline in importance, with a shift
toward local and regional decision-making, increasingly
influenced by environmentally aware citizens, with a trend
toward local self-reliance and stronger communities that focus
on welfare, equality, and environmental protection through local
solutions. The scenario combines a low economic growth with
moderate population growth rates. The demands for goods
are low and production focusses on sustainability with low
levels of energy use or environmental degradation associated
with higher importance for intrinsic and relational values of
nature. Low rates of climate change are expected. Supply of
agricultural products will be organised with regions with low
levels of global trade. A slow technological development and
a sub-optimal land use lead to relatively high rates of land-use
change. Direct exploitation of natural systems will be within
the carrying capacity of natural systems, and risks for invasive
species will be relatively low.
Regional Competition. A regionalized world based on
economic developments. The market mechanism fails, leading
to a growing gap between rich and poor. In turn, this results in
increasing problems with crime, violence and terrorism, which
eventuates in strong trade and other barriers. The effects on
the environment and biodiversity are mixed. Overall, there is
a tendency towards increased security, which can either be
positive (protect biodiversity) or negative (intensify agricultural
production). Particularly in low-income countries, deforestation
and loss of natural areas are a risk. In this scenario, due to a
lack of global co-operation and trade, a high population growth
is expected combined with low economic growth. Thereby, the
demand for goods including agricultural products increases, but
the demand for luxury, energy intensive goods is relatively low,
and thus relatively low climate change is expected. Agricultural
supply will be mainly within regions, which, combined with slow
technological development, will result in lower productivity and
high rate of land-use change. Direct exploitation will continue,
low rates of replacement by cultivation are expected. The risk of
invasive species will be lower than in the archetypes that focus
on globalization.
Business-As-Usual. Assumes that the future can be
characterised by a continuation of historical trends, including
the implementation of international agreements. Sometimes
referred to as a reference scenario, or as a middle-of-the-road
scenario. It can also be considered as a less extreme variant
of the economic optimism archetype. Business-as-usual
is characterized by moderate economic growth, moderate
population growth and moderate globalization. Demands
are not high nor low, and in combination with moderate
technological development, environmental changes will also
be moderate.
the available literature is strongly dominated by studies of
future trajectories of nature, with considerably fewer studies
on NCP and very few studies providing information on
GQL. This may reflect the lack of integrated assessment
tools available to conduct this type of work quantitatively.
This inconsistency of coverage constrained the work in
this chapter, and explains the emphasis put on nature
(section 4.2).
Prosperity based
on economic
efficiency &
Equity & local
Individualism and
safety concerns
No change
Main value in
Instrumental /
Utility value
Instrumental /
Utility value
Intrinsic /
Relational Instrumental /
Utility value
Instrumental /
Utility value
More "efficient"
use of nature
with new
technologies, but
protection is not
Use of nature is
regulated with
reformed polices
nature and
Local sustainable
use of nature
Lack of concern/
low priority for
of nature with
elements of
regulation and
Social principles Individualism Individualism
with elements of
within the
Individualism in a
fragmented world
with elements of
Market oriented
based on profit
Market regulation
based on
efficiency &
Market regulation
and non-market
based on global
sustainability and
oriented to local
and quality of life
oriented with
trade barriers
and growing
asymmetries /
Market oriented
with some
barriers and
some regulation
Approach to
good quality
of life
Material aspects Material aspects,
health and other
GQL components
included in
goals (e.g. SDG)
Respect for
nature at the
global scale is
important for
relationships and
Public security Material aspects,
and other
components such
as health, public
Power relations
among countries
Large countries
Power imbalance
moderated by
Power balanced
by global
institutions and
among and within
High differences
in power among
Large countries
are powerful,
power partially
balanced by
negotiation, high
differences in
power among
Top-down Top-down Horizontal /
Bottom-up /
Top-down with
growing exclusion
of the poorest
(most vulnerable)
regions & social
Private sector Alliance of
governments and
private sector
Balance of
power among
the various
global institutions
Communities National
Governments and
private sector
Private sector
& governments,
with participation
of NGOs
Table 4 1 1 Different guiding principles, values, approaches to good quality of life
(GQL), distribution of power and decision-making approach across scenario
Source Economic
SRES A1F1 B1 (A1T) B2 A2 B2
GEO3/GEO4 Market first Policy first Sustainability
Security first
Global scenario group Conventional
Policy reform New
Millennium Ecosystem
Technogarden Adapting mosaic Order from
OECD Environmental Outlook Trend
Shared Socio-economic
Representative Concentration
Pathways (RCP)
RCP8.5 RCP 2.6 RCP 6.0 RCP 4.5
Roads from Rio/ fourth Global
Biodiversity Outlook
Table 4 1 2 Scenarios from earlier global assessments attributed to archetypes or families.
Source: IPBES, 2016b; van Vuuren et al., 2012.
Scenario All Nature NCP GQL
RCP8.5 237 198 39 0
RCP6.0 9 9 0 0
RCP4.5 50 41 9 0
RCP2.6 150 144 6 0
A1 6 4 1 1
A1b 119 108 8 3
A1B 4 0 4 0
A1F1 76 76 0 0
A1T 1 0 1 0
A2 200 191 7 2
B1 113 106 6 1
B2 123 117 5 1
SSP1 1 0 1 0
SSP2 13 1 12 0
SSP3 2 1 1 0
Table 4 1 3 Classification of studies according to scenario represented along a continuum
from nature via NCP (nature’s contributions to people) to GQL (good quality of
life) focused studies.
The number of papers reported comes from the systematic literature review conducted for this chapter (Appendix A4.1.1).
4.1.4 Projected indirect and direct
drivers of change in scenarios
The main indirect drivers of change of nature and its
contributions to people, and consequently the quality of
life include economic development, demographic trends
and factors, technological development, governance and
institutions, and various socio-cultural aspects such as
worldviews and values. These indirect drivers have multiple
impacts on direct drivers of change, which include climate
change, land-use change, pollution, direct harvesting,
invasive species and disturbance. In each scenario
archetype, assumptions on the indirect drivers lead to
different combinations of direct drivers (Box 4.1.1).
Drivers are always multiple and interactive, so that one-to-one
linkage between particular drivers and specific changes in
ecosystems rarely exists. The causal linkage between drivers
is often mediated by other factors or a complex combination
of multiple factors, thereby complicating the understanding
of causality or attempts to establish the contributions by the
various drivers to changes in nature, NCP and GQL (see also
Bustamante et al., 2018; Elbakidze et al., 2018; Nyingi et
al., 2018; Wu et al., 2018). The cumulative effects of multiple
stressors may not be additive but may be magnified by their
interactions (synergies) and can lead to critical thresholds
and transitions of ecological systems (Côté et al., 2016).
Cascading impacts of co-occurring stressors are expected to
degrade ecosystems faster and more severely (section 4.7 in
Bustamante et al., 2018). Indirect Drivers (including
consideration of diverse values) in
Indirect drivers (also referred to as ‘underlying causes’)
operate diffusely by altering and influencing direct
drivers as well as other indirect drivers (also see chapter
1 in this report and IPBES, 2016b). They influence
human production and consumption patterns with
subsequent environmental implications. Economic drivers,
including trade and finances, and demographic drivers
interact with other indirect drivers such as technology,
governance/institutions and social development including
equity. Archetype environmental scenarios for this
century consider explicit reference to relevant indirect
anthropogenic drivers in different combinations, as
indicated in Table 4.1.4.
Economic development has historically been the key
indirect anthropogenic driver of changes in nature, NCP and
GQL, across all scales (global, regional, national and local).
World GDP (at constant 2010 USD) increased by 6.9 times
between 1960 to 2016 (based on Worldbank, 2017). Taking
a historical perspective, past and prevailing patterns of
production and consumption embodied in global economic
trends have generated growing pressures on natural
resources, the environment, and ecosystem functions.
In all scenarios, world GDP will continue to grow (Table
4.1.5). However, some studies also refer to the plausibility of
sustainable de-growth, as a transformative pathway leading
to a steady-state at a reduced level of economic output
(Schneider et al., 2011).
Economic activities, international trade and financial flows
are closely related, particularly in recent decades due to
increasing economic globalization. These considerably
influence changes in nature, NCP and GQL through various
direct and indirect pathways. In turn, these pathways are
influenced by a number of policy channels and mechanisms,
like trade policies, including incentives (tax exemptions,
subsidies) and trade barriers, the dynamics of foreign debt
and foreign debt service, flows of foreign direct investments,
and monetary policies (dynamic of exchange rates,
interest rates).
Demographic trends are a major indirect anthropogenic
driver of changes in nature, NCP and GQL, across
Scenario All Nature NCP GQL
SSP5 1 1 0 0
BAU 23 20 3 0
Global orchestration 13 11 2 0
Order from strength 12 9 3 0
Technogarden 11 10 1 0
Adapting mosaic 8 7 1 0
Consumption change 6 6 0 0
Global Technology 3 0 3 0
Decentralized solutions 1 1 0 0
Selected indirect drivers Archetype / scenario family
Economic development Very rapid Rapid Ranging from
slow to rapid
Medium Slow Medium
Trade Globalisation Globalisation Globalisation Trade barriers Trade barriers Weak
Technological development Rapid Rapid Ranging from
medium to rapid
Ranging from
slow to rapid
Slow Medium
Population growth Low Low Low Medium High Medium
Policies & institutions
Policies create
open markets
Policies reduce
market failures
Strong global
Local steering Strong national
Table 4 1 4 Selected indirect drivers in archetype scenarios.
Source: Based on Cheung et al. (2016: table 6.3 ); van Vuuren et al. (2012).
all scales (global, regional, national and local). World
population increased by 2.5 times, respectively
between 1960 and 2016 (based on the World Bank
Database, 2017). Population / demographic drivers
consider changes in population size, migration flows,
urbanization as well as demographic variables such as
population distribution and age structure. Urbanisation
driven by growing populations and internal migration
acts as an indirect driver of land-use change through
various ways, including through linear infrastructures
such as transportation networks as well as synergies
with other forms of infrastructure development (IPBES,
2016b). By 2050, all archetype scenarios project great
increase in human population size, while towards the
end of the century, downward trends are projected for
the “economic optimism” (SSP5), “global sustainable
development” (SSP1), “reformed markets” scenarios
(Table 4.1.2, Figure 4.1.2).
Per capita GDP trends combine the impacts of GDP and
population growth on environment. Growing per capita
GDP has historically implied increasing demand of key
natural resources such as food, water and energy with
adverse impacts on ecosystems and biodiversity, due to
the persistence of unsustainable patterns of production
and consumption. Humanity’s demand has exceeded
the planet’s biocapacity for more than 40 years, and the
Ecological Footprint shows that 1.6 Earths would be
required to meet the demands humanity makes on nature
each year, with consumption patterns in high-income
countries resulting in disproportional demands on renewable
resources, often at the expense of people and nature
elsewhere in the world (WWF, 2016).
Technology development can significantly increase
the availability of some ecosystem services, and improve
the efficiency of provision, management, and allocation
GDP PPP in trillion 2000 US$
2050 182-323 181-229 168-251 139-145 106-198 145-241
2100 458-895 427 213-498 310 177-321 310-473
Table 4 1 5 Economic development (in GDP PPP) for the scenario archetypes.
Source: MA, 2005; Nakicenovic et al., 2000; OECD, 2012; Raskin et al., 2002; Riahi et al., 2017; UNEP, 2007). Global GDP
was approximately 50 trillion $ at purchasing power parity in 2000. GDP PPP: Global Domestic Product based on purchasing
power parity.
of different ecosystem services, but it cannot serve as
a substitute for all ecosystem services. Technologies
associated with agriculture and other land uses have a large
impact as drivers of biodiversity and ecosystem change
(IPBES, 2016a).
As part of the problem, some technologies can result
in increased pressure on ecosystem services through
increased natural resource demand as well as lead to
unforeseen ecological risks, particularly natural resource
intensive technologies, as those associated to agricultural
land expansion (e.g., first generation of biofuels when
produced unsustainably). In addition, climate change
is directly related to the use of fossil-fuel-intensive
technologies. As part of the solution, sustainability-oriented
technological innovation may contribute to decouple
economic growth and the consumption of natural resources
through increasing efficiency, resilience and equity (e.g.
agroecological food production systems) (IPBES, 2016a;
Trace, 2016; Vos & Cruz, 2015).
Governance and institutions play an important
role in the management of biodiversity, ecosystem
services and ecosystem functions. Weak governance,
including corruption, frequently leads to environmental
mismanagement as well as the adoption of environmentally
unsustainable policies, and growing conflicts (Pichs-
Madruga et al., 2016). The lack of recognition of indigenous
and local knowledge (ILK) and institutions may also generate
adverse consequences for nature, NCP and GQL as well
as for Indigenous Peoples and Local Communities (IPLCs).
In addition to governments, new actors and coalitions
(e.g. NGOs, researchers, indigenous groups) with different
– and sometimes divergent and conflicting – perceptions
and values are performing critical roles in environmental
decision-making processes.
Social development and culture are critical ingredients
of future scenarios on biodiversity, yet there is a lack of
attention towards understanding how values, norms,
and beliefs affect attitudes and behaviours towards the
environment, and their roles in shaping the future and in
driving transformation pathways. While there has been
advances in methodologies supporting social-ecological
analyses, emphasis has been on measurable indicators
with less attention to the role of sociocultural values and
practices in shaping other indirect drivers of change, and
thus future pathways (Pichs-Madruga et al., 2016).
Social inequity is a key concern in many regions, sub-
regions, countries and territories. In many cases, poverty
conditions correlate with increasing pressures on nature,
but globally per capita consumption of natural resources is
strongly correlated with affluence. World per capita private
consumption, in dollars at constant 2010 prices, rose by
44.5% between 1990 and 2016 (Worldbank, 2017). The
emergence of new waves of affluent consumers is projected
to significantly increase the demand for already limited
natural resources (Myers & Kent, 2003). For this reason, the
impact of consumers’ purchasing power on the demand of
natural resources is receiving growing attention in scenarios.
This discussion is very relevant in the context of the global
Figure 4 1 2
Projected changes in world population according to the fi ve Shared Socio-
economic Pathways (KC & Lutz, 2017).
Note: For the narratives of the SSPs, see O’Neill et al. (2017).
debate on the Sustainable Development Goals (SDGs),
multidimensional progress in human development (UNDP,
2016) and their interlinkages with nature and NCP. Direct Drivers
Climate change
By the end of the 21st century, three of four explored
Representative Concentration Pathways (RCP; van Vuuren
et al., 2011) result in an increase in global average surface
temperatures above 1.5°C compared to the present-day
reference period 1986-2005 (Stocker et al., 2013). Averaged
over years 2046-2065, temperature increases range from
(model median) 1.4°C (RCP4.5) to 2.0°C (RCP8.5) above
the reference period (1986-2005). Only the RCP2.6 scenario
could possibly lead to a below 2°C world, with projected
warming above the reference period from 0.3 to 1.7°C
averaged over the last two decades of the 21st century, and
from 0.4-1.6°C for years 2046-2065. Warming will be larger
over land and by far highest in the Arctic. The frequency
of extreme hot weather events will increase (Stocker et
al., 2013). Precipitation patterns will change in a complex,
spatially non-uniform way.
Based on climate modelling done for the IPCC 5th
assessment report, and recent work presented in the IPCC
special report on 1.5 degrees (IPCC, 2018), limiting warming
to 1.5°C above preindustrial levels will require rapid,
historically unprecedented mitigation efforts (Millar et al.,
2017). Applying a different, statistical modelling approach
found below 2°C warming at the end of the 21stcentury
unlikely, and requiring a much accelerated decline in carbon
intensity compared to the past decades (Raftery et al.,
2017). By 2050, in the RCP2.6 pathway, CO2 emissions
are projected to be lower than they were in 1990. Projected
atmospheric concentrations range from ca. 440 ppm
(RCP2.6) to ca. 540 ppm (RCP8.5) by 2050 to ca. 420-935
ppm by 2100, but uncertainties are of several tens/hundreds
of ppm.
Land-use change
Land-use and land-cover changes have direct and large
impacts on the physical environment. They include
expansion of crops and pastures, as well as intensification
and management changes, mineral and biomass extraction,
urbanization and infrastructure expansion (Geist & Lambin,
2002). Eitelberg et al. (2015) estimated the global potential
for crop area to range from ca. present-day expanse (1500
Mha) to nearly a tripling (5100 Mha), depending on different
future socio-economic and governance assumptions.
Synthesising projected future crop, pasture and forest areas,
Alexander et al. (2017c) showed a huge spread in projected
future land-use change, and found that this spread
depended on the type of scenario, as could be expected,
but also was heavily dependent on the type of model used
to quantify land use for a given scenario (i.e. the same
scenario archetype results in very different land-use change
patterns depending on the underlying model’s assumptions
and structure). Overall, these studies suggest that there
remains a high level of uncertainty in future land-use change
potential and in scenarios of land-use change.
The five main SSP storylines that have been developed in
support of the IPCC can be classified by archetypes (Table
4.1.2), but considerable caution should be exercised when
interpreting land-use projections from the SSP storylines
as being representative of a particular archetype. For
example, the largest declines in global area of forest and
other natural land occur in the reference scenarios (also
referred to as “marker scenarios”) for SSP3, SSP4 and
SSP5 (Popp et al., 2017), i.e. scenarios that emphasise
competition or free markets. However, the range of variation
of the projected change in managed land area by 2100 is
nearly as large within SSPs (i.e. variation due to application
of different IAMs to the same SSP storyline) as it is between
marker scenarios across SSPs (Popp et al., 2017). Given
this large variation within SSPs and high uncertainty in
land-use projections identified by Alexander et al. (2017c),
considerable caution must be exercised when making
the connection between the underlying assumptions of
scenario archetypes (Tables 4.1.1 and 4.1.4) and an
individual projection of land use by a single Integrated
Assessment Model (e.g., Figure 4.1.3).
In the wake of the Paris COP21 agreement, terrestrial
ecosystems will make crucial contributions to meeting
agreed climate mitigation objectives. Achieving the RCP2.6
pathway (or the most recent RCP1.9 pathway, see IPCC,
2018) requires, in nearly all scenarios developed with IAMs,
negative emissions through carbon-dioxide removal. The
majority of this is generally achieved through reforestation,
afforestation and avoided deforestation, as well as bioenergy
plantations coupled with carbon capture and storage
(Anderson & Peters, 2016; Smith et al., 2016). Depending
on how fast fossil fuel emissions decline, substantial
negative emissions to balance continued fossil emissions
need to be achieved by 2050, or even earlier (Anderson
& Peters, 2016) which, if implemented, will have large
consequences for terrestrial ecosystems. Recent results
indicated that SSPs 1, 2, 4 and 5 might be consistent with
low greenhouse gas emissions (i.e., RCP2.6; Kriegler et
al., 2014; Popp et al., 2017) (see also examples in Figure
4.1.3). Despite the very different assumptions contained
in the SSPs (and in the IAMs simulating these) there is
consistent projected decline in food crop and pasture area
at the end of the 21st century, even though demand for crop
and livestock products tend to be larger than today. At the
same time, area under bioenergy plantation increases by
between ca. 200 Mha (SSP1/AIM) and 1500 Mha (SSP4/
The intensity of land-use change can be as important as the
change in area. In particular, the productivity of croplands
is assumed to increase in the future as a result of increased
application of technology, including the use of fertilizers, high
producing varieties, machinery and pesticides. Intensification
has huge impacts on biodiversity in agricultural landscapes,
where for example species richness reduces by more than
50% in intensively used croplands, compared to low input
systems (e.g., Newbold et al., 2015). Intensification will
continue in the coming decades and a recent analysis for
-4 0 4
-19 Mha
+251 Mha
+262 Mha
-1 -0.8 -0.6 -0.4 -0.2 -0.1 0.1 0.2 0.4 0.6 0.8 1
Figure 4 1 3
Projected changes in cropland area used in the Biodiversity and Ecosystem
Services Model Intercomparison Project (BES-SIM, Kim et al., 2018; see box 4.2.5).
Note that the depicted projections are from a single Integrated Assessment Model per SSP and may not be representative of
the range of land-use projections for archetypes to which the SSP is assigned (Table 4.1.2). For instance, trends in agricultural
area used by Krause et al. (2017, 2018) for their RCP2.6 baseline case were opposite (increasing agricultural area) to the trends
seen in the scenario RCP2.6/SSP1 used in BES-SIM. A map with cropland as well as pasture area changes is provided in
Appendix 4.1 (Figure A4.1.12).
the SSP scenarios showed trade-offs between land-use
change and intensification (Table 4.1.6).
To meet the demand of a growing and wealthier population,
increased agricultural production results from land
conversion to cropland in the SSP3/RCP6.0 and SSP5/
RCP8.5 scenarios and from intensification in all scenarios,
where in SSP3/RCP6.0 scenario a relatively low increase of
the yield is assumed.
Pollution here refers to solid and chemical waste of various
kinds, excluding the gases referenced in the Kyoto and
Montreal Protocols. Large increases in waste generation
have occurred in the past decades, with a particular
challenge for persistent organic pollutants (POPs) and
synthetic organic polymers (plastics) which are physically
harmful, chemically toxic, and slow to metabolize (see Solid waste generation rates depend strongly
on urban population growth trends, together with
changing standard of living and societal efforts towards
waste reduction. On current trends, waste production will
attain 11Mt day-1 by 2100, and will continue to rise into
the latter half of this century particularly in sub-Saharan
Africa (Hoornweg et al., 2013). However, socio-economic
pathways could strongly affect waste production trends,
with SSP1 stabilising global waste production by about
2070 at roughly 8.5 Mt day-1 relative to values of 12 Mt day-1
in SSP2 and SSP3 (Hoornweg et al., 2013).
Direct harvesting of natural resources
Scenarios relating to direct harvesting will have complex
relationships with distinct socio-economic futures. In
terrestrial ecosystems, while an increase in human wealth
may reduce direct harvesting of provisioning resources (such
as bushmeat), increasing wealth may increase demands
for some traditional (e.g. medicinal) and “luxury” (e.g. Rhino
horn) resources. On the other hand, marine and freshwater
natural resources might undergo increased fishing pressure
in the face of rising affluence and continuous growth of
human population that is projected to reach 9.8 billion
people by 2050 (UNDESA, 2017). Scenarios of governance
in fisheries management, human consumption of seafood,
improvement of fishing technology (Squires & Vestergaard,
2013) are starting to be integrated into future global scale
projections (section
Invasive Alien Species
Invasive alien species (IAS) are those that have been moved
by direct human actions beyond their native geographic
range, and have established and actively expand geographic
range after introduction (Blackburn et al., 2014). The main
impacts of socio-economic scenarios on IAS are likely to
be through vectors for dispersal (with international trade
and long-distance transport being the most important), and
economic resources to combat IAS. Higher impacts are
thus to be expected under future scenarios of greater global
trade with weaker local governance.
Quantification of the impacts of IAS tends to focus on
adverse ecological effects (Simberloff et al., 2013), including
adverse impacts on ecosystem services. It is thus difficult to
develop a fully integrated understanding of positive, neutral
and negative impacts, though current consensus strongly
suggests overall adverse impacts (Pyšek & Richardson,
2010). For example, invasive plants can cause catastrophic
regime shifts and indigenous diversity reduction (Gaertner
et al., 2014), such as through N-fixing species increasing N
concentrations in nutrient-poor soil (Blackburn et al., 2014),
and by increasing fire frequencies and intensities, or even
introducing novel fire regimes (Pausas & Keeley, 2014).
Cropland in 2015 in km215885409 15885409 15885409
Cropland in 2050 in km215696191 18399153 18507559
Cropland area increase
2015-2050 %
-1.2 15.8 16.55
Crop production increase
2015-2050 %
31.7 40.5 58.4
Yield increase 2015-2050 % 33 21 36
Yield increase per year % 0.95 0.61 1.03
Table 4 1 6 Changes in global cropland area and productivity increase for three SSP
scenarios, as analysed in a model comparison study by BES-SIM.
Invasive animals may cause extreme indigenous diversity
loss particularly if they are predators and invade in islands
(Medina et al., 2011).
The number of documented IAS is most probably a
significant underestimate of the true number, partly because
of inadequate research effort particularly in some developing
countries with potentially high IAS densities (McGeoch et al.,
2010). The IUCN Red List Index indicates that the adverse
impacts of IAS include increased rates of decline in species
diversity (McGeoch et al., 2010).
Disturbance is a fundamental driver of biodiversity, and
ecosystem structure and function, and may strongly control
ecosystem services delivered. Almost all ecosystems
experience episodic events like floods, droughts and
wildfire. Where disturbance is frequent enough, natural
selection both permits nature to adapt, and some species
may even become dependent on disturbance, and enhance
its frequency (Parr et al., 2014). A prime example is wildfire,
which is of global significance in that it is an important
factor in determining local to landscape scale ecosystem
structure over vast areas of the subtropics and tropics.
Without fire, ecosystem structure and function in fire-
prone regions may alter their biodiversity, structure and
function entirely (Bond et al., 2005). Many plant species are
designed to accelerate fire frequency and intensity (Keeley
et al., 2011). Disturbance is thus an important tool available
in the management of biodiversity, ecosystem structure and
function, and the ecosystem services that result (Folke et
al., 2004). Disturbance is likely to be most strongly affected
by climate (especially in case of fire) as well as socio-
economic scenarios. Fire, droughts and flooding would be
expected with higher frequency under low future climate
change mitigation scenarios. However, for fire it has been
argued that changes in human population density, and
shifts in urban to rural lifestyles affect future burnt area to
the same degree as climate change, through reducing fire
spread (Knorr et al., 2016). However, as more people are
projected to live in fire-prone areas, potentially detrimental
impacts on societies may nonetheless increase (Knorr et
al., 2016).
4.1.5 Considering Indigenous
Peoples and Local Communities
(IPLCs) and indigenous and local
knowledge (ILK) in scenarios
The integration of indigenous and local knowledge (ILK)
into scenarios developed at the regional and global
scales, as well as the assessment of the impacts of
scenarios on Indigenous Peoples and Local Communities
(IPLCs), have been limited and remain a key challenge in
scenario development (Hill et al., 2012; Wohling, 2009).
Varying combinations of indirect drivers, and especially
government policy, can disproportionately impact IPLCs
and their livelihoods. This is particularly significant when
considering scenarios as alternative policy or management
options intended to alter the future state of these (system)
components (IPBES, 2016b). The following examples
provide evidence for the potential benefits that could be
gained from a better recognition of and respect for ILK and
IPLCs in conservation of nature, as well as adaptation to
and mitigation of climate change.
Government policies that (i) define agro-industrial plantations
as forests, (ii) change property systems, including
privatization and land titling over areas of customary tenure,
and (iii) incentivize migration to historically low population
density areas, undermine ILK that promote biodiversity
and human well-being, and traditional land-use practices
(Dressler et al., 2017).
Some cases where governments have recognized IPLC
land rights and pursued climate mitigation policies, such
as through REDD+ projects (Reducing Emissions from
Deforestation and Forest Degradation), have led to thus-
far successful collaborations and demonstrated that ILK
could make significant contributions to future forest and
biodiversity conservation (see also review in chapter 6). For
instance, the case of GuateCarbon, which incorporates
the Association of Forest Producers of Petén (ACOFOP, in
northern Guatemala) as full partners alongside government
entities and international NGOs, has proved a potentially
important model for negotiation, benefit sharing, and
monitoring, reporting, and verification that respects local
land-use practices and values (Hodgdon et al., 2013).
Positive livelihood outcomes have accompanied a pattern
of strong forest protection in areas with community-led
management here.
Studies suggest that policy scenarios such as protected
area designation – including territorial recognition for IPLCs –
could play a significant role in avoiding future deforestation,
such as in the Amazon, despite continued pressures to
downgrade, downsize, and degazette protected areas
(PADDD) for infrastructure development and more intensive
land uses (Forrest et al., 2015; Soares-Filho et al., 2010).
For example, a recent Brazilian moratorium on mega-dams
– long demanded by indigenous groups on ecological and
spiritual grounds – could enhance ecosystem protection,
especially if accompanied by increased support for
forest groups (Branford, 2018), despite continuing plans
for inter-modal transport projects essentially promoting
agro-industry and colonization (Molina et al., 2015). While
the Brazilian Amazon has served as an important testing
ground for recognizing the importance of ILK in forest
management and for REDD+, the continued discounting
of ILK systems in broader land-use policy throws doubt
on the long-term viability of such participative initiatives
(Cromberg et al., 2014; Vitel et al., 2013). Specific major
drivers vary by country and by region, but global demand
for basic commodities and national enabling environments
for investment in forest-rich countries will likely continue
to contribute to terrestrial emissions and biodiversity loss
– including through incursions on IPLCs’ traditional lands
and the attendant loss of ILK. Thus, even where REDD+
and conservation initiatives have tried to ensure community
participation, they achieve variable success, in part because
they often fail to address the strongest indirect drivers
of losses of forests, biodiversity and ecosystem services
(Angelsen et al., 2017).
Notwithstanding these limits, the long period of negotiation
over the program internationally and nationally, in
addition to a pivot away from market-based approaches
implementation, has provided IPLCs with opportunities
to insert their priorities (tenure security, Free, Prior and
Informed Consent, social services) into the debate (Angelsen
et al., 2017; Van Dam, 2011). Increasing rates of recognition
of IPLCs’ rights to inhabit and manage their lands alongside
new sources of dedicated funding (such as the UNFCCC’s
Green Climate Fund) could suggest stronger outcomes for
avoided deforestation and ecosystem health.
4.2.1 Impacts of future global
changes on biodiversity:
feedbacks and adaptation capacity Projected negative changes at all
levels of biodiversity
The scientific community has focused on climate change
as a major driver of concern in exploring possible futures
for nature (Table 4.2.1). Based on our systematic literature
review (Appendix A4.1.1), 88% of the global scenario
literature addressed climate change impacts on nature,
followed by 8% and 2% of the papers addressing land-
use change and natural resource extraction, respectively.
A vast majority of the papers addressed single drivers, as
few integrated models are able to represent combination
of drivers and interactions are more complex to implement
(IPBES, 2016b). Of all the scenarios exploring climate
change impacts, only 18% were combined with other
direct drivers of change such as land use or natural
resource extraction.
Invasive alien
Pollution Others
Climate change 569(270) 4(3) 104(36) 12(6) 8(4) 11(8)
Invasive alien
45(19) 7(4) 4(2) 1(1)
Natural resource
16(7) 1(1)
Pollution 1(1) 1(1)
Others 27(8)
Table 4 2 1 Major drivers represented in global change scenarios addressing impacts on
nature at global scale, across terrestrial, freshwater and marine ecosystems.
The number of scenarios published is reported, and in parentheses, the number of scientific papers from the Chapter 4 literature
database (Appendix A4.1.1). Scenarios addressed single drivers (purple cells) or combination of drivers.
Most scenarios of biodiversity change are terrestrial or
marine, while far fewer exist for freshwater (Figure 4.2.1;
IPBES, 2016b). Therefore, most evidence provided in
section 4.2.3 for freshwater biomes is based on local and
regional studies. Overall, relatively few metrics of biodiversity
and ecosystem function have been explored deeply enough
to draw strong conclusions about their interactions in a
globally changing environment.
The systematic literature review indicates that the effects
of global environmental changes on biodiversity are mostly
projected to be negative (Figure 4.2.1) and embrace
all biodiversity levels – from genetic diversity to biomes
(Bellard et al., 2012; Box 4.2.1). Marine systems are
projected to be generally more negatively impacted by
global change drivers than terrestrial systems (Figure
4.2.1). For example, projected changes in species biomass
or abundance cover the spectrum of negative to positive
trends in terrestrial systems (see evidence provided in
sections to, but negative trends stand
out in marine systems (see section 4.2.2). There are a few
metrics, such as terrestrial C pools or organisms’ growth,
where positive trends are the most common response in
the literature (see In case of C-pools this reflects
chiefly the impact of CO2 on photosynthesis and growth,
which in some models outpace the impacts of warming.
In boreal and temperate regions, climate change was also
shown to possibly have positive effects on organisms’
growth, e.g., plant growth (Pretzsch et al., 2014). All
other metrics of biodiversity and ecosystem function are
dominated by projected neutral or negative trends in
response to projected global change drivers. Negative
trends are particularly dominant for indicators of production,
reproduction success, terrestrial species richness and
extinction, marine species biomass and abundance, and
the area and quality of marine habitats.
Figure 4 2 1
Future trends of selected indicators in marine A, terrestrial B and freshwater
ecosystems C, based on global scale scenarios referenced in the literature
database (Appendix A4.1.1), all drivers combined.
The results are extracted from scenarios with increasing pressures from direct drivers (all climate change scenarios and
business-as-usual scenarios for resource exploitation, land-use change and pollution). The selected scenarios were at global
scale. Regional/local scale scenarios were not referenced in the literature database. Colours code the projected trends in the
indicators. N=the number of trends reported and in parentheses the number of papers.
Species distribution
Area and Shift
Species Richness
Habitat Area and Quality
Organisms’ Size
or Growth
Reproduction Rate
Pest, Invasive
or harmful species
C pool
Species distribution
Area and Shift
Species Richness
Habitat Area and Quality
Organisms’ Size
or Growth
Reproduction Rate
Pest, Invasive
or harmful species
C pool
N = 63 (16)
N = 43 (4)
N = 52 (15)
N = 42 (6)
N = 37 (10)
N = 4 (2)
N = 1 (1)
N = 2 (2)
N = 2 (1)
N = 62 (20)
N = 41 (2)
N = 289 (58)
N = 70 (16)
N = 86 (17)
N = 29 (4)
N = 10 (3)
N = 21 (9)
N = 761 (13)
N = 8 (4)
N = 20 (6)
N = 4 (2)
A substantial fraction of wild species is predicted to be
at risk of extinction during the 21st century due to climate
change, land use and impact of other direct drivers (Bellard
et al., 2012; Pimm et al., 2014; Settele et al., 2014; see
sections 4.2.2-4.2.4). In a recent review of published future
global extinction risk, Urban (2015) found that extinction risk
is projected to increase from 2.8% at present to 5.2% at
the international policy target of a 2°C post-industrial rise,
to 8.5% if the Earth warms to 3°C, and to 16% in a high
greenhouse gas emissions scenario (RCP 8.5; 4.3°C rise).
Extinctions might not occur immediately but after substantial
delay called because when a population has been reduced
to very small numbers, it has a high risk to go extinct at
some point in the future (referred to as «extinction debt»).
This means that long-term effects of global change can be
much more severe than short term impacts (Cronk et al.,
2016; Dullinger et al., 2012; Fordham et al., 2016; Hylander
& Ehrlén, 2013).
Notwithstanding a majority of expected negative impacts
of future climate change on biodiversity, Figure 4.2.1
suggests the potential for some positive effects in species
distributions areas and species richness. General poleward
movement of marine and terrestrial species and upward
movement of terrestrial mountain species may lead to
increase in local species richness in high latitudes and in
mountainous regions, while the opposite is projected in
the tropics and flat landscapes (Gilg et al., 2012; Jones &
Cheung, 2015; Settele et al., 2014; Thuiller et al., 2014).
Global scale scenarios can mask the spatial heterogeneity
of projected biodiversity response at finer scales (Urban,
2015; Vellend et al., 2017). For example, the highest species
extinction risk due to climate and land-use changes is
projected in the tropics and polar regions as well as in top
mountain habitats because of projected “novel” climates
in tropics that these regions have never experienced in the
past (Mora et al., 2013a), narrow physiological tolerances
of tropical and polar species, expected disappearance of
polar and top-mountain habitats (Deutsch et al., 2008;
Gilg et al., 2012; Mora et al., 2013a; Pörtner et al., 2014;
Settele et al., 2014) and the highest risk of conversion of
ecosystems to crops and biofuel in the tropics (Kehoe et al.,
2017; Newbold et al., 2015). Biodiversity hotspots are also
projected as subject to high species extinction (Bellard et al.,
2014; see 4.2.2, 4.2.3, 4.2.4).
To account for the spatial differentiation of global changes
impacts on nature, the following sections 4.2.2, 4.2.3, and
4.2.4 cover the outcomes of the literature database analysis
(Appendix A4.1.1), but also include detailed examination of
key studies and specific biomes (IPBES units of analysis).
The major drivers of change and the primary impacts differ
depending on the biome considered (Figure 4.2.2), and
therefore need to be addressed by specific, and sometimes
local, adaptation and mitigation policies. Future biodiversity adaptation
and reorganisation
Species can respond to environmental changes in many
different ways that are not mutually exclusive. In response
to changes in climate, species can adapt to new conditions,
they can shift their geographical distribution following
optimal environmental gradients or can go locally extinct.
A large number of scenarios explore species distribution
shifts. Terrestrial species may respond to climate changes
by shifting their latitudinal and elevation ranges. Marine
species may respond by shifting their latitudinal and depth
ranges. Models predict latitudinal range shifts for plant and
animal species of hundreds of km over the next century
as well as significant range contraction and fragmentation
(Leadley et al., 2010; Markovic et al., 2014; Meller et al.,
2015; Rondinini & Visconti, 2015; Warren et al., 2013).
Comparisons of projected climate velocity (the rate of
movement of the climate across a landscape) and species
displacement rates across landscapes showed that many
terrestrial species (e.g., plants, amphibians, and some small
mammals) will be unable to move fast enough to track
suitable climates under medium and high rates of climate
change (i.e. RCP4.5, RCP6.0, and RCP8.5 scenarios). Most
species will be able to track climate only under the lowest
rates of climate change (RCP2.6) (Settele et al., 2014).
Natural geographical barriers (Burrows et al., 2014) and
human-made habitat disruptions are predicted as important
factors limiting movement of species ranges (Meier et al.,
2012; Schloss et al., 2012).
Species adaptation to novel conditions is likely to mitigate
the predicted impacts of global changes (Hoffmann & Sgrò,
2011; Lavergne et al., 2010; Neaves et al., 2015; Pauls et
al., 2013; Skelly et al., 2007). Models that ignore adaptation
may overestimate extinction probabilities. For example, the
inclusion of local adaptations due to phenotypic plasticity
and microevolution in models of terrestrial carnivore and
ungulate species decreases the expected decline in
population abundance by 2050, from 31–34% to 18%
(Visconti et al., 2016; see Box 4.2.1)
Intraspecific diversity of behavioral, phenological,
physiological and morphological traits allows populations
and species to survive under rapid climate change through
standing genetic variation (GD1 in Box 4.2.1), and provides
material for selection in new conditions (Alfaro et al., 2014;
Hof et al., 2011; Jump et al., 2009). On the one hand,
incorporating intraspecific variation in species models
increases the likelihood of their survival as shown for several
tree species (Benito Garzón et al., 2011; Morin & Thuiller,
2009; Oney et al., 2013). On the other hand, projections
that do not consider probable loss of intraspecific diversity
can underestimate future negative effects on biodiversity.
The loss of genetic diversity is projected for a number of
Figure 4 2 2
Examples of future projected impacts of major drivers of change on nature
(supporting evidence in sections 2.2 and 2.4 of the chapter, and Table A4.2.1 in
Appendix 4.2).
Examples are given for IPBES terrestrial and marine units of analysis (UoA).
species belonging to very different terrestrial and aquatic
taxa and thus, should be recognized as a serious threat to
future biodiversity rescue (Bálint et al., 2011; Jump et al.,
2009; Neaves et al., 2015; Pauls et al., 2013).
Phenotypic plasticity helps to reduce the risk of species
extinction (GD2 in Box 4.2.1) allowing a rapid (within
individual’s lifetime) adjustment of populations to novel
conditions whereas evolutionary responses require several
generations (Chevin et al., 2010). Incorporating phenotypic
plasticity in models predicting future species’ distributions
reduced the extinction risk in southern populations of several
species (Benito Garzón et al., 2011; Morin & Thuiller, 2009).
Rapid adaptive evolution (GD3 in Box 4.2.1) occurring at
similar time scale as global environmental change has the
potential for “evolutionary rescue”, i.e. population survival in
situ due to ongoing selection of standing genetic variations as
well as relatively slower selection of new mutations (Gonzalez
et al., 2013; Hendry et al., 2011; Hoffmann & Sgrò, 2011;
Settele et al., 2014). However, evolutionary responses may be
too slow for species with low capacity for adaptive evolution,
especially under large-scale and rapid environmental changes
(Gienapp et al., 2012; Jump et al., 2006).
Adaptation can cascade to entire communities or
ecosystems, thus maintaining community properties beyond
the level of change in the driver. However, adaptive capacity
is not unlimited and so even evolving systems can eventually
switch to a new state if a change in a driver is too severe or
too rapid. Return to the original system state when change
pressure is removed to the original state can be harder than
would have been the case without evolution, due to the
depletion of the genetic variation (Figure 4.2.3).
Along with the vital importance of preserving the short-term
adaptive capacity of biodiversity, the necessity of long-term
maintenance of further evolutionary processes generating
biodiversity and potential future ecosystem services was
recognized as a key goal that requires preservation of
evolutionary heritage and phylogenetic diversity of the Tree
of Life (Faith, 2015; Faith et al., 2010; Forest et al., 2007;
Mace & Purvis, 2008).
Reorganization of ecological communities and
novel communities: Substantial changes in species
composition and biotic interactions are expected due to
shifts in species distribution (S1 in Box 4.2.1), local species
extinctions, alterations of species abundance, functioning
and phenology (S2 in Box 4.2.1). Projected changes
in species composition can lead to disruptions of food
webs and mutualistic relationships, increased prevalence
of pests and pathogens, introductions of alien species,
biotic homogenization and loss of biological uniqueness of
communities (Blois et al., 2013; Buisson et al., 2013; Thuiller
et al., 2014).
Figure 4 2 3
Potential role of evolution (more generally, “adaptive capacity”) in mediating
tipping points, alternative stable states, and hysteresis.
more diffi cult
is more
Evolution buffers
effect of drivers
(”delays” tipping
without evolution
with evolution
without evolution
with evolution
Novel (no-analog) communities, in which species will
co-occur in historically unknown combinations, are
expected to emerge (Ordonez et al., 2016; Radeloff et
al., 2015; Williams & Jackson, 2007). Novel communities
are expected to become increasingly homogeneous and
shifted towards smaller size species and generalists with
broader ecological niches (Blois et al., 2013; Lurgi et
al., 2012). Novel interactions can strongly affect species
fitness because species will lack a long coevolutionary
history in new conditions (Gilman et al., 2010; see also
Appendix 4.2). The importance of feedbacks
between hierarchical levels of
Some well described feedbacks between different
hierarchical levels and facets of biodiversity are self-
reinforcing and could likely amplify negative effects of global
changes on biodiversity (Brook et al., 2008). Integration of
processes acting at different organizational biodiversity levels
is essential for future predictions of global change impacts
on nature (Mouquet et al., 2015; Thuiller et al., 2013).
The feedback between population size and genetic
diversity (S4 in Box 4.2.1) is known as an extinction vortex
(Frankham et al., 2014) because the reduction in population
size leads to the loss of genetic diversity which in turn, leads
to decrease in population fitness and adaptability and further
reduction in population size. The feedback between species’
range and genetic diversity (S5 in Box 4.2.1) means that
the contraction and fragmentation of species ranges
are expected to cause genetic loss through decrease in
effective population size and extinction of genetic lineages
as well as extinction of local populations with unique genetic
characteristics (Bálint et al., 2011; Pauls et al., 2013).
Genetic loss, in turn, may decrease species adaptability
and migration capacity. The feedback between species
composition and genetic diversity (SD3 in Box 4.2.1) means
that changes in species composition alter the selection
pressure affecting genetic diversity. For example, reduction
in pollinator abundance could lead to selection favoring
self-fertilization in plant populations, leading to a decrease in
genetic diversity (Neaves et al., 2015). Introductions of alien
species may result in hybridization, out-breeding depression
and decrease in genetic diversity of native species.
However, hybridization may also facilitate adaptation to
novel environments (Hoffmann & Sgrò, 2011). Changes in
genetic diversity, in turn, contribute to further disturbance of
species relationships.
The feedback between species composition and single
species extinctions (SD4 in Box 4.2.1) make changes in
species composition and single-species extinctions modify
the web of interactions at the community level and lead to
cascading and catastrophic co-extinctions called “chains
of extinction” (Bellard et al., 2012; Brook et al., 2008). The
loss of key species as well as invasions and proliferation
of pests and pathogens can have the most drastic effects.
Failing to account for changes in biotic interactions could
cause models to under- or overestimate extinction risks
(Gilman et al., 2010). The feedback between species
composition and species’ capacity to track climate change
(SD5 in Box 4.2.1) implies that interspecific interactions can
modulate the outcome of species range shifts. Mutualistic
interactions, such as plant-pollinator relations, may fail
in tracking fast environmental change (Lavergne et al.,
2010). Competition and predation can both hamper and
facilitate range shifting (Holt & Barfield, 2009; Svenning
et al., 2014). Interactions can slow climate tracking and
produce more extinctions than predicted by models
assuming no interactions (Urban et al., 2013). Moreover,
interspecific interactions can modulate the direction of
species range shifts, for example, species may shift
downslope due to competitive release at the lower margin
of species distribution (Lenoir et al., 2010). Changes in
species distribution, in turn, contribute to further changes
of species composition. The feedback between landscape
homogenization and species extinctions (ED2 in Box 4.2.1)
involves that predicted biotic homogenization and loss of
biological uniqueness of communities within a region (Blois
et al., 2013; Buisson et al., 2013; Thuiller et al., 2014)
can synchronize local biological responses to disturbance
across individual communities and thus, compromise the
potential for landscape- and regional-level disturbance
buffering (Olden, 2006). Taxonomic homogenization of
communities can reduce resistance of a landscape to
future invasions (Olden, 2006). As a result, local extinctions
of native species and invasions of alien species should
be expected that, in turn, will contribute to further biotic
homogenization (for details, see Appendix 4.2).
4.2.2 Marine ecosystems Global state and function of
marine ecosystems and future drivers of
The ocean is central to regulating the Earth’s climate.
The ocean absorbs around 25% of the anthropogenic
emissions of CO2 (Le Quéré et al., 2016), leading to ocean
acidification with a decrease in surface seawater pH of
0.1 units since the beginning of the industrial era (Orr et
al., 2005). The ocean absorbs 93% of the Earth’s excess
heat energy, resulting in warming of 0.11°C per decade
in the upper 75m of the ocean between 1971 and 2010
(Rhein et al., 2013). Oceans are essential to life and
provide major services to human societies. Marine
phytoplankton produce about half of the global O2 (Pörtner
et al., 2014). The ocean supports fisheries and aquaculture
activities and produced on average 104.3 million tons
Box 4 2 1 The main interrelations and feedbacks between hierarchical levels that are
important for the future of biodiversity.
Direct drivers of global change affect all levels of biodiversity,
either directly (coloured arrows) or indirectly through feedbacks
(grey arrows). Even one-way interactions are important for
biodiversity response, while self-reinforcing feedbacks can
potentially signifi cantly increase expected negative effects of
global change drivers (for details, see Appendix 4.2).
Effects of changes in genetic and phenotypic diversity
GD1 – adaptation of populations to new conditions through
standing genetic and phenotypic variations
GD2 – adaptation of populations due to phenotypic plasticity
GD3 – adaptive evolution, “evolutionary rescue” of populations
and species
Effects of changes in functioning, population size and
range of individual species
S1 – changes in local species composition due to alteration of
species range (shift, change in area, fragmentation)
S2 – changes in local species composition due to local species
extinctions and alteration of species abundance and functioning
(including changes in phenology)
S3 – changes in ecosystem structure and functioning due to
changes in key species abundance and functioning
S4 – changes in genetic diversity due to changes in population
S5 – changes in genetic diversity due of alteration in species
range (shift, change in area, fragmentation) and dispersal ability
Effects of changes in local species diversity, species
composition and interspecifi c relations
SD1 – weakening and destabilization of ecosystem functioning
due to loss of local species diversity
SD2 – biotic homogenization as a result of species shift, local
species extinctions and invasions
SD3 – changes in selection pressure because of alteration of
species composition and interspecifi c relations (including effects
of alien species invasions)
SD4 – species extinctions as a result of cascading effects of
alteration of species composition
SD5 – impact of alteration of species composition on species
capacity to track climate change
Effects of changes in structure and functioning
of ecosystems
E1 – the contribution of individual ecosystems to the total
landscape/seascape ecosystem functioning
E2 – disappearance of the most vulnerable ecosystems in
landscapes/seascapes and regions
E3 – reduction of species population size, reduction and
fragmentation of species’ ranges and disruption of population
structure because of habitat loss and fragmentation
Effects of changes in diversity of ecosystems,
heterogeneity of landscapes and seascapes
ED1 – weakening and destabilization of the total landscape/
seascape functioning because of loss of ecosystem/
habitat diversity
ED2 – infl uence of landscape heterogeneity on local
species persistence
ED3 – infl uence of landscape heterogeneity on genetic diversity
and evolution
per year of fish and invertebrates from 2009-2014, which
represented approximately 17% of the animal protein
consumed by humans (FAO, 2016). Oceans supports
rapid socioeconomic development and growth of human
population on coastlines, with increasingly intensive, multiple
uses leading to heavily degraded habitats (Spalding et
al., 2014; Wong et al., 2014). Marine populations and
communities have been impacted at unprecedented
rates by climate change (mainly in the form of ocean
warming, ocean acidification, deoxygenation, and sea level
rise) and direct anthropogenic activities (mainly in the form
of fishing, pollution, and habitat degradation) (Chapter 2;
Hoegh-Guldberg et al., 2014; Poloczanska et al., 2016;
Pörtner et al., 2014).
Globally, none of these pressures are projected to decrease
in the future. Earth System Models have been used to
project future environmental conditions (IPCC, 2013),
showing that the state of the future ocean will strongly
depend on the amount of carbon emitted in the coming
decades (Gattuso et al., 2015; IPCC, 2018). Climate change
is, among other drivers, the main driver considered in global
scale scenarios (Table 4.2.2).
Mean sea surface temperature is projected to increase
by +2.7°C in 2090-2099 as compared to 1990-1999 for the
high emission scenario (RCP8.5), whereas the warming is
limited to +0.71°C for the more stringent RCP2.6 emission
scenario (Bopp et al., 2013); model-mean values from the
Coupled Model Intercomparison Project 5). At the regional
scale, stronger warming occurs in the tropics, in the North
Pacific and in the Arctic Ocean, with the sea surface
warming more than +4°C at the end of the 21st century
under RCP8.5 (Bopp et al., 2013; Collins et al., 2013).
As global temperatures rise, so does the mean sea level
due primarily to the thermal expansion of ocean water and
by melting of glaciers, ice caps and ice sheets. A sea level
model calibrated with empirical data and forced by the IPCC
high emission scenario (RCP8.5) projects a sea level rise
(SLR) of 52-131 cm by 2100 relative to year 2000 (Kopp et
al., 2016).
A broadly uniform decrease of the mean sea surface
pH of -0.33 pH units (model-mean) by the 2090s relative
to the 1990s is predicted under RCP8.5 (Bopp et al.,
2013), which is accompanied by a decrease in carbonate
ion concentration and in the saturation states of calcium
carbonates (e.g., calcite, aragonite), essential components
of shells or skeletons of many marine organisms. The
volume of undersaturated waters with respect to aragonite
is projected to increase between 1990 and 2100 from 76%
to 91% of the global ocean under RCP8.5 (Gattuso et
al., 2015).
Earth system models also project decreasing global
ocean oxygen due to climate change. The mechanisms
at play are a reduction of oxygen solubility due to ocean
warming and the combination of increased stratification
and reduced ventilation that prevents the penetration
of oxygen into the deep ocean (Breitburg et al., 2018).
Deoxygenation will continue over the 21st century
irrespective of the future scenario, with decreases of global
O2 of -1.8% and -3.45% (model-mean) under RCP2.6 and
RCP8.5, respectively (Ciais et al., 2013), with a stronger
drop for the North Pacific, the North Atlantic, and the
Southern Ocean (Bopp et al., 2013). Despite a consistent
global deoxygenation trend across models, there is as yet
no consensus on the evolution of hypoxic and suboxic
Direct drivers of change
Global scale
Open ocean
Polar seas
Shelf ecosystems
Deep sea
Tropical coral
Rocky and
sandy shores
Kelp forests
Climate-related drivers of change
Ocean warming 45%
Ocean acidification 8%
Deoxygenation 4%
Sea ice melt 2%
Sea level rise (SLR) 16%
Extreme events 3%
Direct human-mediated drivers of change
Fishing 16%
Pollution 5%
Maritime transport
Species introduction
Land-use change 1%
Coastal development 1%
Oil and gas extraction, mineral mining
Main direct impacts on nature
Habitat degradation
Biodiversity decline
Species invasion / range shift
Shifts in food webs and biogeochemical cycles
Table 4 2 2 Major climate-related and direct human-mediated drivers of change impacting
marine ecosystems (by IPBES subunits) as highlight in this chapter’s sections to
Cells are colored when there is substantial evidence from the reviewed scenarios and models that drivers have a major impact
on one of the marine ecosystems. Where the information exists, the second column of the table reports the percentage of
marine global scale scenarios implementing changes in the drivers and quantifying impacts on nature, based on our literature
database (Appendix A4.1.1).
waters due to uncertainties in potential biogeochemical
effects and in the evolution of tropical ocean dynamics
(Cabré et al., 2015). Along coastlines, deoxygenation and
the increase of hypoxic “dead zones” are largely driven by
direct human activities (which combine with sea warming),
with rivers draining large nitrogen and phosphorus loads
from fertilized agricultural watersheds, and from sewage,
aquaculture and atmospheric nitrogen deposition,
causing eutrophication and subsequent aerobic microbial
decomposition (Glibert et al., 2018; Levin et al., 2009;
Rabalais et al., 2009).
Future climate change will hence alter marine habitats and
modify biogeochemical cycles. Recent modelling work has
shown that climate change may continue to produce more
hostile conditions and threaten vulnerable ecosystems and
species with low adaptive capacity (Gattuso et al., 2015;
Hoegh-Guldberg et al., 2014; Mora et al., 2013a; Pörtner et
al., 2014; Wong et al., 2014).
Adding to future climate change and potentially amplifying
impacts on marine ecosystems, direct human-mediated
pressures will likely intensify in future. An increase in
fisheries and aquaculture production is plausible as
a response to increasing demand for fish and seafood
(Chapter 11 of the World Ocean Assessment, UN, 2017)
which is expected to arise as a result of population growth
and increasing average income that allows for augmenting
the proportion of fish in the diet (World Bank, 2013). Under
assumptions of increasing technological efficiencies and
increasing demand for fish, the FAO and OECD project
that total world marine seafood production (fishery plus
aquaculture) would exceed 120 million tons in 2025,
or plus 17% relative to 2013-2015. Diverse forms of
pollution (excessive nutrient loads, toxic contaminants,
persistent organic pollutants, plastics, solid waste) will
likely continue to pervade marine ecosystems in the future,
constituting additional threats to living organisms (Bergman
et al., 2012; Geyer et al., 2017; Lamb et al., 2018; Sutton
et al., 2013; Worm et al., 2017). The oceans are sinks
for landborne and airborne inputs of persistent pollutants
which can both travel great distances in the near-surface
water masses (Eriksen et al., 2014) of the open ocean,
and sink into the deeper ocean (Chapter 20 of the World
Ocean Assessment, UN, 2017). In coastal oceanic waters,
increasing nutrient loads and pollution in combination with
warming will likely stimulate eutrophication and increase
the extent of oxygen minimum zones (Breitburg et al.,
2018; Rabalais et al., 2009).
The impacts of global change on marine biodiversity will vary
geographically, with latitudinal gradients of expected in many
global scale scenarios (Gattuso et al., 2015), and depending
on the type of ecosystems (Table 4.2.2). Major drivers of
change in the open ocean pelagic ecosystems that are
included in global scale models and scenarios are climate-
related drivers (sea warming, acidification, deoxygenation),
and fisheries exploitation. Additional future threats included
in scenarios for shelf ecosystems are sea level rise, extreme
events, nutrient pollution and coastal development which
may cause degradation, fragmentation and loss of habitats
(Table 4.2.2).
Future scenarios of climate change impacts on marine
biodiversity at global scales are the most documented in the
literature (78% of the scenarios in our literature database
Table 4.2.2). They will therefore form the main content
of this section (section, with evidence provided
by type of ecosystems (IPBES units of analysis). The rest
of the drivers are much less, or not at all, represented in
scenarios projecting impacts on marine biodiversity at global
scale, even though their historical and current impacts on
biodiversity have been shown to be significant. Moreover,
there are relatively few global scale scenarios involving
multiple pressures on marine ecosystems and biodiversity
(23% of the marine scenarios involve a combination of
multiple drivers in our global scale literature database), so
in addition to updating recent global assessments with
the latest modelling and scenarios work, sections
to report evidence from more local studies of how
direct anthropogenic drivers may combine with climate
change in impacting future marine biodiversity. Future climate change impacts
on marine biodiversity and ecosystem
functioning Climate change impacts in open
ocean ecosystems
Low trophic levels
Net Primary Production (NPP) by marine phytoplankton
is responsible for 50% of global carbon fixation through
photosynthesis, but is also the basis of marine food webs,
controlling the energy and food available to upper trophic
levels. Earth System Models project a mean decrease of
NPP in 2100 under all RCP greenhouse gas emissions
scenarios, ranging from -3.5% to -9% under RCP2.6 (low
emissions) and RCP8.5 (very high emissions), respectively
(Bopp et al., 2013), though there is significant variation
between individual model projections. The global decrease
of NPP is accompanied by a change in the seasonal timing
of peak NPP, with an advance by ~0.5–1 months by 2100
globally, particularly pronounced in the Arctic (Henson et
al., 2013).
The projections are heterogeneous over space with general
agreement that NPP is expected to decrease in the tropics
and in the North Atlantic, and increase at high latitudes
(Bopp et al., 2013; Boyd et al., 2014; Steinacher et al.,
2010). Some regional discrepancies between models
exist, with nonlinear dynamics making some projections
uncertain. In the tropics, the mechanisms at play are largely
model-dependent, with both stratification–driven reduction
in nutrient availability and increases in grazing and other
phytoplankton loss processes (Laufkötter et al., 2015). This
results in large inter-model differences, with the decline in
tropical NPP being projected between -1 and -30% by 2100
under RCP8.5 (Kwiatkowski et al., 2017). Using satellite-
based observations of ocean–colour and an emergent-
constraint relationship, the uncertainties in the decline of
tropical NPP have been reduced with an estimated decline
of -11±6% in 2100 for a business-as-usual scenario
(Kwiatkowski et al., 2017).
In the Arctic, some models project an increase in NPP
because of the loss of perennial sea-ice and an increase of
light availability, whereas other models simulate a decrease
due to increasing ocean stratification and decreasing nitrate
availability (Vancoppenolle et al., 2013). In the Southern
Ocean, models project a zonally-varying response of NPP to
climate change, with a decrease in the subpolar band (50°S
and 65°S), but increases in the Antarctic (south of 65°S) and
in the transitional band (40°S-50°S) (Leung et al., 2015).
Mechanisms at play are changing light availability and iron
supply by sea ice melting (Wang et al., 2014).
Under the SRES A1B scenario, the reduction in zooplankton
biomass was projected to be higher than for primary
production in 47% of the ocean surface particularly in the
tropical oceans, implying negative amplification of ocean
warming through bottom-up control of the food web
(Chust et al., 2014). This impact differs regionally with
positive amplification of zooplankton biomass in response
to the increase of NPP in the Arctic and Antarctic oceans,
thereby increasing the efficiency of the biological pump
in those regions. Other changes in species composition
can be expected under future climate change, such as
shifts from diatom-dominated phytoplankton assemblages
with high POC export efficiencies to smaller, picoplankton
communities characterized by low export efficiencies (Morán
et al., 2015; Smith et al., 2008).
In addition to warming and changes in ocean stratification/
circulation, ocean acidification is also expected to influence
metabolic processes in phytoplankton and zooplankton
species. Laboratory and mesocosm experiments have
shown contrasting responses for different plankton types
under elevated CO2 concentrations, with a stimulating
influence for nitrogen-fixing cyanobacteria (Hutchins et al.,
2007, 2013) and pico-eukaryotes (Bach et al., 2017), but
potential detrimental effects on growth and calcification
rates for some of the main calcifying phytoplankton
(Meyer & Riebesell, 2015). Other potential effects of ocean
acidification include a reduction in microbial conversion of
ammonium into nitrate (Beman et al., 2011), which could
have major consequences for oceanic primary production
and potentially less carbon export to the deep sea. A recent
modeling study incorporating differing growth responses
of phytoplankton types to increased pCO2, has suggested
that acidification effects may even outrank the effects of
warming and of reduced nutrient supply on phytoplankton
communities over the 21st century (Dutkiewicz et al., 2015).
Higher trophic levels
Most published global scale scenarios of change in
higher trophic levels in response to climate change rely on
correlative models examining changes in species’ spatial
distribution (64% of publications on the effect of climate
change on marine biodiversity at global scale in our literature
database, Appendix A4.1.1). These “Species Distribution
Models” (SDMs) (also called ecological niche models or
climate envelope models) analyze the statistical relationship
between species occurrences and a set of environmental
variables (Araújo & New, 2007; Thuiller et al., 2009). SDMs
do not typically consider species adaptation nor the effects
of species interactions.
Using species distribution models for projecting future
climate-induced changes, the main findings at the global
scale are that species will shift their distribution poleward
(Cheung et al., 2009), likely resulting in an increase in
species richness and species invasions in high latitude
regions (the Arctic and Southern Ocean) and conversely
a decrease of species richness in the tropics and the
equator (García Molinos et al., 2016; Jones & Cheung,
2015; Pörtner et al., 2014) and in semi-enclosed seas (e.g.,
Mediterranean Sea, Ben Rais Lasram et al., 2010). A mean
latitudinal range shift of 25.6 km per decade to 2050 was
projected under the high emission scenario RCP8.5, which
reduced to 15.5km per decade under RCP2.6 (Jones &
Cheung, 2015).
Distributional shifts of marine species are the most clearly
detectable pattern that can currently be assigned to climate
change, or more specifically to sea surface temperature
change (García Molinos et al., 2016). This is related to the
sensitivity of marine ectotherms, which constitute the bulk
of high trophic level species, to temperature change. But
ocean warming can trigger additional adaptive responses
such as phenological shifts and physiological changes
in growth and reproduction. It is expected that animals
inhabiting temperate latitudes, where seasonality is strong,
will better adapt to a changing climate whereas polar
stenotherm species will be more vulnerable to warming
(Pörtner et al., 2014). Tropical species, in addition to having
narrow thermal windows, inhabit the warmest waters and
are thus near physiological temperature tolerance limits
that lower their adaptive capacity (Storch et al., 2014)
At low latitudes, open-ocean oxygen-minimum zones
(OMZ) constitute an additional threat to marine organisms,
especially in the eastern tropical Pacific (Cabré et al., 2015)
and along major eastern boundary upwelling systems
(Gilly et al., 2013). The horizontal and vertical expansion of
already large OMZs will potentially affect marine populations
dramatically, through shifts in their spatial distribution and
abundance, as well as altered microbial processes and
predator-prey interactions (Breitburg et al., 2018; Gilly et al.,
2013). The shoaling of the upper boundary of the OMZs can
also trap fish in shallower waters, compressing their habitat,
and thereby increasing their vulnerability to predation and
fishing (Bertrand et al., 2011; Breitburg et al., 2018).
In addition to correlative species distribution models, there
are recently developed integrated modelling approaches
(e.g., end-to-end models combining the physics of the
ocean to organisms ranging from primary producers to top
predators) considering the multiple responses of marine
populations to climate change (based on e.g., physiological
rates, trophic interactions, migration behavior), as well
as essential food web knock-on effects and adaptive
mechanisms to move towards more realistic projections of
marine biodiversity (Payne et al., 2016; Rose et al., 2010;
Stock et al., 2011; Tittensor et al., 2018a; Travers et al.,
2007). At regional and local scales, such models have been
developed with more detailed representation of multiple
taxa of commercial interest or of conservation concern
than at the global scale, where the few existing end-to-end
models represent ecosystems and biodiversity through
large functional groups (e.g. fish biomass, pelagic biomass,
biomass in different size classes) or are focused on single
key species. A global scale end-to-end model run under the
worst-case scenario (RCP8.5) projected that the biomass of
high trophic level organisms would decrease by 25% by the
end of the century (Lefort et al., 2015). This first estimate,
which has been recently confirmed by an ensemble of
global marine ecosystem models (Box 4.2.2), suggests that
the response of high trophic levels amplifies the decrease of
biomass projected for phytoplankton and zooplankton.
Global scale models project that ocean warming may
shrink the mean size of fish by the end of century (Cheung
et al., 2013; Lefort et al., 2015) and lead to smaller-sized
infaunal benthos globally (Jones et al., 2014). This trend
is very robust to the model used in the different studies,
as well as to the mechanisms involved: the decrease in
mean size could be either due to the combined effects of
future warming and deoxygenation on animal growth rates
(Cheung et al., 2013), the combined effects of warming
and food limitation (Lefort et al., 2015), or to the limiting flux
of particulate organic matter from the upper ocean to the
benthos (Jones et al., 2014).
Air-breathing marine species
Marine turtles are particularly vulnerable to climate change
as, being ectotherms, their behavior, physiology, and life
traits are strongly influenced by environmental factors
(Janzen, 1994; Standora & Spotila, 1985). Arguably, the
most detectable impacts will occur during the terrestrial
reproductive phase: incubating eggs are vulnerable to
sea-level and extreme weather events (Fish et al., 2005;
Fuentes et al., 2010), while future changes in temperature
and rainfall at nesting beaches will likely reduce hatching
success and emergence, cause a feminization of turtle
populations, and produce hatchlings with higher rates of
abnormalities (Fisher et al., 2014; Mrosovsky & Yntema,
1980). Future changes in temperature are expected to
impact the frequency and timing of nesting (Fuentes &
Saba, 2016; Limpus & Nicholls, 1988; Saba et al., 2007),
as well as marine turtle distribution (McMahon & Hays,
2006; Pikesley et al., 2015; Witt et al., 2010). Foraging
specialists (i.e. leatherbacks) might be more susceptible to
climate change impacts on the marine food web relative to
foraging generalists (i.e. loggerheads) due to a lesser ability
to switch prey type (Fuentes & Saba, 2016). Ultimately,
impacts will depend on populations’ resilience and ability
to adapt. Some marine turtle populations are already
responding to climate change by redistributing their nesting
grounds and shifting their nesting phenology (Pikesley et al.,
2015). However, it is still unclear whether marine turtles will
be able to fully adapt since climatic changes are occurring
more rapidly than in the past and are accompanied by a
variety of anthropogenic threats (e.g., fisheries by-catch,
pollution) that make them more vulnerable and decrease
their resilience (Fuentes et al., 2013; Poloczanska et
al., 2009).
Seabirds responses to future climate change are
commonly predicted using species distribution models.
Shifts and contractions in foraging habitat could be
particularly problematic for seabirds by increasing energetic
expenditures. For example, the summer foraging areas
for king penguins are predicted to shift southward in
response to an intermediate warming scenario (SRES
A1B), doubling the travel distance to optimal foraging
areas for breeders with likely negative consequences for
population performance (Peron et al., 2012). Poleward shifts
in foraging areas are also projected for seven Southern
Ocean albatross and petrel species under a range of
emission scenarios, with associated range contractions
of up to 70% for wandering and grey-headed albatross
by 2050 (Krüger et al., 2018). For other species (e.g., the
endangered Barau’s storm petrel), climate-driven shifts and
contractions in wintering range are predicted but the overall
population consequences are unclear (Legrand et al., 2016).
Fewer studies have coupled mechanistic population models
with climate projections to estimate future population
trajectories. Cassin’s auklets are predicted to decline by
11-45% by 2100 under a mid-level emission scenario, due
to increased sea surface temperatures and changes in
upwelling dynamics within their foraging range (Wolf et al.,
2010). Contrasting responses to future climate scenarios
were reported in three seabirds (albatrosses and petrel),
Box 4 2 2 Ensemble model projections of marine ecosystem futures under climate change.
Model intercomparison studies use a common set of input
conditions to force a suite of potentially very different models to
then produce an ‘ensemble’ of outputs. These outputs can be
compared to examine differences among models, and provide
a multi-model mean and range of uncertainty for end users.
While such studies are a common tool in the Earth system and
climate modelling communities, their application to biodiversity
and ecosystems, particularly in the marine realm, remains
relatively new.
Fish-MIP (Tittensor et al., 2018b) is the first model
intercomparison project examining the impacts of climate
change on fisheries and marine ecosystems at regional to
global scales using a common set of climate change scenarios.
There have been many different attempts to model the ocean
ecosystem resulting in a large diversity of models with various
purposes – from examining species distributions to ecosystem
structure to fisheries catch potential (Tittensor et al., 2018b).
Fish-MIP provides a common simulation framework and
standardized forcing variables to provide consistent inputs
to these models and prescribes a common set of consistent
outputs for analysis. In the first round of Fish-MIP, the focus
was on examining climate change (rather than fisheries)
impacts on marine animal biomass over the 21st century at
both regional and global scales. Here, marine animal biomass
includes mostly fish, but in some models, invertebrates and
marine mammals are also considered.
The results across six global marine ecosystem models
Macroecological) that were forced with two different Earth-
system models (ESMs) and two emission scenarios (RCPs 2.6
and 8.5) show that ocean animal biomass will likely to decline
over the coming century under all climate change scenarios
(Figure 4.2.4; Lotze et al., 2018; Tittensor et al., 2018b). The
ensemble model means show steeper declines under RCP8.5
(highest emission scenario) than RCP2.6 (high mitigation
scenario), and steeper declines when forced with the ESM
IPSL-CM5A-LR than GFDL-ESM2M. The trajectories from
different ESMs and RCPs remain relatively similar until about
2030 to 2050, after which they begin to diverge markedly.
Thus, by 2100, the model-mean animal biomass is projected
to decline between 3% and 23% (Figure 4.2.4). These
declines are largely driven by a combination of increasing water
temperature and declining primary productivity, and are likely
to impact ecosystem services including fisheries (Blanchard et
al., 2017).
Spatial maps of ensemble projections (Figure 4.2.5; Lotze et
al., 2018; Tittensor et al., 2018b) show broad-scale decreases
in animal biomass in tropical and many temperate regions, and
potential increases in polar regions. While ensemble projections
Figure 4 2 4
Ensemble projections of global ocean animal biomass under different
scenarios of climate change.
Projections represent the multi-model means of six global marine ecosystem models forced by marine environment
change projected by two different Earth-system models: GFDL-ESM2M (solid lines) and IPSL-CM5A-LR (dashed lines)
and two greenhouse gas emission scenarios: RCP2.6 (low emissions; blue) and RCP8.5 (very high emission; red) with no
shing signal imposed (i.e., changes are due only to climate). Shaded areas represent one inter-model standard deviation
(ecosystem models). All percentage changes are relative to a 1990-1999 baseline. The vertical grey line separates historical
and future projections for climate forcing; the vertical dashed orange line represents the 2030 target year for the Sustainable
Development Goals. Data source: Tittensor et al. (2018b); Lotze et al. (2018).
RCP 2.6
RCP 8.5
Change in total biomass [%]
RCP 2.6
RCP 8.5
across many models are more likely to capture plausible trends
than any single model, there was more variation among models
in polar and some coastal regions, suggesting that there is
greater uncertainty about projected outcomes.
The results shown here for global marine ecosystem models are
helpful for describing the global trends but may not capture the
complex dynamics at local and regional scales. Forthcoming
analyses should therefore compare regional projections based
on regional scale models and global models and examine the
variability between regional models to provide projections and
measures of uncertainty at scales better matched to the needs
of resource managers. Moreover, different scenarios of fishing
pressure need to be incorporated to examine interactions
between fishing and climate change impacts.
Figure 4 2 5
Global ensemble mean spatial patterns of change in global ocean animal
biomass under RCP2.6 (low greenhouse gas emissions; top) and RCP8.5 (very
high emissions; bottom) forced by GFDL-ESM2M (left) and IPSL-CM5A-LR
(right) Earth System Models.
Percentage changes are relative to a 1990-1999 baseline. Data source: Tittensor et al. (2018b); Lotze et al. (2018).
owing to differences in life histories and distribution area
(Barbraud et al., 2011). These studies have identified strong
non-linearities in demographic responses, suggesting the
potential for threshold effects under future climate extremes
(Pardo et al., 2017).
Marine mammals, as homeotherms, are physiologically
buffered from some direct effects of temperature rise.
Rising ocean levels from ocean warming and ice melt will
likely lead to a loss of land or ice-based habitat available
for breeding or pupping, particularly for marine mammals
on low-lying atolls or ice-dependent breeders (Baker et
al., 2006; Laidre et al., 2015). A global assessment of
climate change effects on marine mammals used a range
of climate scenarios (warming between 1.1°C and 6.4°C) to
qualitatively rank negative population effects for all marine
mammal species (MacLeod, 2009). It showed that species
tied to land, ice, or facing geomorphic barriers were most
likely to be affected. Climate change impacts in shelf
Tropical Coral Reefs
An unprecedented 3-year (2014-2017) marine heat wave
have damaged most of coral reefs on Earth (75%) with still
unassessed social-ecological consequences (Eakin et al.,
2018). Thermal stress disrupts the relationship between
corals and their algal symbionts, with bleached corals being
physiologically damaged and suffering severe mortality
rate. The number of years between recurrent severe coral-
bleaching events has diminished fivefold in the past four
decades, from once every 25 to 30 years in the early 1980s
to once every 5.9 years in 2016 (Hughes et al., 2018). A
full recovery of mature coral assemblages, source of reef
biodiversity and productivity, generally takes from 10 to 15
years for the fastest growing species (Hughes et al., 2018).
Many reefs, including those of the iconic and well-protected
Great Barrier Reef, have experienced a shift from dominance
of branching tabular species that build 3-dimensional
habitats, towards corals with simpler morphological
characteristics (Hughes et al., 2018). A trophic model
showed that a loss of coral complexity could cause more
than a 3-fold reduction in fishery productivity (Rogers et al.,
2014), due to the preferential settling of juvenile fishes in
unbleached coral habitat (Scott & Dixson, 2016).
In addition to thermal stress, ocean acidification represents
a major threat to marine calcifier organisms like corals,
particularly those building large but low-density skeletons.
A decrease of pH by 0.4 units (expected under RCP8.5;
Hoegh-Guldberg et al., 2014) would translate into a coral
habitat complexity loss of 50%, inducing a decrease in
species richness by 30% for both fish and invertebrates
(Sunday et al., 2017). A seawater pH lowered by just 0.14
units (RCP2.6) would induce a loss of 34% net community
calcification (Albright et al., 2018). Projections anticipate a
shift from a state of net accretion to net dissolution before
the end of the century (Eyre et al., 2018). Anoxic events are
also rapidly increasing in prevalence worldwide and cause
underestimated mass mortality on coral reefs (Altieri et
al., 2017).
To better anticipate and simulate the potential futures of
coral reef habitats, two complementary approaches have
been used. First, laboratory and field experiments try to
estimate the tolerance, acclimatization and adaptability
of coral species and their symbionts to environmental
changes. One of the most striking studies demonstrates
that progressive acclimatization, even to temperatures
up to 35°C, can achieve the same heat tolerance as
expected from strong natural selection over many
generations (Palumbi et al., 2014). This suggests that at
temperatures beyond the thermal limits of coral species,
the rate and speed of temperature change is key to explain
coral bleaching. Experiments also allow testing of the
interactions of multiple stressors. For instance, a 3-year field
experiment deciphered the mechanisms by which elevated
temperatures exacerbate overfishing and nutrient pollution
effects on corals by increasing coral–algal competition
and reducing coral recruitment, growth and survivorship
(Zaneveld et al., 2016).
Second, models attempt to simulate the futures of tropical
coral reefs under various scenarios. A simulation based on
genomic models predicting future evolution and persistence
in a high-latitude population of corals from Cook Islands
(South Pacific) showed a rapid evolution of heat tolerance
resulting in population persistence under mild warming
scenarios (RCP2.6 and RCP4.5) though this adaptation
would not be rapid enough to prevent extinction under
more severe scenarios (RCP6.0 and RCP8.5; Bay et al.,
2017). Other studies based on niche models, that can
also integrate adaptation capacity related coral cover to
environmental variables allowing for projections at global
(Logan et al., 2014) and regional (Ainsworth et al., 2016)
scales. For instance, coral cover on the Great Barrier Reef
was projected to remain lower than 5% before the end
of the century under a high emission scenario (RCP8.5)
(Ainsworth et al., 2016).
Rocky and sandy shores
Straddling the intersection between land and ocean,
rocky and sandy shores are the dominant components of
coastlines globally, are the most accessible of the marine
biomes and supply services in terms of coastal protection,
direct provisioning (food and materials), recreation (tourism,
fishing), spiritual and cultural purposes, and substrate for
aquaculture and infrastructure.
These ecosystems are vulnerable to sea-level rise which
adds to the height of sea-level extremes, such as during
storm surges, and can exacerbate projected changes
in wave impacts (Hemer et al., 2013). Sea level rise can
affect the dynamics of the morphology of beach systems,
as well as increasing coastal inundation risk, leading to
erosion in many cases, as well as increasing threats to
nesting beaches for turtles and seabirds, dune vegetation
and coastal infrastructure and assets (e.g., de Winter &
Ruessink, 2017; Jevrejeva et al., 2016; Pike et al., 2015).
Evidence of species responses to warming oceans are
recorded from sandy and rocky shores globally, showing
that barnacles, mollusks, crabs and macroalgae have
shifted their distributions in response to recent warming
(e.g., Johnson et al., 2013; Pitt et al., 2010; Poloczanska
et al., 2013; Schoeman et al., 2015; Wethey et al., 2011).
For example, the cold-water barnacle Semibalanus
balanoides may disappear from south-western English
shores by 2050 (Poloczanska et al., 2008). The frequency
of temperature extremes is projected to increase in the next
few decades, particularly during summer in regions such
as the Mediterranean (Kirtman et al., 2013), with potential
high ecosystem impact as large-scale mortalities of intertidal
species have been recorded during extreme heat events
(Garrabou et al., 2009; Wernberg et al., 2013). In south-east
Australia, the temperature-driven range extension of the
sea urchin Centrostephanus rodgersii has led to the loss
and overgrazing of kelp beds and a reduction in associated
biodiversity (Johnson et al., 2011; Ling et al., 2015).
Forests of kelp, large brown temperate-coast marine algae,
are themselves directly impacted by climate change. Under
RCP2.6 and RCP8.5 scenarios, models of kelps in the
North Atlantic incorporating changes in temperature, salinity,
and sea ice cover predict northern movement and range
contraction by 2090 (Assis et al., 2017a, 2017b, 2016;
Raybaud et al., 2013). Under RCP8.5, areas such as the
Gulf of Maine, Southern Europe, and the northwestern coast
of Africa would be bereft of kelps (Assis et al., 2017a), a
trend which in some of these systems is already observed
now (Filbee-Dexter et al., 2016; Krumhansl et al., 2016).
The Arctic, conversely, is projected to gain kelps, which
is consistent with observations of kelp increases in areas
that are decreasing in sea-ice cover and hence increasing
in light availability (Bartsch et al., 2016). The area gained is
not projected to counterbalance the area lost. Similarly, in
Japan, models project its southernmost species, Ecklonia
cava, to colonize new northern habitats that are currently
occupied by colder water kelps, due to a combination of
shifting temperatures and increases in grazing by warm
water fishes under all RCP scenarios by 2090. Further
scenario-based modeling efforts are needed for Australia,
New Zealand, the Southern Atlantic, and the Pacific Coasts
of the Americas, where models of climate change’s future
impacts on kelps have been less explored. While modeled
predictions typically report declines or polar movement,
the observed long-term trajectories of kelp forests are
currently mixed (Krumhansl et al., 2016). In some cases,
such as South Africa, this is due to local cooling (Blamey
et al., 2015; Bolton et al., 2012). In others, climate driven
range expansions of urchin predators has also driven local
increases (Fagerli et al., 2014), although the longevity of
this trend is unclear as they can be overridden by physical
drivers (Moy & Christie, 2012).
Coastal wetlands
Coastal wetlands are found along coastlines globally,
and include salt marshes (mostly found along temperate,
boreal and arctic coastlines), mangroves (mostly found in
tropical and subtropical areas), tidal flats, and seagrasses.
They form essential marine vegetated habitats for carbon
sequestration, and coastal protection against increased sea
level rise (SLR) and natural hazards (Alongi, 2008; Duarte et
al., 2013; Fourqurean et al., 2012). They also host a great
diversity of species, playing a major role as nursery and
breeding areas for a wide variety of marine fauna organisms
(Heck Hay et al., 2003), including migratory ones such as
coastal birds (Nuse et al., 2015) or coral reef fish species
(Harborne et al., 2016). Climate changes in the form of
warming, sea level rise and increased extreme events
(e.g. hurricanes) may increase the vulnerability of these
ecosystems in the future. Vegetated coastal habitats are
already declining globally (Duarte et al., 2005), and many
species are threatened with extinction (Polidoro et al., 2010;
Short et al., 2011). The recent IPCC report on « Global
warming of 1.5°C » (IPCC, 2018) assessed that at global
warming limited to 1.8°C above the pre-industrial level, the
risks to mangroves will remain medium (e.g., not keeping
pace with SLR; more frequent heat stress mortality) whereas
seagrasses are projected to reach moderate to high levels of
risk (e.g., mass mortality from extreme temperatures, storm
damage) (Hoegh-Guldberg et al., 2018).
Sea level rise can have large impacts on coastal ecosystems
because of the flat, gentle slope of much coastal land.
Although coastal wetlands are dynamic ecosystems that
can adapt to sea level rise, their capacity to do so is
limited, regionally differentiated and is affected by many
human activities (Kirwan & Megonigal, 2013; Schuerch et
al., 2018; see The response of wetlands to sea
level rise involves landward migration of vegetated areas,
and submergence at lower elevations (Wong et al., 2014).
Acceleration of sea level rise threatens future wetlands
capacity to adapt with occurrence of horizontal retreat,
and vertical drowning, when accretion of sediment and
organic matter cannot keep pace with SLR (Spencer et
al., 2016). A meta-analysis estimated that under RCP2.6,
60% of the saltmarshes will be gaining elevation at a rate
insufficient to keep pace with SLR by 2100, and the loss
could reach 90% under high SLR (RCP8.5) (Crosby et al.,
2016). Such high SLR (1m by 2100) could put at risk 68%
of coastal wetlands in developing countries (Blankespoor
et al., 2014). By contrast, a just published integrated
model, taking into account the capacity of wetlands to
both expand horizontally by inland migration and build up
vertically by sediment accretion, projected less pessimistic
impacts of SLR with the loss of global coastal wetlands
area ranging between 0 and 30% by 2100, depending
on the RCP considered (Schuerch et al., 2018). Sea level
rise and storm surges cause salinity intrusion inland, that
can impact coastal and freshwater wetlands, with various
effects such as decreased inorganic nitrogen removal,
decreased carbon storage, and increased generation of
toxic sulphides (Herbert et al., 2015). Increased salt and
sulphide concentrations induce physiological stress in biota
and ultimately can result in large shifts in communities and
associated ecosystem functions. Because impacts of sea
level rise are so prominent in coastal wetlands (Jennerjahn
et al., 2017), the impacts of temperature rise have been
relatively less explored despite their importance in terms of
ecosystem structure and function (Gabler et al., 2017).
Submerged plants such as seagrass are highly impacted
by temperature extremes. Warming-induced deterioration
of seagrass ecosystems has been observed over recent
decades in the West Atlantic, Mediterranean, and
Australia, with summer temperature spikes often leading to
widespread seagrass mortality (Fraser et al., 2014; Jordà et
al., 2012; Moore & Jarvis, 2008; Short & Neckles, 1999). In
the western Mediterranean Sea, a model relating mortality
rates to maximum sea temperature projected that seagrass
meadows may become functionally extinct by 2050–2060,
under the SRES A1B emission scenario (Jordà et al.,
2012). Climate warming is also affecting other components
of seagrass ecosystems, notably via ‘tropicalization’—
increasing representation of tropical species—among
seagrass-associated fish communities (Fodrie et al.,
2009), with the potential to reduce seagrass biomass and
habitat complexity as tropical herbivorous fishes increase
(Heck et al., 2015). Among the most serious concerns is
rising frequency of disease epidemics and prevalence of
pathogens, which are associated with warming in many
systems, and that could trigger widespread die-offs of
seagrass (Altizer et al., 2013; Harvell et al., 2002; Kaldy,
2014; Sullivan et al., 2013).
Under elevated mean global temperatures, mangroves
are expected to displace salt marshes in many areas
as the limits to mangrove growth imposed by cold
events decrease (Short et al., 2016). Mangroves in the
southeastern US have been projected to expand in area
(Osland et al., 2013), consistent with observed trends
across five continents over the past 50 years (Cavanaugh et
al., 2014; Saintilan et al., 2014). These projections overlook
important differences among mangrove species, and also
depend on mangroves’ ability to successfully migrate
landward (Di Nitto et al., 2014), and to build up sediment
or continue to receive allochtonous sediment inputs from
estuarine or freshwater sources at rates apace with SLR
(Lovelock et al., 2015; Parkinson et al., 1994). In coastal
settings experiencing erosion, an expansion of mangroves
is highly unlikely. On the other hand, expansion is seen in
areas of accelerating sediment deposition due to upstream
land-use changes (Godoy & de Lacerda, 2015). Species
distribution modeling studies have projected geographically
dependent shifts in community composition and species
richness under climate change scenarios (Record et al.,
2013). While species richness is projected to increase in
SE Asia, South America, eastern Australia and parts of
the African coasts, it will likely decline in Central America
and the Caribbean, partly linked to increased intensity and
frequency of tropical storms, as well as in northern Australia
(Record et al., 2013).
Under increased CO2, the productivity of wetlands
vegetation (seagrass, mangrove trees, saltmarsh plants)
is expected to increase in the future (Wong et al., 2014).
Seagrasses are likely to be among the species that perform
better in a more acidified ocean, because their growth can
benefit from increasing dissolved CO2 (Koch et al., 2012).
This simulation result is supported by greater growth rates
reported around natural marine CO2 seeps, where seagrass
sequestered considerably more carbon below-ground
under acidified conditions, suggesting a possible feedback
to reduce the impacts of CO2 injection into marine waters
(Russell et al., 2013). However, there is limited evidence that
elevated CO2 will increase seagrass resistance to warming
(Jordà et al., 2012). For mangroves, increased CO2 has
been linked to variable responses in net primary productivity,
with decreased NPP projected for Laguncularia racemosa
and increased NPP for Rhizophora mangle (Farnsworth
et al., 1996; Snedaker & Araújo, 1998). Such variation
may be due in part to methodological differences, but
may also reflect important variations in regional conditions
(McKee, 2011). Climate change impacts in deep
The deepsea (defined here as >200m depth) covers about
60% of global ocean area and represents the largest
ecosystem in the world (Smith et al., 2009; Watling et al.,
2013), accounting for more than 95% of the volume of
the Earth’s oceans. Deep sea ecological processes and
characteristics (e.g., nutrient cycling, productivity) underlie
the healthy functioning of ocean ecosystems and provide
valuable services to mankind (Thurber et al., 2014).
Many observational studies have shown that present-day
climate change is already impacting deep sea environments
due to increased temperature (Purkey & Johnson, 2010),
deoxygenation (Helm et al., 2011; Keeling et al., 2010;
Stramma et al., 2008, 2012), lowered pH of intermediate
deep-waters (Byrne et al., 2010), and altered particulate
organic carbon (POC) flux to the seafloor (Ruhl & Smith,
2004; Smith & Stephenson, 2013). Elevated seafloor
temperatures (3.7°C at the bathyal seafloor by 2100 under
RCP8.5; Mora et al., 2013b; Sweetman et al., 2017) will
lead to warming boundary currents which has the potential
to massively release methane from gas hydrates buried
on margins (Johnson et al., 2015; Phrampus & Hornbach,
2012), especially in the Arctic, with simultaneous effects
on water column de-oxygenation and ocean acidification
(Biastoch et al., 2011; Boetius & Wenzhöfer, 2013). Along
canyon-cut margins such as those that occur in the western
Mediterranean, warming may additionally reduce density-
driven processes, leading to decreased organic matter
transport to the seafloor (Canals et al., 2006).
Climate change is also likely to increase wind-driven
upwelling in eastern boundary currents, stimulating
photosynthetic production at the surface (Bakun,
1990; Bakun et al., 2015; Wang et al., 2014). This new
production may, however, decay as it sinks and increase
biogeochemical drawdown of O2. Upwelling may also
bring low-O2, high-CO2 water onto the shelf and upper
slope (Bakun, 1990; Bakun et al., 2010; Feely et al., 2008;
Sydeman et al., 2014; Wang et al., 2014). The expansion of
hypoxic zones is expected to affect many aspects of deep-
sea ecosystem structure and function (Gooday et al., 2010).
As O2 levels decline, many species of deep water
octocorals (including gorgonians and pennatulaceans)
which provide habitat for a diverse array of invertebrates,
are expected to decrease in abundance (Buhl-Mortensen
et al., 2010; Etnoyer & Morgan, 2005; Murray Roberts et
al., 2009). Acidification of deep waters has been projected
to negatively impact cold-water stony corals (Scleractinia),
particularly in the North Atlantic (Tittensor et al., 2010).
Single stressors like warming will also limit tolerance
windows for other stressors such as low O2 or low pH
(Pörtner, 2012; Pörtner & Knust, 2007).
With the projected global reduction in the biomass of
phytoplankton in the upper ocean (Bopp et al., 2013;
section, the flux of particulate organic carbon
(POC) to feed open ocean seafloor communities is
expected to decrease, causing potential alterations of
the biomass, composition and functioning of the benthic
communities. Reductions in seafloor POC flux will be most
drastic in the oceanic gyres and equatorial upwelling zones,
with the northern and southern Pacific Ocean and southern
Indian Ocean gyres projected to experience as much as
a 32–40% decline in POC flux by the end of the century
(CMIP5, RCP8.5; Mora et al., 2013b; Sweetman et al.,
2017). Recent studies have suggested that the NE Atlantic
Ocean could also undergo similar reductions in POC flux
(Jones et al., 2014). The abyssal ocean is highly sensitive to
changes in the quantity and quality of POC flux that could
affect the biomass of benthic microbial and faunal biomass,
and cause dramatic reductions in the sediment mixed-layer
depth, benthic respiration, and bioturbation intensity (Jones
et al., 2014; Smith et al., 2008; Sweetman et al., 2017).
These changes have the potential to feed back on global
carbon cycling and ultimately C-sequestration (Thurber et
al., 2014). Climate change impacts in polar seas
Rising temperatures are projected to reduce sea ice extent
and volume in the Arctic and Antarctic, some of the fastest
warming places on Earth (IPCC, 2013). The rapid rate at
which sea ice retreats in polar seas implies major changes
to be expected in the future for biodiversity and ecosystem
function (Gutt et al., 2015; Larsen et al., 2014; Wassmann et
al., 2011). All components of the food webs will potentially
be impacted, from phytoplankton to top predators, and from
pelagic to benthic species.
Multiple lines of evidence show that ice-melting is likely to
increase primary productivity in polar seas due to increased
light availability, although this could be dampened by a
decrease in nutrient supply due to enhanced water column
stratification that is expected from warming and freshening
of surface waters (section; Hoegh-Guldberg et
al., 2018; Larsen et al., 2014). It has also been shown that
the increased production of floating icebergs, enriched with
terrigenous material, might significantly elevate nutrient
levels and primary production (Smith et al., 2007). However,
while primary production may increase in polar seas in the
future, warmer waters can cause a shift in the composition
of the zooplankton community, such as the shift from
Calanus glacialis towards dominance of the smaller, less
energy-rich Calanus finmarchicus in Arctic waters (Kjellerup
et al., 2012), with potential huge consequences up the
food chain. By contrast, in coastal areas, the production
and transport of organic matter to the seafloor may decline
because glacial meltwater and erosion of melting tundra
(Węsławski et al., 2011) will likely enhance water column
turbidity, which results in decreased water column light
levels (Grange & Smith, 2013; Sahade et al., 2015). The
increased sedimentation in deep coastal areas, particularly
in Arctic fjords, may also smother or clog the breathing
and feeding apparatus of sessile suspension-feeders (e.g.,
corals and sponges), induce O2 stress, but may favour
ophiuroids and capitellid polychaetes (Sweetman et al.,
2017; Wlodarska-Kowalczuk et al., 2005).
Changes in primary production and resulting POC flux
to the seafloor will have impacts on ecosystem structure
and function. Elevated POC flux increases the abundance
and diversity of benthic communities, the prevalence of
habitat-forming taxa (sponges, benthic cnidarians), and the
extension of species ranges into deeper waters (De Rijk et
al., 2000). It could also trigger the switch from dominance by
bacteria to dominance by metazoans for processing benthic
organic matter with bottom-up consequences on the
food-web (Sweetman et al., 2014). Changing ice regimes
may also result in physical disturbance of the deep sea, as
large icebergs can scour the sediment down to 400m on
the Antarctic shelf, enhancing seafloor heterogeneity and
creating hard substrates for sessile megafauna (Meyer et
al., 2015, 2016; Schulz et al., 2010). In the longer term,
iceberg scouring and dropstone deposition will tend to
elevate diversity on regional scales through (re)colonization
processes, although the immediate effect of scouring will be
local elimination of many species (Gutt & Piepenburg, 2003;
Gutt et al., 1996; Thatje et al., 2005).
Sea ice melting is also expected to impact species up
the food-web, and especially those marine mammals
and seabirds depending on ice as haul-outs, but future
scenarios are available for just a few emblematic species.
Demographic models predict that changes in Antarctic
sea ice will substantially reduce the abundance of global
emperor penguin (Aptenodytes forsteri) by 2100 under a
mid-range emission scenario (Jenouvrier et al., 2014), even
when complex dispersal processes are included (Jenouvrier
et al., 2017). A high probability of extinction is foreseen for
the polar bear (Ursus maritimus) subpopulation of southern
Beaufort under SRES A1B scenario by the end of the
century, due to the decrease in the cover, the duration
and the thickness of sea ice (Hunter et al., 2010), but low
probability of extinction has been attributed for all polar
bears in the Arctic (Larsen et al., 2014). However, a recent
study showed that the high-energy requirements of polar
bears could endanger their survival in extended ice-free
periods (Pagano et al., 2018).
Ocean acidification is another major stressor which will be
enhanced in polar regions because of the higher capacity
of seawater to absorb CO2 at low temperatures, resulting
in lower pH and under-saturated waters in aragonite and
calcite (Hoegh-Guldberg et al., 2014; Orr et al., 2005). This
may impact the growth and survival of calcifying shelled
organisms such as Arctic pteropods, foraminifera in the
Southern Ocean, and the recruitment of Antarctic krill
(Euphausia superba), all of those species being essential
prey species at the basis of food-webs (Kawaguchi et al.,
2013; Larsen et al., 2014; Trathan & Hill, 2016). Adding to
the negative impacts of acidification, a combination of ice
retreat and changes in primary production is projected to
decrease Antarctic krill suitable habitat and survival rate
(Piñones & Fedorov, 2016) with potential cascading effects
on their many predators (Trathan & Hill, 2016). Future impacts of fisheries
exploitation on marine ecosystems
In addition to exposure to climate change, marine animal
populations will likely undergo increased fishing pressure as
a result of increasing demand for fish products (World Bank,
2013) particularly in the developing world (Figure 4.2.6;
FAO, 2016). This will largely be driven by growth of human
population that is projected to reach 9.8 billion people by
2050 (UNDESA, 2017) and by income growth in low- and
middle-income countries (Vannuccini et al., 2018). The rate
of increase in demand for fish has been more than 2.5 per
cent per year since 1950 and is likely to continue in the
future (HLPE, 2014). The world fish production (capture and
aquaculture) was projected to increase by 17% between the
base period (2013-2015) and 2025 (FAO, 2016). With the
growing demand, commercial fishing activities are likely to
expand to all areas of the globe.
Scenarios that include governance in fisheries
management, human consumption of seafood, and
advancement of fishing technologies (Squires &
Vestergaard, 2013) are starting to be integrated into global
scale projections. For example, a simple surplus production
model applied to a set of 4713 fisheries worldwide showed
that a business-as-usual fisheries management scenario
would increase the proportion of overexploited populations
by ca. 30% in 2050 (Costello et al., 2016). In contrast, in a
scenario where long-term economic benefits are optimized,
such as through rights-based fisheries management, the
majority of exploited fish populations (98%) would recover
to a healthy status, with a median time of recovery of
about 10 years. Similarly, under the high emission scenario
RCP8.5 and the SSP3 scenario (characterized by low
economic development and a large increase in human
population), maximizing the long term economic yield of
the fishery was projected to increase the biomass of the
skipjack tuna population (Dueri et al., 2016). Recently, it
was shown that reforming fisheries by adopting an optimal
harvest policy that maximizes long-term economic benefits
and that adapts its management strategy to climate-
induced changes in fish biomass and spatial distribution
could offset the detrimental impacts of climate change on
future fish biomass and catch under most RCP greenhouse
gas emission scenarios, except RCP8.5 (Gaines et al.,
Figure 4 2 6
Projections of additional fi sh consumed in 2025 (from fi sheries and aquaculture)
per world region.
Developing countries are projected to eat 93 percent of the additional fi sh available for human consumption. Source: OECD
and FAO (FAO, 2016).
Latin America
and Caribbean
North America
Viet Nam
Other Asia
2018). This important finding needs to be consolidated by
further investigations in a context where fisheries maximum
catch potential is projected to decrease by 2.8-5.3% and
7-12.1% by 2050 relative to 2000 under RCP2.6 and
RCP8.5, respectively (Cheung et al., 2018).
In addition to climate change (see, heavy fishing
also impacts fish size, decreasing both the maximum size
of species and the biomass of large-sized species because
(i)high-value target species are generally larger, (ii) fishing
gear is size-selective and often designed to remove larger
fish, (iii) older and larger fish in a population become fewer
as a result of accumulation of fishing mortality rate through
time, and (iv) large species are more vulnerable because
their life-history traits are generally linked to lower potential
rates of increase (Shin et al., 2005). Under heavy fishing,
a SRES A1B climate change scenario was reported to
magnify the reduction in fish size (Blanchard et al., 2012).
This shift towards smaller fish size and higher growth rates
could ultimately increase the variability of fish biomass
(Hsieh et al., 2006).
Species targeted by fisheries are not the only species
impacted by different fishing scenarios. Long-lived and
vulnerable species such as marine mammals, turtles
and birds suffer from direct impact of fish harvest though
bycatch, and so their future is tightly linked to the long-term
fishing strategies adopted. The interaction with climate
change is complex to resolve but some studies have
started addressing the potential synergistic effects. Some
models based on species distribution projected that climate
change will alter the future distribution of both fisheries and
seabird populations, altering the rates of future bycatch
and hence seabird mortality rates (Krüger et al., 2018). For
some species, spatial overlap with fisheries may decline,
reducing rates of incidental mortality associated with
human activity. However, for two highly threatened seabird
species (grey-headed and wandering albatross), severe
range reductions and increased overlap with fisheries
are projected.
In addition to scenarios of fishing management, the
future status of wild fish populations cannot be envisaged
without considering alternative scenarios of aquaculture
development which will play a major role in sustaining the
supply of seafood products and the maintenance of per
capita fish consumption (Delgado et al., 2003; FAO et
al., 2018). But the development of aquaculture is partly
dependent upon the exploitation of low trophic level fish
species which supply fishmeal for farmed fish.
Aquaculture development could potentially reduce fishing
pressure on wild fish populations, but not to an extent
that could compensate for projections of increases in
demand for seafood products and fishing technology,
both of which result in increased fishing pressure (Quaas
et al., 2016). Taking into account projections in human
population, climate change (IPCC A1B scenario), and
technological development in aquaculture, a bio-economic
model projected that if fishmeal prices increase, this would
encourage fishers to maximize their short-term economic
profits and exceed yearly quotas, leading to collapse of
exploited fish populations (Merino et al., 2012). Given the
current increasing trends of fishmeal prices (Merino et
al., 2010), this implies that compliance to strict fisheries
management and market stabilization measures need to be
seriously considered to maintain exploited populations at
sustainable levels. Likewise, another bio-economic model
run under contrasted archetype scenarios suggested that
relative to climate change impacts, fisheries regulation is
the most important factor in determining the future of fish
populations (Mullon et al., 2016). However, the interplay
between drivers of change cannot be ignored in fisheries
management strategies (see example in Box 4.2.3). A
multi-model ensemble approach allowed to show that the
risk of negative synergistic effects between changes in
primary production and in fishing effort was higher for small
forage fish species (Fu et al., 2018).
Box 4 2 3 Synergistic impacts of multiple drivers on tropical coral reefs.
Tropical coral reefs share a history of strong dependence on
natural and human systems (Maire et al., 2016) that must
be accounted for in attempts to maintain long-term human
development and well-being, and marine biodiversity (Cinner
et al., 2016). Indeed, coral reefs support the nutritional and
economic needs of people in many developing countries.
Their exceptional biodiversity translates directly into biomass
production and thus food security (Duffy et al., 2016). However,
coral reefs face multiple and considerable challenges from
ocean warming (see, ocean acidification, pollution,
overexploitation and destructive fishing practices. More than
80% of the world’s coral reefs are severely over-fished or
have degraded habitats, thus imperiling the livelihood and
sustenance of coastal human populations (McClanahan et
al., 2015). This negative spiral is likely to accelerate in the
future due to the synergistic effects of climate change and
direct human impacts. For example, nutrient loads from
the land increases the vulnerability of corals to bleaching
(Vega Thurber et al., 2014). Plastic debris were estimated
to increase coral susceptibility to diseases from 4% to 89%
with structurally complex corals being eight times more likely
to be affected by plastic (Lamb et al., 2018) inducing a loss
of fish productivity (Rogers et al., 2014). Tipping points exist
at which coral reef ecosystems can shift to being dominated
by macroalgae (Holbrook et al., 2016), with low resilience,
reductions in biodiversity and degradation of the many
ecosystem services they provide, such as reef-associated
fisheries and tourism. However, there are opportunities for
improving the status of coral reefs by the combined action of
reducing both greenhouse gas emissions and overfishing of
species which help the recovery of coral reefs by grazing their
algal competitors (Figure 4.2.7; Kennedy et al., 2013). Robust,
integrated models that can account for combinations of
multiple impacting drivers are still lacking, but these are needed
to simulate the dynamics of coral reef social-ecological systems
on a long-term basis and better anticipate their futures. This
challenge is even more difficult given the multispecies nature
of fisheries, the complexity of trophic interactions, and the time
scales on which different processes determine the trajectories
of coral reef social-ecological systems and the boundaries
beyond which they collapse.
Figure 4 2 7
Future carbonate budgets (proxy for net production of corals skeletons) of
Caribbean coral reefs under climate change and acidifi cation scenarios (top
panel: high RCP8.5 greenhouse gas emission scenario, bottom panel: strong
mitigation RCP2.6 emission scenario), without or with local conservation of
grazing fi sh (parrot fi sh symbol in B, D, G, H).
Initial conditions of reefs are either degraded with 10% coral cover ( A, B, E, F) or healthier with 20% coral ( C, D, G,
H). Vertical blue bars indicate point at which the projected budget becomes negative (erosion of corals skeleton exceeds
production). Source: Kennedy et al. (2013).
TIME (yrs)
TIME (yrs)
645 Future impacts of pollution on
marine ecosystems Persistent organic pollutants and
plastics: another ‘Silent Spring’?
Over the last century the human enterprise has fundamentally
altered the planet by releasing large quantities of persistent
organic pollutants (POPs) into the environment. These
synthetic organic compounds have harmful and toxic
properties and are not readily metabolized by bacteria
or other life forms, thus prolonging their presence in the
environment. Concerns about their effects on wildlife and
people were first raised by Rachel Carson’s book ‘Silent
Spring’ (Carson, 1962), highlighting the devastating effects
of organochlorine POPs on birds and aquatic animals in
particular. As a result, many POPs were tightly regulated or
banned under the Stockholm Convention (UNEP, 2001), and
their production has ceased or decreased for most listed
substances. Large historical burdens of these pollutants
still circulate in the environment however (Harrad, 2009),
and novel substances get synthesized at a rapid pace, with
potentially harmful effects.
Synthetic organic polymers (plastics) form another class
of pollutants that share certain properties with POPS
in that they persist and accumulate in the environment,
can be transported over long distances (reaching remote
polar regions for example; Science for Environment
Policy, 2017), and can have harmful effects on wildlife and
people. In contrast to POPs, their production numbers
are much higher overall and still increasing, thus global
concerns about plastic pollution now match or exceed
those for other POPs, particularly with respect to the
marine environment which forms a sink for discarded
plastic waste (Jambeck et al., 2015; Worm et al., 2017).
Annual plastic production now exceeds 330 million metric
tons (Mt) (PlasticsEurope, 2015), with a cumulative burden
of 8300 Mt produced since 1950 (Geyer et al., 2017),
approximately 6300 Mt of which has been discarded (9%
recycled, 12% incinerated, and 79% ended in landfills or
the natural environment). If current production and waste
management trends continue, roughly 12,000 Mt (million
tons) of plastic waste will be in landfills or in the natural
environment by 2050 (Figure 4.2.8). If evenly spread
around the globe, this would equal a burden of ~24 tons
of plastic waste for each square kilometre of land and
sea surface. This level of pollution in terms of volume and
persistence has no previous analogue in human history.
Negative impacts on the planet and people are becoming
more profound (Figure 4.2.9) as exposure to plastic
pollutants intensifies. As an example, about 90% of
seabirds examined today have plastic in their gut, with
Figure 4 2 8
Current global trends and likely future trajectories of total plastic waste
generation and management.
After data in Geyer et al. (2017).
Primary waste
All waste discarded
All waste incinerated
All waste recycled
100% expected to be exposed by 2050 (Wilcox et al.,
2015). Sea turtles are similarly affected (Schuyler et al.,
2015), as are at least 693 other marine species that have
been recorded to be compromised by plastic pollution
(CBD, 2016). Much of the plastic is released as or broken
down into small microplastic (1 µm-1mm) or nanoplastic
(<1µm) particles. While the harmful effects of microplastic
debris are well understood, the long-term effects of the
smallest fragments are only now emerging (Galloway &
Lewis, 2016), including their tendency to interact with
other pollutants (GESAMP, 2015), facilitate diseases
(Lamb et al., 2018), and transmit through the food chain
(Figure 4.2.9).
Clearly, another ‘Silent spring’ scenario seems plausible, if
effects on numerous wildlife species continue to accelerate
further. Because plastic persists and accumulates in the
environment in similar ways POPs do, a zero-net-release
policy that builds upon the successful Stockholm Convention
(SC) on Persistent Organic Pollutants (POPs) may be a
promising strategy to mitigate the risk posed by current and
future levels of plastic pollution. Yet, in contrast to traditional
POPs, which are largely emitted by industry, plastic pollution
touches every person’s life, and requires a broader societal
effort including designers, producers, regulators, and
consumers of plastic products to engage in comprehensive
solutions (GESAMP, 2015; Worm et al., 2017). Nutrient loads and eutrophication
Numerous model projections show that coastal zones in
many world regions are almost certain to see increases
in nitrogen (N) and phosphorus (P) from increasing river
loads in the coming decades (Sutton et al., 2013; Figure
4.2.10). In contrast, silica (Si) river export is decreasing
globally as a result of retention in the increasing number
of reservoirs in the world’s river systems and this trend
will also continue in many parts of the world. The result of
these simultaneous changes of N, P and Si will continue
to alter nutrient stoichiometry, affecting not only total algal
growth but also biodiversity in coastal waters, including
the propensity for harmful algal blooms (HABs). The
enhanced primary production in coastal surface waters
can cause eutrophication, with subsequent sinking of
excess degradable organic matter to bottom waters
where aerobic microbial decomposition reduces oxygen
concentration. The decline in oxygen concentrations due
to nutrient loads in coastal waters will likely be exacerbated
with climate change, due to decreased oxygen solubility in
warmer waters and decreased oxygen transport to deeper
waters because of stronger stratification of the water
column (Breitburg et al., 2018). The expansion of areas
of low oxygen will impact marine biodiversity at all levels
from individuals’ physiology and behavior, to populations’
demography and range shifts with consequences for
Figure 4 2 9
Possible pathways by which plastic pollutants of different size classes enter the
food chain and propagate to higher trophic levels, including humans.
After Worm et al. (2017).
Microplastics (1 µm–5 mm)Nanoplastics (< 1 µm) Mesoplastics (5–200 mm) Macroplastics (> 200 mm)
Plankton hsiF eavraL airetcaB
Filter feeders
slammam eniraM sdribaeS
srotaderp poT sredeef tisopeD Benthic predators
10 µm
species assemblages and food-webs (Levin et al., 2009;
Pörtner et al., 2014).
Storylines developed by the IPCC and the Millennium
Ecosystem Assessment and translated into changes of the
main anthropogenic drivers, i.e. economic development,
demography and land use (Alcamo et al., 2007), have
been applied to project conditions to 2050. Although
each storyline has different assumptions, they show major
increases in N and P river export especially in South and
Eastern Asia, in South America and Africa where fertilizer
use will likely increase to support the population, and
where urbanization and lagging treatment of wastewater
and sewage connection will lead to increasing nutrient
discharge to surface water (e.g., Glibert et al., 2018). In
contrast, stabilized or decreasing trends in nutrient loads
are projected in Europe, North America and Australia owing
to the development of improved wastewater treatment
systems, and improved nutrient management reducing
NH3 volatilization, leaching and run-off. In these regions,
improvements in hypoxia and frequency or magnitude of
HABs may be realized.
However, the trajectory of nutrient loads is additive with
other global changes, such as temperature rise, which will
alter stratification of the water column, availability of nutrients
and their forms and ratios, and pCO2, among other factors
(e.g., Boyd & Doney, 2003). Recent models supported
evidence for increased eutrophication together with climate
changes, and therefore the propensity for the worsening of
HABs and/or hypoxia by the end of the century (Sinha et
al., 2017). Multiple combined changes such as increases
in nutrient pollution, in global temperature and in reservoir
capacity resulting in increased retentiveness of rivers, require
proactive management to stabilize or reduce the impacts
of eutrophication, including hypoxia and the frequency
of HABs. Future impacts of coastal
development on marine ecosystems
Direct human-related drivers of change such as
urbanization, coastal development, and land-use change
will bring challenges to coastal ecosystems in addition
to climate change. Coastal populations are increasing
disproportionately relative to the global population increase.
Many of emerging cities are on the coast and their growth
will add to the 75% of the world’s mega-cities which are
already coastally located (World Economic Forum’s Ocean
Programme, 2017). Over 2.6 billion people live on or near
the coast, many in developing countries where dependence
on coastal resources may be high and demand for multiple
benefits such as food, coastal protection and income,
will continue to grow as human populations expand (Bell
et al., 2009; Sale et al., 2014). Some 1.36 billion live on
tropical coasts, and this is projected to grow to 1.95 billion
Figure 4 2 10
Trends in global mineral fertilizer consumption for nitrogen and phosphorus and
projected possible futures.
Source: Sutton et al., 2013.
by 2050, with associated pollution and eutrophication of
coastal waters and degradation of coastal ecosystems
(Sale et al., 2014). Urbanization and coastal development
can restrict the capacity of coastal ecosystems to adapt
to rising sea levels e.g. through the “coastal squeeze”
(Wong et al., 2014). Along urbanized coastlines, the
resilience of wetlands to SLR will depend on the availability
of accommodation space (Schuerch et al., 2018) and
sediment supply (Lovelock et al., 2015) which are reduced
by anthropogenic infrastructure barriers (e.g., flood
protection structures, roads, settlements). Future expansion
of coastal development will also bring risks to iconic and
threatened species. For example, the expansion of artificial
lighting at night from coastal development interrupts the
sea-finding behaviour of sea turtle hatchlings and ultimately
survivorship (Gaston & Bennie, 2014; Kamrowski et
al., 2014).
Future projections show a multiplicity of human stressors
acting simultaneously with direct climate-induced changes
on social-ecological systems. Stressors from population
growth and coastal development such as nutrient run-
off, urbanization, and land-use change are expected to
increase and combine with climate stressors such as sea
level rise and warming to exacerbate risks for rocky and
sandy shores, and seagrasses (Box 4.2.4). Models show
that mangroves are particularly threatened by projected
coastal development, with the main direct drivers including
the expansion of aquaculture (prevalent in both Asia and
Latin America) and agriculture (mostly rice cultivation and
pasture), extraction of timber and related forest products
(e.g., for charcoal and domestic construction), and
infrastructure development and alterations of freshwater
flows (e.g., for due to settlements, transportation networks
or dams) (Roy Chowdhury et al., 2017). Under projected
changes, coastal adaptation options will involve increasingly
difficult trade-offs in future among multiple development and
biodiversity objectives (Mills et al., 2015).
4.2.3 Freshwater ecosystems Freshwater biodiversity and
current threats
Freshwater ecosystems provide fundamental services to
humans such as food, water, nutrient retention, recreation,
and climate regulation. Globally, freshwaters (i.e. rivers,
lakes, wetlands) represent less than 0.02% of Earth’s water
volume and cover only about 0.8% of Earth’s surface
(Dawson & Dawson, 2012). However, an estimated 129,000
species live in freshwater ecosystems, representing ~8%
of Earth’s described species (Balian et al., 2008; Figure
4.2.11). The relative contribution of freshwater ecosystems
to global biodiversity is thus extremely high (Tedesco et al.,
2017; Wiens, 2016). Climate, productivity and area size
drive freshwater diversity patterns globally despite profound
functional differences between taxa (Moomaw et al., 2018;
Tisseuil et al., 2013).
Box 4 2 4 Synergistic impacts of multiple pressures on seagrass meadows.
Direct human-related drivers of change such as urbanization,
coastal development, and land-use change will bring
challenges to coastal ecosystems. For seagrasses, key threats
include sediment and nutrient run-off from upstream land-
use change, physical disturbance, algal blooms, and invasive
species, as well as climate warming and disease (Orth et al.,
2006; Waycott et al., 2009). Requirements for clear water and
low nutrient concentrations make seagrasses vulnerable to
eutrophication, as nutrient and sediment loading reduce light
availability and favor faster-growing algae (Burkholder et al.,
2007; Duffy et al., 2013). The protected embayments in which
seagrasses grow best are also prime real estate for coastal
and harbor development. As a result seagrasses are declining
worldwide, and roughly 30% of global seagrass cover has been
lost since the first estimates were made in the late 19th century,
with loss rates increasing in recent decades (Waycott et al.,
2009). Ten of the 72 known seagrass species on earth are at
elevated risk of extinction and three species are classified as
Endangered (Short et al., 2011).
Perennial organisms such as seagrasses are vulnerable to
human disturbance and, under repeated impacts, often
yield dominance to faster growing, opportunistic species
such as fleshy and filamentous algae. In the Baltic Sea, for
example, dominance by eelgrass and rockweed has yielded
over recent decades to accumulations of ephemeral algae
(Bonsdorff et al., 1997). Long-term field monitoring suggests
that exploitation of piscivores such as cod in offshore waters
has released the smaller inshore fishes—mesopredators—
from top-down control, and their consumption of grazing
invertebrates indirectly led to algal blooms and decline
of perennial seagrasses (Eriksson et al., 2011). Coastal
vegetation, including seagrasses, protects coastal human
communities against storm damage, and the continuing
decline of these natural barriers will likely be aggravated by
SLR. Coastal habitat loss exacerbates damage from storms
and flooding in coastal communities (Gedan et al., 2011).
Mapping the risk of such hazards along the coastline of the
USA shows that, under several projected climate scenarios,
the number of people, especially the poor and elderly, and the
total value of residential property exposed to hazards could
be reduced by half by preserving existing coastal habitats
(Arkema et al., 2013).
Current major threats to freshwater biodiversity include
climate change, habitat modification and pollution from land-
use, habitat fragmentation and flow regime homogenization
by dams, non-native species, increased eutrophication
resulting from nutrient and organic discharges, water
abstraction, and overexploitation (Young et al., 2016).
Those threats currently affect freshwater biodiversity and
functioning to varying degrees (Carpenter et al., 2011;
Vörösmarty et al., 2010), and their additive and potentially
synergistic effects may further threaten future freshwater
biodiversity and resources (Collen et al., 2014; Knouft &
Ficklin, 2017). Future climate change impacts
on freshwater biodiversity and
ecosystem functioning
The lowest greenhouse gas emissions scenario is the only
scenario not expected to threaten much of global freshwater
biodiversity in 2050 through direct effects of climate
change. Under all other scenarios, freshwater biodiversity
is expected to decrease proportionally to the degree of
warming and precipitation alteration. All water body types
on all continents are likely to be affected. Warmer waters
will alter community structure, food webs, body sizes, and
Figure 4 2 11
Global diversity maps (species richness and endemicity) for freshwater fi shes,
aquatic amphibians, aquatic mammals, crayfi sh and aquatic birds.
For comparison purpose, the diversity descriptor values of each taxon are rescaled between 0 and 100. Study based on the
global distributions of 13, 413 freshwater species among fi ve taxonomic groups (i.e. 462 crayfi sh, 3263 amphibians, 8870 fi sh,
699 birds and 119 mammals) and conducted on 819 river drainage basins covering nearly 80% of Earth’s surface. After Tisseuil
et al. (2013).
[0 : 21·03[
[21·03 : 27·18[
[27·18 : 37·7[
[37·7 :45·45[
[45·45 : 51·6[
[51·6 : 56·7[
[56·7 : 61·07[
[61·07 : 68·27[
[68·27 : 81·12[
[81·12 : 100]
[0 : 11·16[
[11·16 : 14·42[
[14·42 : 24·11[
[24·11 : 32·4[
[32·4 : 39·3[
[39·3 : 46·06[
[46·06 : 54·47[
[54·47 : 63·44[
[63·44 : 74·68[
[74·68 : 100]
[0 : 1·09[
[1·09 :5·36[
[5·36 : 10·02[
[10·02 : 15·85[
[15·85 : 23·06[
[23·06 : 30·75[
[30·75 : 40·65[
[40·65 : 55·31[
[55·31 : 90·05[
[90·05 : 100]
[0 : 13·42[
[13·42 : 24·06[
[24·06 : 38·97[
[38·97 :47·27[
[47·27 : 54·27[
[54·27 : 61·12[
[61·12 :66·8[
[66·8 : 74·34[
[74·34 : 83·2[
[83·2 : 100]
[0 : 3·06[
[3·06 :6[
[6 : 9·4[
[9·4 : 13·43[
[13·43 : 17·99[
[17·99 : 24·43[
[24·43 : 33·93[
[33·93 : 49·7[
[49·7 : 77·34[
[77·34 : 100]
[0 : 12·81[
[12·81 : 16·55[
[16·55 : 22·96[
[22·96 : 27·68[
[27·68 : 37·19[
[37·19 : 46·66[
[46·66 : 55·81[
[55·81 : 67·19[
[67·19 : 84·37[
[84·37 : 100]
[0 : 1·77[
[1·77 : 5·13[
[5·13 : 10·94[
[10·94 : 17·81[
[17·81 : 24·97[
[24·97 : 34·73[
[34·73 : 43·54[
[43·54 : 55·85[
[55·85 : 80·32[
[80·32 : 100]
[0 : 5·97[
[5·97 : 11·86[
[11·86 : 19·44[
[19·44 : 26·99[
[26·99 : 34·28[
[34·28 : 42·38[
[42·38 : 50·59[
[50·59 : 60·38[
[60·38 : 77·67[
[77·67 : 100]
[0 : 0·08[
[0·08 : 5·73[
[5·73 : 11·79[
[11·79 : 18·47[
[18·47 : 26·92[
[26·92 : 37·37[
[37·37 : 49·28[
[49·28 : 63·09[
[63·09 : 79·86[
[79·86 : 100]
[0 : 0·99[
[0·99 : 3·64[
[3·64 : 7·07[
[7·07 : 13·03[
[13·03 : 21·6[
[21·6 : 33·07[
[33·07 : 49·46[
[49·46 : 63·54[
[63·54 : 87·82[
[87·82 : 100]
species ranges — especially in regions where semi-arid and
Mediterranean climates currently occur as well as high-
mountain ecosystems. In addition to reduced biodiversity
and ecosystem functioning, warmer and less water will lead
to species extinctions because of habitat shrinkage.
Scenarios of climate change impacts on global freshwater
ecosystem biodiversity and functioning were reviewed
by Settele et al. (2014). Climate change alters freshwater
ecosystems and their biodiversity by changing (1)
temperatures, (2) water availability and (3) flow regimes
through changes in precipitation (Döll & Zhang, 2010; Knouft
& Ficklin, 2017) and/or temperature (Blöschl et al., 2017).
Increased water temperatures often lead to progressive
shifts in the structure and composition of assemblages
because of changes in species metabolic rates, body size,
migration timing, recruitment, range size and interactions
(Daufresne et al., 2009; Myers et al., 2017; Parmesan,
2006; Pecl et al., 2017; Rosenzweig et al., 2008; Scheffers
et al., 2016). There is already evidence of regional and
continental shifts in freshwater organism distributions
following their thermal niches (Comte et al., 2013), local
extirpations through range contractions at the warm edges
of species’ ranges (Wiens, 2016), and body size reductions
(Daufresne et al., 2009). Warmer water temperatures also
enhance microorganism metabolism and processing of
organic matter (unless dissolved oxygen is limiting), causing
eutrophication when nutrient levels are high (Carpenter et
al., 2011; Mantyka-Pringle et al., 2014) as well as increased
omnivory. Warming also induces phenological mismatches
between consumers and resources in highly seasonal
environments, potentially destabilizing food-web structure
(Woodward et al., 2010a).
The strongest temperature increases are projected for
eastern North America (0.7 to 1.2 °C under RCP2.6 and
RCP8.5, respectively, by 2050), Europe (0.8 to 1.2 °C),
Asia (0.6 to 1.2°C), southern Africa (>2.0°C under RCP8.5)
(van Vliet et al., 2016b) and Australia (CSIRO & Bureau of
Meteorology, 2015). Moderate water temperature increases
(<1.0°C) by 2050 are predicted for South America and
Central Africa (Van Vliet et al., 2013; van Vliet et al., 2016b).
Changes in water temperature are projected to lead to
local or regional population extinctions for cold-water
species because of range shrinking especially under the
RCP 4.5, 6.0 and 8.5 scenarios (Comte & Olden, 2017).
Most lowland-tropical freshwater species are expected to
tolerate warmer conditions where water is sufficient (Comte
& Olden, 2017).
Decreased water availability and altered flow regimes
reduce habitat size and heterogeneity. This increases
population extinction rates because the probability of
species extinctions increases with reduced habitat size
(Tedesco et al., 2013). Climate change can also alter flow
regime seasonality and variability (e.g., Blöschl et al., 2017;
Döll & Zhang, 2010) and increase flow intermittency (Pyne
& Poff, 2017). This would lead to decreased food chain
lengths through loss of large-bodied top predators (Sabo
et al., 2010), altered nutrient loading and water quality
(Woodward et al., 2010b), and/or pushing taxa into novel
trajectories from which they may not recover (Bogan &
Lytle, 2011). However, whatever the RCP scenario, climate
change impacts on the timing of seasonal streamflow are
found to be generally small globally (Eisner et al., 2017). Yet,
relative to water availability and according to the wet-wetter/
dry-dryer mechanism (Gudmundsson et al., 2017; Held &
Soden, 2006; Wang et al., 2017), more severe water stress
in current drylands is expected in the future. Although under
RCP2.6 the distributions of water availability may change
little by the end of the 21st century, RCP4.5, 6 and 8.5
scenarios are expected to induce substantial shrinking of
water drainage where semi-arid and Mediterranean climates
currently occur. Reduced water availability in those regions,
including shifts from permanence to intermittency, will
generate population extirpations of all types of freshwater
organisms (Jaeger et al., 2014), leading to global net
biodiversity losses because endemism is usually high in
those regions. For example, projected fish extinction rates
from drainage shrinking under the high emission SRES A2
scenario in river basins worldwide show that among the
10% most-altered basins, water availability loss is likely to
increase background extinction rates by 18.2 times in 2090
(Tedesco et al., 2013; Figure 4.2.12). Also, in glacier-fed
high-mountain ecosystems, significant changes to snow
and glacier melt regimes, including glacier disappearance,
have already been observed (Leadley et al., 2014) and are
expected to continue (Kraaijenbrink et al., 2017). This leads
to reduced water availability and declines in biodiversity
through local population extirpations and species extinctions
in regions of high endemicity in all water body types.
Besides biodiversity losses, losses of glacial ice in closed
drainages and flows in semi-arid regions (Vörösmarty et al.,
2010) will substantially decrease water for agriculture, power
and public water supply, thereby increasing economic
vulnerability in the affected regions (e.g., Moon, 2017).
Wetlands, including peatland and permafrost regions,
sequester carbon in their soils. But when confronted to
warming, drying and conversions to agriculture, wetlands
are expected to release CO2, CH4, and N2O. Global warming
alone is projected to contribute 1.6 x 108 kilotons of carbon
from melting permafrost to the atmosphere and CH4
emissions from freshwater wetlands are projected to nearly
double by 2100 (Moomaw et al., 2018). Such changes are
very likely to impact biodiversity negatively due to habitat
loss and reduced water quality, which increase the risk
of extinctions and extirpations of wetland endemic and
dependent species (Segan et al., 2016).
651 Future land-use change
impacts on freshwater biodiversity and
ecosystem functioning
Land use will likely increase the risk of eutrophication, leading
to local population extinctions, changes in community
structure and consequent modification of the food-web,
ecosystem temporal instability, and establishment and
spread of pathogens and toxic cyanobacteria blooms
globally. Land use will become especially problematic in the
emerging tropical economies because of increased human
population density and weak pollution controls. Increasing
pollution and eutrophication will degrade water quality, impair
biological resource availability, reduce nutrition in developing
countries, and reduce recreational opportunities and tourism
income. Globally increased toxic cyanobacteria blooms and
pathogens will increase health risks for people and livestock.
These risks will most affect closed water bodies and
estuaries, but rivers will also be threatened. The additional
impact of future increasing use of pesticides in agriculture is
hard to quantify due to a lack of scenario studies.
Land use, especially croplands, mining and urbanization, will
affect freshwater ecosystems and associated biodiversity
through two main pathways. First, further increased water
and groundwater withdrawals are expected to decrease
Figure 4 2 12
Global patterns of proportional increase or decrease in freshwater fi sh extinction
rates between current climatic conditions and future (2090) under the most
‘pessimistic’ IPCC SRES scenario (A2).
Negative values of projected change in extinction rate depict drainage basins where extinction rates may decrease, while
positive values depict drainage basins where extinction rates may increase. 91 949 river drainage basins covering ~99% of the
terrestrial surface. After Tedesco et al. (2013).
]5%; 10%]
]10%; 25%]
]25%; 50%]
]50%; 75%]
]75%; 100%]
]100%; 535%]
0%; 5%]
]5%; 10%]
]10%; 25%]
]25%; 50%]
]50%; 100%]
]100%; 1000%]
]1000%; 10000%]
]10000%; 110000%]
habitat (water) availability for freshwater organisms leading
to increased population extinction rates in rivers and lakes
or direct extinctions from wetland conversions (Gardner et
al., 2015; Tilman et al., 2001). The problem is exacerbated
in semi-arid regions where water withdrawals lead to some
rivers and lakes drying ro