ArticlePDF Available

Ecological monitoring with citizen science: The design and implementation of schemes for recording plants in Britain and Ireland

Article

Ecological monitoring with citizen science: The design and implementation of schemes for recording plants in Britain and Ireland

Abstract and Figures

Interest in citizen science has been increasing rapidly, although the reviews available to date have not clearly outlined the links between the long-established practice of recording plant species' distributions for local and national atlases, or other recording projects, and the gradual development of more structured monitoring schemes that also rely on volunteer effort. We provide a review of volunteer-based plant monitoring in Britain and Ireland, with a particular focus on the contributions of expert volunteers working with biological recording schemes and natural history societies; in particular, we highlight projects and practices that have improved the quality of data collected. Although the monitoring of plant distributions at larger scales has led to numerous insights into floristic change and its causes, these activities have also led to the recognition that knowledge of species' abundances at finer-scales often provides a more powerful means of detecting and interpreting change. In the UK, this has led to the development of a new, abundance-based 'National Plant Monitoring Scheme'. We outline this new structured scheme, and review some of the design considerations that have been made during its development. New monitoring projects require a clear justification, and the launch of a new scheme is also an opportune moment to review whether some basic assumptions about the collection of monitoring data can withstand scrutiny. A distinction is often made between monitoring that is focused on answering particular, focused questions, and that which is more generally seeking to detect changes; for example, in species' distributions or abundances. Therefore, we also review the justification for such general 'surveillance' approaches to the monitoring of biodiversity, and place this in the context of volunteer-based initiatives. We conclude that data collected by biological recorders working within atlas or monitoring scheme frameworks will continue to produce datasets that are highly valued by governments, scientists, and the volunteers themselves.
Content may be subject to copyright.
Ecological monitoring with citizen science: the design
and implementation of schemes for recording plants in
Britain and Ireland
OLIVER L. PESCOTT
1
*, KEVIN J. WALKER
2
, MICHAEL J. O. POCOCK
1
,
MARK JITLAL
1
, CHARLOTTE L. OUTHWAITE
1
, CHRISTINE M. CHEFFINGS
3
,
FELICITY HARRIS
4
and DAVID B. ROY
1
1
Biological Records Centre, Centre for Ecology and Hydrology, Maclean Building, Benson Lane,
Crowmarsh Gifford, Wallingford, Oxfordshire, OX10 8BB, UK
2
Botanical Society of Britain and Ireland, Suite 14, Bridge House, 12 Station Bridge, Harrogate,
HG1 1SS, UK
3
Joint Nature Conservation Committee, Monkstone House, City Road, Peterborough, PE1 1JY, UK
4
Plantlife, 14 Rollestone Street, Salisbury, Wiltshire, SP1 1DX, UK
Received 4 March 2015; revised 29 April 2015; accepted for publication 29 April 2015
Interest in citizen science has been increasing rapidly, although the reviews available to date have not clearly
outlined the links between the long-established practice of recording plant species’ distributions for local and
national atlases, or other recording projects, and the gradual development of more structured monitoring
schemes that also rely on volunteer effort. We provide a review of volunteer-based plant monitoring in Britain
and Ireland, with a particular focus on the contributions of expert volunteers working with biological recording
schemes and natural history societies; in particular, we highlight projects and practices that have improved the
quality of data collected. Although the monitoring of plant distributions at larger scales has led to numerous
insights into floristic change and its causes, these activities have also led to the recognition that knowledge of
species’ abundances at finer-scales often provides a more powerful means of detecting and interpreting change.
In the UK, this has led to the development of a new, abundance-based ‘National Plant Monitoring Scheme’. We
outline this new structured scheme, and review some of the design considerations that have been made during
its development. New monitoring projects require a clear justification, and the launch of a new scheme is also an
opportune moment to review whether some basic assumptions about the collection of monitoring data can
withstand scrutiny. A distinction is often made between monitoring that is focused on answering particular,
focused questions, and that which is more generally seeking to detect changes; for example, in species’
distributions or abundances. Therefore, we also review the justification for such general ‘surveillance’ approaches
to the monitoring of biodiversity, and place this in the context of volunteer-based initiatives. We conclude that
data collected by biological recorders working within atlas or monitoring scheme frameworks will continue to
produce datasets that are highly valued by governments, scientists, and the volunteers themselves. ©2015 The
Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115: 505521.
ADDITIONAL KEYWORDS: atlas – biological recording – botany – long-term monitoring – participatory
monitoring – surveillance monitoring – volunteer surveying.
INTRODUCTION
Long-term ecological monitoring schemes with volun-
teer participants are a large part of the landscape of
what is now often referred to as ‘citizen science’; that
is, the participation of people who are not profes-
sional scientists in activities that contribute to scien-
tific research (Dickinson, Zuckerberg & Bonter, 2010;
Dickinson & Bonney, 2012). Although such initia-
tives may be increasingly linked to technological
advances (August et al., 2015), technology is not an
essential component of citizen science as now under-
stood. In many cases, existing schemes, or historical
*Corresponding author. E-mail: olipes@ceh.ac.uk
505©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
Biological Journal of the Linnean Society, 2015, 115, 505–521. With 1 figure.
short-term initiatives, considerably pre-date the
popularization of this term in the 1990s (Bonney,
Shirk & Phillips, 2013). Indeed, despite some asser-
tions that, up to the mid-1990s, ‘volunteer-led data
collection efforts were relatively few in number’ and
that ‘most ... focused on monitoring the quality of
lakes, streams and rivers’ (Bonney et al., 2013), data
collection conducted largely by volunteers across
large areas has been a standard part of natural his-
tory in Britain and Ireland since at least the 1962
Atlas of the British Flora (Harding & Sheail, 1992;
Preston, Croft & Pearman, 2002b; Preston, 2013).
This, in turn, has its roots in the network of mainly
amateur botanists who assisted in the elucidation of
the biogeography of the British and Irish flora in the
19th and early 20th Centuries (Allen, 1986). Its his-
tory in Britain and Ireland can be traced even fur-
ther back, through to the huge contributions made
by the earlier naturalists who laid the foundations
for our understanding of the taxonomy and composi-
tion of our flora and fauna (Ray, 1660, 1670; Allen,
1976; Oswald & Preston, 2011). These types of his-
toric observations still have great value today for
helping us to understand long-term environmental
change (e.g. Preston, 2000, 2003; Walker, 2003a, b;
Walker, Preston & Boon, 2009), and it is now
increasingly recognized that biological recording has
played a pioneering role in involving nonprofession-
als in the collection of scientific data (Pocock et al.,
2015). Other volunteer-based recording projects with
early origins in Britain and Ireland include the UK
Butterfly Monitoring Scheme (UKBMS), begun in
1976 (Pollard & Yates, 1993), and the British Trust
for Ornithology (BTO)/Joint Nature Conservation
Committee/Royal Society for the Protection of Birds
Breeding Bird Survey (BBS), which has its origins in
the Common Bird Census begun in the early 1960s
(Harris et al., 2014).
The term ‘monitoring scheme’ can refer to a wide
range of activities, conducted at different scales, by
amateurs or professionals, focusing on many differ-
ent aspects of the environment (Spellerberg, 2005;
Lindenmayer & Likens, 2010b; Gitzen et al., 2012;
Pocock et al., 2015b). The production of species’ dis-
tribution atlases can itself be considered as a form of
monitoring, particularly where data can be assigned
to different time periods (Telfer, Preston & Rothery,
2002; Spellerberg, 2005; Tulloch et al., 2013),
although atlas initiatives can typically only report on
distributional change. Atlas (sometimes also called
‘cross-sectional’) projects, although potentially extre-
mely valuable with respect to collecting information
on the distributions and large-scale trends shown by
species (Spellerberg, 2005; Robertson, Cumming &
Erasmus, 2010; Porter & Leach, 2013; Hill &
Preston, 2014; Stroh et al., 2014), are not typically
run on short time-scales according to tightly con-
trolled data collection protocols (Brotons, Herrando
& Pla, 2007). By contrast, monitoring schemes that
allow the estimation of time trends in the abun-
dances of animals or plants that can be regularly
updated generally provide a more sensitive measure
of change for the reporting demands of national or
regional governments (Jones et al., 2011). From this
point on, we reserve the term ‘monitoring scheme’
for these ‘BBS’-type programmes, at the same time
not forgetting that other types of monitoring are pos-
sible. Existing British and Irish monitoring schemes,
then, include the UKBMS and UK BBS previously
mentioned, as well as the National Bat Monitoring
Programme (Barlow et al., 2015), Wetland Bird Sur-
vey (Austin et al., 2014), and Wider Countryside But-
terfly Survey (Brereton et al., 2011). Many of these
volunteer-based schemes have comparable pro-
grammes in other European countries (Schmeller
et al., 2009; see also http://www.floron.nl for current
Dutch initiatives). National monitoring schemes that
combine information from professional and volunteer
surveyors also exist, such as in Switzerland (Pear-
man & Weber, 2007; FOEN, 2014). Monitoring
schemes that are still in the developmental or scop-
ing stages in the UK include the National Plant
Monitoring Scheme (Walker et al., 2010a, 2015) and
a monitoring scheme for pollinators and pollination
services to crops, although we note that volunteer-
based, distribution-focused biological recording activ-
ities for these taxa (although not on pollination per
se) have existed in Britain and Ireland for many
years (Harding & Sheail, 1992; Spellerberg, 2005;
Pocock et al., 2015).
It is also important, however, to remember that
the distinction between an atlas project and a moni-
toring scheme may often be blurred. Information on
the presence of species at larger spatial scales can
normally be derived from finer-scale monitoring
scheme projects, whereas abundance data can also
now be linked to larger-scale detection/nondetection
data collected as part of an atlas project (Pagel et al.,
2014). In addition, a wide variety of techniques can
also now be used to extract estimates of trends from
biological records data that may be relatively
unstructured en masse, or semi-structured where
atlas-focused recording schemes encourage the use of
a particular protocol (Isaac et al., 2014). Indeed, the
use of recording protocols and support networks for
atlas projects (Dines, 1996; Walker et al., 2010b;
Balmer, Gillings & Caffrey, 2013) can help to break
down the distinction between monitoring schemes
and atlas-focused fieldwork, where the end result is
higher-quality, more structured data. Similar sugges-
tions, based largely on the premise of incorporating
elements of monitoring schemes into the design of
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
506 O. L. PESCOTT ET AL.
atlas projects, were recently made by Tulloch et al.
(2013). These suggestions included fine-scale data
collection; replication over time covering a range of
habitats; and communication of data needs to volun-
teers. Regional coordinators for the curation of data
and quality checking were also highlighted as desir-
able. Most of these ideas have been appreciated by
many British and Irish biological recording schemes
for some time (Preston et al., 2002b; Pocock et al.,
2015a,b). The suggestions of Tulloch et al. (2013)
were motivated by the finding of the greater scien-
tific impact of monitoring scheme data. Regardless of
whether one accepts the premise that the impact of a
national, often largely volunteer-based piece of work,
such as an atlas, can be adequately quantified by sci-
entific citation counting, the introduction of more
structure to volunteer-based survey work is clearly
very important for data quality and, consequently,
for the robustness of any ecological conclusions
drawn (Walker et al., 2010b).
Here, we (1) review volunteer-based plant monitor-
ing in the broad sense in Britain and Ireland, partic-
ularly focusing on its contributions to our knowledge
of ecological change during the late 20th Century; (2)
outline approaches towards an abundance-based
plant monitoring scheme; and (3) review the justifi-
cation for such general ‘surveillance’ approaches to
the monitoring of biodiversity.
VOLUNTEER-BASED PLANT MONITORING
PROJECTS IN BRITAIN AND IRELAND
Where the recording of plant distributions and abun-
dances are concerned, the Botanical Society of Brit-
ain and Ireland (BSBI), in collaboration with the
national Biological Records Centre of the UK, has a
long history of learning from the challenges posed by
volunteer-based projects covering large areas (Rich &
Woodruff, 1990; Perring, 1992; Croft & Preston,
1999; Spellerberg, 2005; Braithwaite, Ellis & Preston,
2006; Preston, 2013; Pocock et al., 2015). For example,
the Atlas of the British Flora (Perring & Walters,
1962) started life as a descriptive, phytogeographical
project (Perring, 1992), although it was subsequently
used to identify widespread declines in various spe-
cies; these insights have come to be widely considered
as one of the most significant applications of the data
collected (Rich & Woodruff, 1990; Porter & Leach,
2013; Preston, 2013), and led to the first Red Data
book of British plants (Perring & Farrell, 1977). The
uses to which the Atlas of the British Flora data were
put clearly highlight the potential value of distribu-
tion mapping for conservation. This particular case
can be seen partly as the product of a particular his-
torical and ecological situation: the periods across
which the 1962 Atlas data were compiled were a time
of great change in the British countryside (Perring,
1970), and particular habitats were being lost at an
unprecedented rate (Fuller, 1987; Hooftman &
Bullock, 2012). This meant that the relatively coarse
scale of the Atlas (10 910 km) was able to reveal
large changes in the distributions of many plants
(Perring, 1970); these types of data are still regularly
used to produce conservation assessments at a
national scale (Stroh et al., 2014; Van Maes et al.,
2015).
The far-reaching legacy of the 1962 Atlas (Preston
et al., 2002b; Preston, 2013) led to discussion during
the 1980s concerning the need for an updated atlas,
aiming to provide further information on plant distri-
butions for both conservation and statutory demands
(Rich & Woodruff, 1990). A report of the time notes
the general feeling in the plant recording community
that a new atlas survey during the late 1980s might
coincide with a slowing of agricultural intensifica-
tion, and that it might therefore be better to wait for
a period of ‘relative stability’ to maximize the longev-
ity and relevance of the maps created (Rich &
Woodruff, 1990). This highlights one tension for
co-ordinators of atlas projects: the snapshot captured
by surveyors may miss a period of widespread
change, leading to a resource with the potential to
mislead its users. This may be one reason why some
studies report that monitoring schemes appear to be
more influential than atlases in the scientific litera-
ture (Tulloch et al., 2013). In the case of the BSBI,
the decision to resurvey for a new national plant
atlas was not taken lightly: Rich & Woodruff (1990)
note that there was much discussion in the 1980s
about the appetite for a new volunteer survey. These
doubts, coupled with worries that the distribution of
many species might not have changed much since
1962, led to the planning of a different type of
survey, namely the ‘BSBI Monitoring Scheme’ (Rich
& Woodruff, 1990, 1996; Preston et al., 2002b). This
was designed not only to collect information at a
finer scale (2 92 km) than the 1962 Atlas, using a
regular, systematic grid sample, but also to allow
comparisons to be made with the Atlas. The scheme
was therefore intended both as a first step towards a
new national plant atlas, and as the basis of a new
type of scheme that would be run periodically: one
that integrated detection/nondetection type atlas
data with a rigorous sampling framework (Rich &
Woodruff, 1990). This demonstrates that the BSBI
Monitoring Scheme coordinators were identifying
multiple uses for their future dataset, as well as
considering the characteristics required for a scheme
to provide answers to questions about changes in
plant distributions, and providing important insights
into the capacity for future volunteer-based atlas
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 507
surveying (Rich & Woodruff, 1996). Moreover,
between one-quarter and one-eighth of species inves-
tigated were found to have undergone significant
change, depending on the region, despite the avowed
lack of sensitivity of the project by the coordinators
(Rich & Woodruff, 1996). Five general trends were
also highlighted, including widespread losses of
grassland taxa and large declines in arable weeds,
which were later confirmed during analyses of the
subsequent national New Atlas (Preston, Pearman &
Dines, 2002c; Preston et al., 2002a, 2003). This dem-
onstrates that the 19871988 BSBI Monitoring
Scheme was successful in detecting at least the lar-
ger changes that had occurred in the landscape since
the completion of the 1962 Atlas (Preston et al.,
2002b).
To our knowledge, the BSBI Monitoring Scheme is
the only example of a volunteer-based, systematic
monitoring scheme for plants that has sampled such
a large area (Britain and Ireland, an area of
315 130 km
2
; Rich & Woodruff, 1996), and it must
certainly be the earliest such scheme successfully
completed. However, the BSBI Monitoring Scheme
appears to have been largely overlooked by the wider
ecological monitoring and citizen science literatures;
for example, of the 110 papers indexed by Google
Scholar as citing Rich & Woodruff (1996) as of
December 2014, none are general reviews of volun-
teer-based ecological monitoring or atlas projects,
despite reviews of this area being published with
increasing frequency (Cohn, 2008; Bonney et al.,
2009; Silvertown, 2009; Devictor, Whittaker &
Beltrame, 2010; Dickinson et al., 2010, 2012; Robert-
son et al., 2010; Hochachka, Fink & Zuckerberg,
2012; Miller-Rushing, Primack & Bonney, 2012; Tul-
loch et al., 2013). This neglect may be partially
because the botanical findings were only reported in
two scientific papers (Rich & Woodruff, 1996; Rich,
Beesley & Goodwillie, 2001), with other commentar-
ies on the findings being published largely in the
grey literature (Rich & Woodruff, 1990; Palmer &
Bratton, 1995; Rich, 1996; but see also Pearman
et al., 1998). The project coordinators also empha-
sized the possible biases in the data collected (Rich
& Woodruff, 1992; Rich, 1998), demonstrating the
desire to ensure that the findings were as robust as
possible. This lead to further work on systematic
sampling for volunteer-based plant surveying (Rich
& Smith, 1996), resulting in an attempt at record-
ing a local flora with uniform effort (Rich et al.,
1996). In this sense, the Monitoring Scheme coordi-
nators were amongst the earlier innovators in an
area now receiving much more attention; namely
that of dealing with bias in distribution data (Dick-
inson et al., 2010; Robertson et al., 2010; Isaac
et al., 2014). The BSBI Monitoring Scheme was not
a one-off, and the initial intention to carry out
repeat surveys was fulfilled in 20032004 when the
BSBI follow-up project ‘Local Change’ was carried
out in Britain using the same methodology and grid
squares as those of the earlier scheme (Braithwaite
et al., 2006). This re-survey allowed the first analysis
of national change in plant distributions at the
292 km scale. Seven-hundred and fifty volunteer
recorders took part.
The Local Change survey was successful in creat-
ing distribution-based trend statistics for a large
number (N=725) of species. Species were also clus-
tered into groupings, based on an a priori habitat
classification and geographical distribution, for the
purposes of creating more robust analyses of change
(Braithwaite et al., 2006). Considerable effort was
also invested in adjusting for differences in recorder
effort between the Monitoring Scheme and Local
Change surveys (Braithwaite et al., 2006, pp. 353
368). Local Change was highly successful in detect-
ing signals of ecological change: it not only supported
many of the earlier conclusions of the 2002 New
Atlas (e.g. the loss of species of infertile habitats
such as dwarf shrub heath; Preston et al., 2002c;
Braithwaite et al., 2006), but also revealed new
trends, such as the first evidence for a ‘modest’ recov-
ery in the weed flora of arable fields (Braithwaite
et al., 2006: 323). However, as for the BSBI Monitor-
ing Scheme, the Local Change project has not been
widely cited as an example of a successful and innova-
tive volunteer-based monitoring project: of the 70 cita-
tions of Braithwaite et al. (2006) on Google Scholar as
of December 2014, none are from recent reviews cover-
ing the general areas of volunteer-based monitoring or
citizen science. The authors of the BSBI Local Change
report also reviewed the survey’s contribution to UK
plant monitoring and found that, in terms of species
distribution types (e.g. rare, local, widespread, etc.)
and habitat coverage, it filled a niche that was
complementary to other existing professional and
volunteer-based projects (Braithwaite et al.,2006:
327). This complementarity is clearly an important
consideration for projects aiming at longevity and
efficiency, and the recognition of the unique contribu-
tion of the Local Change project has recently been
re-affirmed by the BSBI, with a decision to re-run the
survey in the early 2020s.
In addition to the innovations characterized by
the 2 92 km, repeated, systematic surveys dis-
cussed above, traditional large-scale atlas data (i.e.
10 910 km) in broad date classes (Perring & Walters,
1962; Preston et al., 2002b) are being increasingly
used to draw inferences about trends. Methods of
analyzing this type of data have recently been
reviewed and evaluated by Isaac et al. (2014). Several
of these techniques were developed in the context of
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
508 O. L. PESCOTT ET AL.
botanical data (Telfer et al., 2002; Hill, 2012) and are
now being used increasingly on plant (Preston et al.,
2002c; Stroh et al., 2014; Hill & Preston, 2014; Hill &
Preston, 2015; Pescott et al., 2015) and animal data
sets (Balmer et al., 2013; Cham et al., 2014; Fox
et al., 2014). Other methods of analyzing floristic
change, such as looking at changes in extrapolated
occupancy probabilities using kriging (Groom, 2013)
or other spatial methods (Le Duc, Hill & Sparks,
1992; Firbank et al., 1998), have also been enhanced
by the existence of the regular 2 92 km grid sur-
veyed by the BSBI Monitoring Scheme and Local
Change projects. It is clear from the applications seen
in the ecological literature that, in Britain and Ire-
land, largely volunteer-collected atlas data have con-
tributed an enormous amount to the national
monitoring capacity (cf. Schmeller et al., 2009) and to
ecological research (e.g. McClean et al., 2011; Powney
et al., 2014). Although at least partly recognized
(Spellerberg, 2005), this fact has not received the
attention that it deserves in the citizen science litera-
ture. Recent historical overviews of the subject have
not clearly drawn out the continuous thread linking
early biological recorders and taxonomists, such as
John Ray, to current day recording activity (cf.
Miller-Rushing et al., 2012), have focused largely on
birds (Dickinson et al., 2010) or on North American
contributions (Cohn, 2008; Miller-Rushing et al.,
2012). As noted by Pocock et al. (2015a) in the gen-
eral context of biological recording, this has tended to
mean that ‘the distinctive attributes and successes’ of
British and Irish volunteer-based plant monitoring
have been ‘largely untold outside of the UK’; no doubt
similar statements could be made about long-running
schemes on the continent (Schmeller et al., 2009).
Nonetheless, all of the schemes discussed in the pres-
ent review fit within standard definitions of ecological
monitoring [e.g. Suter II (1993: 505) defines monitor-
ing as the ‘measurement of environmental character-
istics over an extended period of time to determine
status or trends in some aspect of environmental
quality’] and they have also all been dependent on
largely voluntary biological recording activity. The
unique contributions of plant recording to the devel-
opment of volunteer-based monitoring deserve to be
more widely recognized (Preston, 2013).
It may be that atlas projects and related activities
are seen as lacking the rigour of abundance-based
monitoring schemes with more statistical designs
(Tulloch et al., 2013); however, it is inevitable that
there will always be a desire, either from funders or
from within a biological recording scheme or society
itself, for interpretation and commentary on results
from atlas or other cross-sectional projects, and, as
has been emphasized above, many recording
schemes already incorporate numerous ‘monitoring
scheme’ elements of rigour in their make-up, result-
ing in data that tend towards being semi-structured
rather than opportunistic. Regardless of commentar-
ies in the academic literature, it is probably inevita-
ble that atlas data will continue to be used to
answer questions about environmental change: the
emphasis on collecting and publishing atlas data
within recording schemes attests to its popularity
with amateur naturalists (Pocock et al., 2015a).
Indeed, in Britain and Ireland, the number of
botanical county or regional atlases featuring distri-
bution maps shows no signs of slowing (Fig. 1), and
it is clear that the collected data have numerous
potential uses (Hill, 2003). Analyses of change using
atlas data also have the advantage of longer avail-
able time-series in many instances, albeit poten-
tially with increased uncertainty around historical
estimates of frequency.
DESIGNING NEW VOLUNTEER-BASED
SURVEILLANCE MONITORING FOR PLANTS
In addition to the many uses of atlas-type plant dis-
tribution monitoring in Britain and Ireland, recently,
and partly as a result of the Local Change survey
(Braithwaite et al., 2006), there has been growing
recognition of the fact that observations of plant
trends across finer scales of space and time would
also be desirable, particularly where there is a wish
0
10
20
30
40
50
60
70
1970 1980 1990 2000 2010
Year
Total
Total
5 km
2 km
1 km
Figure 1. The cumulative number of plant distribution
atlases published for British and Irish counties and
regions at different grid sizes. The first local Flora featur-
ing grid-based distribution maps was published in 1967
by J. G. Dony (1967).
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 509
to monitor population changes, or changes in species
groups, at the scale of the habitat, or a desire to pro-
duce annual trends. Observations at this level are
more likely to provide early warnings of negative
trends or, indeed, of conservation successes. The
need for a standardized approach to plant monitoring
that would provide timely and robust estimates of
status and trends was therefore identified as a high
priority within the UK government’s Terrestrial Bio-
diversity Surveillance Strategy (JNCC, 2008). A scop-
ing study to investigate how this might be achieved
soon followed (Walker et al., 2010a). Walker et al.
(2010a) reviewed a number of possible approaches to
the deployment of volunteer-based plant monitoring
at the habitat scale, drawing on existing schemes
such as the BSBI Monitoring Scheme/Local Change,
and approaches used in other species groups. Statis-
tical design is obviously central to the robustness of
any monitoring scheme, and, unless a project is
solely aimed at public engagement, a defensible sta-
tistical foundation will be essential (Legg & Nagy,
2006; Jones, 2013). Below, a review is provided of
the central requirements relating to the development
of a nascent annual monitoring scheme for plants in
the UK, namely, the new National Plant Monitoring
Scheme (Walker et al., 2015).
SCHEME DESIGN
The power of a scheme to detect change
The statistical power to detect a specified level of
change is often highlighted as a key preliminary for
the design of any new monitoring scheme (Legg &
Nagy, 2006), and much work has been carried out in
recent years aiming to develop new models for esti-
mating power that are appropriate to different types
of monitoring data (e.g. Roy, Rothery & Brereton,
2007; Irvine & Rodhouse, 2010). Given the numerous
sources of variance that may influence the ability of a
large-scale, long-term, volunteer-based survey to
detect change, a key challenge is to derive sensible
parameters for power analyses. Variables that are
typically explored include the number of years of
monitoring; the number of sites; the number of visits
within years; the effect size(s) to be detected; site
revisit schedules; and variation affecting the trends
exhibited across sites and species. For larger volun-
teer-based schemes, the choice of which variables to
include in a prospective power analysis (and at what
values) may be subject to considerable discussion. For
example, for a scheme that intends to monitor aggre-
gated trends across multiple species, how should
trends simulated for power analyses be distributed
across species? Power to detect an aggregated trend
may be high if all species are assumed to follow a par-
ticular trajectory, whereas power may be low where
species have randomly assigned trends. Similarly, for
a single species across sites, the variance in a simu-
lated trend will be a key determinant of power. Exist-
ing data are indispensable for estimating sensible
variance parameters to investigate but, even where
suitable data are available, key factors may still vary
between the data utilized and the future reality of
the scheme (Urquhart, 2012). For example, power
analyses for the new UK National Plant Monitoring
Scheme utilized long-term data from the Countryside
Survey (http://www.countrysidesurvey.org.uk) to esti-
mate variance parameters (O.L. Pescott, M. Jitlal
and S.N. Freeman, unpubl. data), although Country-
side Survey data are collected by professional ecolo-
gists, and may underestimate the amount of
variation exhibited by a volunteer-based scheme. Of
course, the ‘true’ variance in trends across sites may
also change over time.
Site revisit schemes can also be important for
power, particularly over relatively short time-scales.
For example, sampling schemes that revisit sites in
the medium-term generally have higher power that
those that do not revisit sites, or only revisit them in
the very short-term (Urquhart, 2012). Current
recommendations highlight the need to revisit sites
after a sufficiently long period for significant change
to have occurred, but also to sample a large number
of different sites. Sampling larger numbers of sites
improves estimates of site variance, and leads to
more precise estimates of trend sizes (Urquhart,
2012). One way of fulfilling these dual aims would be
to have a rotating panel-like scheme for site revisits,
supplemented with a set of sites that are visited each
year (Urquhart, 2012). However, even with this
intention, such a scheme is unlikely to be fully met
in a volunteer context, due to surveyor drop-out, or
differing preferences for visiting the same site every
year versus visiting a number of sites on a rotating
basis. Urquhart & Kincaid (1999), however, showed
that random revisit schedules, with a mix of visits to
new sites and revisits to existing sites, comes close to
achieving the power of more systematic panel
designs.
In general, recent approaches to power analyses
using simulation-based estimates have increased the
ease and flexibility with which the ecologist can esti-
mate power (Gelman & Hill, 2007; Bolker, 2008;
Irvine & Rodhouse, 2010). Analyses can now rou-
tinely use these approaches to include random effects
to account for variation in responses (Miller & Mitch-
ell, 2014; Johnson et al., 2015). However, the incor-
poration of variance parameters in simulated data
can sometimes yield misleading results. For example,
the use of the Poisson distribution to simulate count
data, combined with estimates of variances in trends
from existing data, can lead to the counter-intuitive
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
510 O. L. PESCOTT ET AL.
result that power increases with increasing variance;
for example, for scenarios where initial counts and
effect sizes (i.e. trends) are both small. Under these
circumstances, simulated declines can lead to a small
proportion of sites increasing to unrealistically large
values. These outlying sites can then lead to the
detection of significant increasing trends. The use of
sophisticated models for estimating power clearly
requires vigilance, as well as a constant need to
check that simulated data are biologically realistic
and that appropriate models are being used, particu-
larly where investigated variances are derived from
small existing datasets. Finally, a proliferation of
modelled scenarios may actually make it harder to
decide whether a study is likely to be under-powered
or not, although it is also possible that bringing
uncertainty around power analyses into the open
may help to reduce unrealistic expectations.
Dealing with potential biases
Bias in monitoring schemes can arise in a number of
ways, and may well be greatest for volunteer-based
schemes. For example, organizers may find it chal-
lenging to persuade surveyors to visit random loca-
tions, particularly in an intensively managed
landscape where the chances of encountering many
target species may be low. However, specifically tar-
geting nature reserves or other high-quality habitats
leads to a biased picture of the countryside that is
likely to be inappropriate for a scheme producing
data to support mandated monitoring (Jones et al.,
2011). This was a potential criticism of the original
UKBMS, in which participants selected their own
survey locations, typically in areas of relatively high
species richness for butterflies (Roy et al., 2015). But-
terfly Conservation, the Centre for Ecology & Hydrol-
ogy, and BTO recently addressed this challenge by
launching a new scheme, the Wider Countryside But-
terfly Survey (WCBS), as a component of the UKBMS
(Brereton et al., 2011). The WCBS requires surveyors
to visit a random (1 91 km) grid-cell at least twice a
year. It predominately samples populations of com-
mon species, highlighting a trade-off between a repre-
sentative design (i.e. a random sample of the
countryside) versus the need to monitor rarer species
that tend to be restricted to higher-quality semi-natu-
ral habitat (i.e. those targeted by the traditional tran-
sect location of the UKBMS). Initial results from the
WCBS suggest that the trends are similar with either
approach (Roy et al., 2015). Given existing knowledge
of the importance of high-quality habitat for many
specialist butterflies (Warren et al., 2001), and the
fact that they are generally highly mobile organisms
(Thomas, 2005), this result is not unexpected; how-
ever, without the WCBS, it would be impossible to
know with any degree of confidence whether nature
reserves were different from the wider countryside
for commoner butterfly species, or to detect increases
in the ranges or abundances of our rarer species
occurring outside of the potentially biased set of
traditional UKBMS sites.
One option for reducing bias from a set of sites, at
the same time as retaining a focus on higher-quality
sites that may lead to increased volunteer retention,
is to create a weighted random selection, where the
weights are derived from some measure of habitat
quality. Existing land cover maps can be used to cre-
ate weightings that integrate national frequencies of
‘high value’ habitat types with the areas of these
types within a grid cell; the resulting weights can
then be used in analyses, thus adjusting for the
inherent bias (Yoccoz, Nichols & Boulinier, 2001).
Although appealing, and often useful, this approach
does have its limitations. Land cover maps are not
perfect (Gerard et al., 2012), and this approach will
only be as successful as the quality of the land cover
map used and the appropriateness of the land cover
categories mapped for the monitoring question being
addressed. Even if the bias in larger-site selection is
accounted for in this way, bias can also arise from
what occurs at the finer-scale of monitoring activity.
For example, the placement of a transect or plot
within a site may also lead to inaccurate assess-
ments of a species’ status: surveys that are along
rights of way are likely to overestimate the abun-
dances of species of edges or other linear features,
particularly for sessile organisms.
Surveys designed to detect both increases and
decreases in some resource are also likely to be
affected by surveyors self-selecting survey points: for
example, the selection of the most typical example of a
vegetation type is likely to mean that it contains most
of the species that the surveyor is being asked to moni-
tor; consequently, these patches are more likely to
‘decline’ in status on average (an example of the well-
known phenomenon of ‘regression to the mean’). The
challenge with all these types of bias is that the extent
to which they may or may not affect results is essen-
tially unknowable without conducting a second, unbi-
ased, survey. The importance of getting it right first
time can be appreciated. All of this argues for the
unbiased allocation of survey points within a site, par-
ticularly for sessile organisms, although this is chal-
lenging for surveys by volunteers, particularly where
much of a landscape is in private ownership. Obtain-
ing unbiased trends, then, should be a key consider-
ation of any monitoring scheme (Bart & Beyer, 2012),
although, in cases where volunteers are contributing
to a project, it is likely that bias of some sort or
another cannot ever be entirely removed, except per-
haps where projects focus on small areas. Techniques
such as double-counting (i.e. two volunteers recording
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 511
the same square independently), as is being imple-
mented for the Dutch Butterfly Monitoring Scheme
(C. van Swaay, pers. comm.), combined with occu-
pancy modelling may help to alleviate certain forms of
potential bias (Chen et al., 2009).
Selecting the species to monitor
It is well established that monitoring schemes often
focus on subsets of species as a result of resource con-
straints (Manley et al., 2004) or to enable wider par-
ticipation in a scheme. Even where experienced
volunteers are encouraged to record all species
encountered in a group, low encounter rates for rare
species may still restrict the number of species for
which trends can be produced (Braithwaite et al.,
2006). For the new UK National Plant Monitoring
Scheme, a particular focus has been on encouraging
broad participation. This has led to a scheme design
that attempts to appeal to different skill levels: an
entry level, composed of short lists of around 15
plants per habitat; an intermediate level, with lists of
up to 30 species per habitat, including harder to iden-
tity groups such as grasses, sedges, and ferns; and an
expert level, where all species within a plot are
recorded. With 28 habitats included in the scheme
(Walker et al., 2015), this means that the entry and
intermediate levels contain approximately 200 and
400 species, respectively. One UK public engagement
scheme, ‘Wildflowers Count’, led by the conservation
charity Plantlife, found that slightly shorter overall
lists (99 species, in this case) were popular with vol-
unteers, although many participants asked for more
species aiming to increase the number of encounters
in the field. A list of 400 species for an intermediate
level of volunteer expertise may appear excessive,
although it should be remembered that a volunteer is
unlikely to encounter more than a few target habitats
within a single site (here, a 1 91 km square of the
British or Irish national grid).
Another challenge for a volunteer-based monitoring
scheme is the selection of a set of species that balances
the demands on volunteer identification effort noted
above with the need to select species indicating a wide
range of ecologically informative but potentially
unpredictable changes in a plant community. Ran-
dom, or weighted random, selections of species can be
made with respect to certain traits, coupled with fil-
ters that use information on identifiability or national
frequency to eliminate very rare or critical species
(Walker et al., 2010b). Information on plant species
associations with national vegetation classifications
can also be used to restrict habitat species pools if
the habitats to be monitored can be equated with
these. Ultimately, where volunteer uptake is essen-
tial for the success of a scheme, we have found that
the combination of the preceding steps with a group-
led expert review is the most pragmatic route
through species selection. Species’ trait representa-
tions can be investigated post-selection to evaluate
the likelihood that the final species set selected is
biased in some way towards detecting, or failing to
detect, certain types of environmental change.
Reporting results
An important aspect of monitoring scheme success is
the timely reporting of results (Devictor et al., 2010).
Information obtained from monitoring schemes is
increasingly reported in the form of biodiversity indi-
cators, which are measures that are intended to
illustrate trends in biodiversity over time (Va
ck
a
r
et al., 2012). These are often shown in an aggregated
form to illustrate overall change or can be disaggre-
gated into various sub-indices, depending on the
data, to look at changes within certain species
groups or across specific habitat types. Robust (i.e.
unbiased and sensitive) biodiversity indicators are
becoming increasingly demanded by policy makers so
that assessments of human impact can be made and
progress towards environmental targets can be moni-
tored (Butchart, Walpole & Collen, 2010). The recent
Convention on Biological Diversity in Aichi set five
strategic goals and twenty targets with the aim of
stopping biodiversity loss by 2020 (Convention on
Biological Diversity, 2010). In particular, Target 12,
which states that ‘[b]y 2020 the extinction of known
threatened species has been prevented and their con-
servation status, particularly of those most in
decline, has been improved and sustained’, requires
the use of robust indicators to determine which spe-
cies are declining, to monitor their status, and to
assess whether this target has been reached. In the
UK, a suite of indicators has been produced to moni-
tor progress towards the Aichi targets and are pro-
duced annually in the ‘Biodiversity Indicators in
Your Pocket’ report (Defra, 2014). A number of attri-
butes are regarded as necessary for an ideal biodi-
versity indicator, including the need to: (1) be
quantitative in that the measure is accurate with
estimates of error; (2) be indicative of a wider range
of species than those included within it; (3) be timely
in the reporting of changes in trends; and (4) be rele-
vant to policy (Gregory et al., 2005; Jones et al.,
2011). Clearly, all design aspects of a scheme must
be carefully considered for these attributes to be
confidently defended.
IS ALL ‘SURVEILLANCE’ MONITORING
INEFFICIENT?
Although we have discussed scheme design in the
context of producing robust trends, the initial deci-
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
512 O. L. PESCOTT ET AL.
sion to commission a monitoring scheme can itself be
considered as a design choice (McDonald-Madden
et al., 2010). The typical proposition of many atlases
and monitoring schemes, namely that ‘relevant,
up-to-date information is essential for determining
priorities and policies for conservation’ (Rich &
Woodruff, 1996), would appear to be uncontroversial;
however, recently, a subset of the literature on biodi-
versity monitoring has focused on investigating the
effectiveness of different types of monitoring, particu-
larly where cost is concerned (Nichols & Williams,
2006; Wintle, Runge & Bekessy, 2010; Possingham,
Fuller & Joseph, 2012b; Possingham et al., 2012a).
Some of this literature essentially challenges those
who promote monitoring that is not specifically
focused on answering pre-specified questions to jus-
tify such an approach (Yoccoz et al., 2001; Nichols &
Williams, 2006; Lindenmayer & Likens, 2010a); this
type of monitoring has typically been referred to as
‘surveillance’ or ‘omnibus’ monitoring.
Those who are largely critical of surveillance moni-
toring suggest that it is too unfocused to reliably pro-
vide conclusions of scientific value to feedback into
the achievement of desired goals, particularly con-
cerning the management of habitats and species
(Nichols & Williams, 2006). Question-based monitor-
ing (QBM; sometimes also called process-based or
targeted monitoring) has typically been put forward
as a more desirable alternative (Table 1) (Nichols &
Williams, 2006; Lindenmayer & Likens, 2010a). Crit-
ics of surveillance monitoring suggest that the stated
aims of such projects could often be achieved for less
effort using targeted surveys conducted by small pro-
fessional teams (Nichols & Williams, 2006). Other
criticisms have focused on the challenge of separat-
ing the effects of multiple drivers (Yoccoz et al.,
2001), pointing out that surveillance monitoring
often relies on induction to reach conclusions
(Table 1). Possingham et al. (2012a, b) have recog-
nized a larger range of the potential benefits arising
from different types of monitoring, noting that moni-
toring that is not specifically focused on experimental
management, or on informing periodic management
decision making, may still yield benefits in the areas
of informing policy makers, educating or engaging
the public, or detecting unanticipated ecological
change. We see these last three benefits as those that
most volunteer-based monitoring schemes in Britain
and Europe have taken as their main motivations.
Accordingly, two of the key propositions concerning
the choice between QBM and surveillance monitoring
are that: (1) trend detection unlinked to action may
be inefficient; and (2) surveillance monitoring could
potentially be justified by the detection of ‘ecological
surprises’ (Nichols & Williams, 2006; Lindenmayer
et al., 2010; Wintle et al., 2010). Below, we briefly
evaluate the relevance of these considerations to
volunteer-based monitoring as it has proceeded in
Britain and Ireland to date.
TREND DETECTION UNLINKED TO ACTION MAY BE
INEFFICIENT
A monitoring scheme may identify a trend, and so a
logical question to ask is: ‘what next?’ Unless moni-
toring is specifically designed to detect such a trend,
then a plan to address this trend is likely to take
time to formulate. Is this really a weakness of the
monitoring? The link between management action
and trend detection is obviously of most relevance to
monitoring focusing on a limited set of processes or
quantities that can conceivably be managed by a set
of actions applied across the area monitored. Unfo-
cused surveillance monitoring projects that cover
large areas (e.g. those monitoring national trends),
make observations at large spatial grain (e.g. atlas
projects), or with broad taxonomic coverage, are unli-
kely to be amenable to the planning of clear actions
that can be launched when certain trends are discov-
ered, not least because of the practical problem of
launching a management action across a habitat-
type or landscape largely in private ownership. For
example, evidence from the UK Breeding Bird Sur-
vey has revealed large declines in woodland birds,
which has been linked to a decline in certain types of
traditional woodland management (Fuller et al.,
2007). Although woodland management is not consid-
ered to be the only driver of declines in woodland
birds, this knowledge could be used to promote
increased woodland management as a desirable
activity for avian biodiversity conservation in British
woodlands; however, the launching of specific conser-
vation activities as a result of this finding is neither
direct nor straightforward.
The desire for trend detection resulting from
monitoring to be always linked to management
responses parallels the long-standing debate between
the relative worth of observational-inductive and
hypothetico-deductive studies in ecology. Nichols &
Williams (2006) adduce the ‘strong inference’ para-
digm of Platt (1964) as support for their view of mon-
itoring, in that all monitoring should be conducted in
a framework that leads to the rejection of hypothe-
ses, allowing a gradual approach to the underlying
reality. Others have pointed out that Platt’s strong
inference is only one possible approach to science; for
example, Hilborn & Mangel (1997: 23) emphasize
that ‘many ecological studies are motivated by prob-
lems where such clear experimentation and “hard
data” are often not possible ... or lead to other diffi-
culties’ (for a detailed treatment of these issues, see
Pickett, Kolasa & Jones, 2007). Observational data
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 513
Table 1. The general features of question-based and surveillance monitoring. Adapted from Wintle et al. (2010) and
Field et al. (2007). As noted by Wintle et al. (2010), these are the two ends of a continuum, and mixed, intermediate
strategies are also possible
Project area
Question-based monitoring
(QBM) Surveillance monitoring (SM) Additional comments
Focus Targeted to improve
management by learning
about pre-specified
processes
May have a specific scientific
purpose (e.g. monitoring
optimized to discern
between competing
hypotheses)
Generally not based on any one
particular management
problem or specific scientific
question
Management-targeted
monitoring may be challenging
for SM with multiple species
and a broad scope
Design Sampling optimal to address
specific hypotheses or to
estimate a state
High statistical power to
differentiate between
specific hypotheses or to
achieve precise estimates of
state variables
Sampling not optimized to a
particular purpose, although
trend detection is often given
as a rationale
Potentially low power to
differentiate between
hypotheses or to estimate
trends
Where SM is volunteer-based,
power to detect a trend may
be high, although we recognize
that adequate volunteer
engagement then becomes a
key challenge
Generally well stratified,
replicated and exhibiting
low bias and/or variance for
the specified purpose
Often poorly stratified, not
replicated, and having a
biased sampling frame or one
with high variance
Replication is not necessarily a
problem for SM where projects
have a large extent and are
volunteer-based
Bias in sampling frames can be
addressed at the project
design stage in both QBM and
SM initiatives
Variable sample sizes,
depending on the question
and the available resources
More often characterized by
large sample sizes
Logical
approach
Deductive reasoning Inductive reasoning
A priori hypotheses
articulated
A priori hypotheses either not
specified or vague. The
hypothesis is often that
certain species or habitats are
likely to change in interesting
ways in the future
The fundamental objective of SM
is often the conservation of all
species; monitoring can
subsequently be focused on
learning about drivers of
decline (Wintle et al., 2010)
Scope Typically has a narrower
geographical and/or
temporal scope
Fewer species, few state
variables, fewer covariates
Often broad in geographical
scope or long-running with no
pre-defined end-point
(Potentially) many species, many
state variables, many
covariates
Data collection Generally collected by
professional scientists.
Expensive per data point
Generally collected by a mix of
professional scientists and
volunteer observers
Cheap per data point
Potential for
volunteer
contribution
Likely to be restricted to the
most engaged volunteers
(even amateur experts in
the identification of species
may be deterred by
rigorous sampling
protocols)
Often accessible to a broad
audience with varied skills,
although species identification
requirements may limit
engagement to expert
amateurs unless training is
provided
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
514 O. L. PESCOTT ET AL.
can be used to generate hypotheses, particularly if
organized in a clear conceptual framework that high-
lights patterns; for example, see studies of biogeo-
graphical relationships (Preston & Hill, 1997) or
physio-ecological trait relationships (Waite & Sack,
2010). Indeed, Proctor (2010), commenting on Waite
& Sack (2010), notes that ‘there are occasions when
thorough exploration of an extensive block of data
produces ideas and insights that no amount of
hypothesis and test would have hit upon. The two
approaches are complementary. The hypothesistest
approach is effective, satisfying and progressive, as
long as you have a good supply of hypotheses. How-
ever, we also need systematic bodies of data, and the
ideas that come from exploring them’.
Rackham (2006: 404) and Peterken & Backmeroff
(1988) have made similar cases for long-term studies
in woodlands. We suggest that surveillance monitor-
ing fits into this mould (Wintle et al., 2010): the pat-
terns and data that arise from surveillance
monitoring can stimulate more focused studies on
the causes of ecological change. These are studies
that would be unlikely to be designed in a hypothet-
ic-deductive framework based around adaptive man-
agement, due to the resource constraints that would
stop this being a reality for all but the few areas of
habitat receiving regular management by conserva-
tionists. Particularly, volunteer-based surveillance
monitoring enables the net to be cast wide for rela-
tively few resources. The adaptive management/
QBM paradigm is likely to be most appropriate for
relatively well-resourced conservation professionals,
managing populations that are amenable to manage-
ment actions that can be easily deployed over a tar-
get area (e.g. the optimal harvesting of commercially
valuable animals). For the rest of the landscape, we
suggest that volunteer-based surveillance is likely to
produce data that inform policy makers and inspire
further, targeted research, or policy actions, rather
than directly triggering management interventions
at the level of a particular set of sites.
SURVEILLANCE MONITORING COULD BE JUSTIFIED BY
THE DETECTION OF ECOLOGICAL SURPRISES
Wintle et al. (2010) have cogently argued for the
wise use of surveillance monitoring, with a specific
focus on the idea that one of the justifications of sur-
veillance monitoring is the detection of unexpected
patterns of change: the ‘unknown unknowns’, or ‘sur-
prises’ sensu Hilborn (1987). After a short review of
real and hypothetical examples of surveillance-type
projects yielding new information, Wintle et al.
(2010) attempt to derive a model that would clarify
the circumstances under which investment in sur-
veillance monitoring over QBM would be justified.
This model is based within a decision-theoretic
framework; it considers the relative probabilities
that QBM or surveillance monitoring leads to the
achievement of a management goal or to the detec-
tion of an unforeseen pattern; information on the rel-
ative rewards of achieving management goals, the
costs of missing a harmful ecological novelty, and the
frequency with which harmful novelties occur are
also all required. The framework can in theory be
used to estimate the division of funds between QBM
and surveillance monitoring.
The framework of Wintle et al. (2010) is elegantly
constructed but, in reality, the parameterization of
the model would be extremely difficult. One particu-
lar obstacle in many cases would be the framing of
management rewards and the costs avoided by
detecting an unforeseen event: one could imagine a
targeted management experiment for a particular
species yielding population increases but, for most
mandated surveillance monitoring, it is much harder
to link the unobserved ‘cost’ of an unanticipated
change with subsequent action that would unambig-
Table 1. Continued
Project area
Question-based monitoring
(QBM) Surveillance monitoring (SM) Additional comments
Public impact Potentially less amenable as
a tool for community
engagement
Community engagement may be
one of the primary objectives
Communicating
key results
Straightforward where a
project has clarity of
purpose
Choice of result(s) typically
post hoc, and can be
challenging where resulting
data are not amenable to
statistical analysis
If volunteer-based SM is set up
with the purpose of
monitoring species of
conservation interest, then the
key results (trends in
distribution and/or abundance)
are likely to be clear
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 515
uously result in the direct avoidance of that cost.
Wintle et al. (2010) recognize these challenges, sug-
gesting that ‘the most tractable use of the model is
as a communication tool for increasing the clarity of
thought and discussion about the purpose and design
of the monitoring scheme’. As they emphasize, a sur-
veillance monitoring scheme makes choices about
what is important by virtue of its existence; in cases
where these choices can be narrowed down to specific
species or habitats, then a question-based, targeted
form of monitoring may reveal itself to be more effi-
cient (e.g. Pullin & Woodell, 1987). However, if we
were able to include additional benefits of surveil-
lance monitoring in such a decision framework, such
as the increase in a public’s level of investment in
environmental issues (Levins, 2003; Miller, 2005),
the highlighting of large-scale trends that are not
easily amenable to site-level management solutions
(e.g. climate change or nitrogen deposition) or the
fact that volunteer-based surveillance monitoring is
likely to be more heavily subsidized by the voluntary
activities of interested individuals across a large area
(Schmeller et al., 2009), then we might find that sur-
veillance monitoring is preferred over QBM in a lar-
ger range of scenarios.
These points present a challenge to that part of
the ecological literature that deals with the imple-
mentation of and the quantification and communica-
tion of results from government-mandated,
surveillance-type monitoring, without questioning
whether its logical foundations are well-justified. As
previously noted, Tulloch et al. (2013) dealt with the
question of whether atlas or monitoring scheme
approaches to volunteer-based monitoring are most
efficient (although their metric of impact investi-
gated was restricted to scientific citations). The more
general question of whether funding agencies should
support these types of volunteer-based surveillance
monitoring was not specifically addressed, although
it is highly likely that such activities represent excel-
lent value for money (Schmeller et al., 2009), even
though such monitoring does not typically focus its
activities on QBM. Possingham et al. (2012b) make
the case for greater attempts to formalize the various
benefits arising from monitoring activity that has
the primary goals of informing policy makers, edu-
cating, or detecting unexpected ecological change;
these authors speculate that, in schemes where these
benefits increase over time, cost-effectiveness will
also increase. Possingham et al. (2012b) conclude
that a greater effort should be made to quantify
these benefits, with the aim of building up a body of
data that allows ‘conservation returns on invest-
ments’ to be routinely calculated and compared
between projects, regardless of how those returns are
garnered. Of course, such an activity would itself
require the monitoring of a scheme’s impacts (e.g. on
awareness, engagement or its detection of ecological
surprises) using some formalized tool or framework.
For a volunteer-based organization engaged in bio-
logical recording and natural history, mobilizing vol-
unteers nationally on relatively straightforward
projects may often be a much more likely scenario
than participation in focused hypothesis- or question-
driven monitoring of the type that is often framed as
a direct rival to surveillance monitoring (Nichols &
Williams, 2006). Although it may appear that we
have set up a straw man in subjecting volunteer-
based atlas projects and monitoring schemes to criti-
cisms often directed at professional-led surveillance
monitoring, these types of citizen science projects
still fall into this category, and therefore the criti-
cisms are still relevant: for example, that volunteers
are wasting their time collecting information with
little conservation relevance, or that the collected
data are not as good as they could be for the effort
expended. The recognition that other benefits should
be formally taken into account (Possingham et al.,
2012a, b) should be welcomed as a useful addition to
the debate, and one that is applicable to a broader
range of monitoring activities; however, we should
also be aware that formalizing the collection of fur-
ther data on broader impacts may itself drain conser-
vation resources. Lindenmayer (2012) suggests that
an efficient way forward would be to find novel ways
to combine data from both surveillance and question-
based monitoring. The increasing popularity of
hierarchical Bayesian modelling for combining data
collected at different scales and with different obser-
vation processes is likely to prove useful for this
purpose (Pagel et al., 2014).
WHITHER VOLUNTEER-BASED
SURVEILLANCE MONITORING?
A range of opinions concerning the utility of surveil-
lance-type, long-term ecological monitoring schemes
exist. Many of the strongest objections come from
those that espouse more focused, hypothesis-led or
adaptive-management driven monitoring, and who
put forward reasonable doubts about its general cost-
effectiveness. We suggest that each type of monitor-
ing is complementary to the other, rather than being
mutually exclusive: in the UK, key drivers of habitat
change, nitrogen deposition for example, have been
demonstrated by focused small-scale research
(Stevens et al., 2004), as well as by large-scale, vol-
unteer-based recording (McClean et al., 2011). The
British and Irish experience suggests that flagship,
volunteer-based, long-term recording and monitoring
can contribute to a set of foci for national conserva-
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
516 O. L. PESCOTT ET AL.
tion (Defra, 2014), and an increased awareness by
people of their local environment, as long as contin-
ual efforts are made to engage new audiences. These
may then translate into new funds for conservation
and a greater desire for land management that sup-
ports biodiversity (Miller, 2005; Pretty, 2012). These
impacts are hard to capture, and citation-counting
exercises are unlikely to do them justice.
G. C. Druce (1932), in his The Comital Flora of the
British Isles (a list of which plants occurred in which
counties of Britain and Ireland, compiled using data
collected by professional and amateur botanists) was
already commenting on the uses of such a list for
documenting local extinctions, noting that he had
‘shown ... how many plants [had] been lost to our
Flora in the past’, and concluding, ‘I shudder to
think of the tale which may be shown even in the
year 2000’. Druce’s prediction has largely been borne
out, and further data collected, again, by a combina-
tion of professionals and expert amateurs have pro-
vided the evidence documenting impoverishments of
large parts of Britain’s plant diversity (Preston,
2000; Preston et al., 2002c; Walker et al., 2009;
McClean et al., 2011). Statements, then, such as
those that claim we are ‘just beginning to see the
multiple benefits of using data from citizen science
programs to monitor changes in the environment’
(Tulloch et al., 2013) can be seen to lack a historical
perspective. The data that have allowed conclusions
to be drawn about the current state of our plant bio-
diversity have accumulated gradually over time, and
have largely been the result of curiosity-driven sur-
veillance monitoring, combined with occasional, more
structured projects (Rich & Woodruff, 1990, 1996;
Braithwaite et al., 2006); these have often been orga-
nized by largely volunteer-based organizations that
would be highly unlikely to turn their focus com-
pletely towards conservation management. It is very
likely that continued input from volunteer experts in
numerous areas of surveillance monitoring will
continue to produce datasets that are highly valued
by governments, scientists, and the volunteers
themselves (but see Ellis & Waterton, 2004), and
that the development of variations on the monitoring
scheme or atlas will continue to contribute as much
to ecology and conservation in Britain and Ireland as
they have over the past 50 years (Roy et al., 2014;
Pocock et al., 2015).
ACKNOWLEDGEMENTS
We thank Chris D. Preston for many helpful com-
ments on the text. We also thank Stuart Newson
and one anonymous reviewer for comments that
improved the paper. We are indebted to the following
individuals who contributed to the development of
the National Plant Monitoring Scheme: Adam But-
ler, David Elston, David Noble, Ian Strachan, Iain
Diack, Louise Marsh, Lynn Heeley, John Redhead,
Peter Henrys, Quentin Groom, Richard Jefferson,
Stephen Freeman, Steve Buckland, Steve Langton,
Stuart Smith, Sue Southway, Susie Jarvis, and Tre-
vor Dines. The Biological Record Centre receives
support from the Joint Nature Conservation Commit-
tee and the Natural Environment Research Council
(via National Capability funding to the Centre for
Ecology and Hydrology, project NEC04932).
REFERENCES
Allen DE. 1976. The naturalist in Britain: a social history.
Princeton, NJ: Princeton University Press.
Allen DE. 1986. The botanists. Winchester: St Paul’s Bibliog-
raphies.
August T, Harvey M, Lightfoot P, Kilbey D, Papadopoulos
T, Jepson P. 2015. Emerging technologies for biological
recording. Biological Journal of the Linnean Society 115:
731749.
Austin GE, Calbrade NA, Mellan HJ, Hearn RD, Stroud
DA, Wotton SR, Holt CA. 2014. Waterbirds in the UK
2012/13: the Wetland Bird Survey. Thetford: BTO, RSPN
and JNCC, in association with the WWT.
Balmer DE, Gillings S, Caffrey B. 2013. Bird Atlas 2007
11: the breeding and wintering birds of Britain and Ireland.
Thetford: BTO Books.
Barlow KE, Briggs PA, Haysom KA, Hutson AM,
Lechiara NL, Racey PA, Walsh AL, Langton SD. 2015.
Citizen science reveals trends in bat populations: the
National Bat Monitoring Programme in Great Britain. Bio-
logical Conservation 182: 1426.
Bart J, Beyer HL. 2012. Use of citizen-science monitoring
for pattern discovery and biological inference. In: Gitzen
RA, Millspaugh JJ, Cooper AB, Licht DS, eds. Design and
analysis of long-term ecological monitoring studies. Cam-
bridge: Cambridge University Press, 253278.
Bolker BM. 2008. Ecological models and data in R. Prince-
ton, NJ: Princeton University Press.
Bonney R, Cooper CB, Dickinson J, Kelling S, Phillips
T, Rosenberg KV, Shirk J. 2009. Citizen science: a
developing tool for expanding science knowledge and scien-
tific literacy. BioScience 59: 977984.
Bonney R, Shirk J, Phillips TB. 2013. Citizen science. In:
Gunstone R, ed. Encyclopedia of science education. Dordr-
echt: Springer, 13.
Braithwaite ME, Ellis RW, Preston CD. 2006. Change in
the British flora 19872004. London: Botanical Society of
the British Isles.
Brereton TM, Cruickshanks KL, Risely K, Noble DG,
Roy DB. 2011. Developing and launching a wider country-
side butterfly survey across the United Kingdom. Journal
of Insect Conservation 15: 279290.
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 517
Brotons L, Herrando S, Pla M. 2007. Updating bird spe-
cies distribution at large spatial scales: applications of habi-
tat modelling to data from long-term monitoring programs.
Diversity and Distributions 13: 276288.
Butchart S, Walpole M, Collen B. 2010. Global
biodiversity: indicators of recent declines. Science 328:
11641168.
Cham S, Nelson B, Parr A, Prentice S, Smallshire D,
Taylor P. 2014. Atlas of dragonflies in Britain and Ire-
land. Shrewsbury: Field Studies Council.
Chen G, Kery M, Zhang J, Ma K. 2009. Factors affecting
detection probability in plant distribution studies. Journal
of Ecology 97: 13831389.
Cohn JP. 2008. Citizen science: can volunteers do real
research? BioScience 58: 192197.
Convention on Biological Diversity.2010. Report of the
tenth meeting of the conference of the parties to the conven-
tion on biological diversity. Annex. Decision No. X/2. Mon-
treal, Canada: UNEP/CBD.
Croft JM, Preston CD. 1999. Recording studies and inter-
pretative projects in the British Isles, 198397. Acta Botan-
ica Fennica 162: 3541.
Defra.2014. UK biodiversity indicators 2014: measuring pro-
gress towards halting biodiversity loss. London: Defra.
Devictor V, Whittaker RJ, Beltrame C. 2010. Beyond
scarcity: citizen science programmes as useful tools for con-
servation biogeography. Diversity and Distributions 16:
354362.
Dickinson JL, Shirk J, Bonter D, Bonney R, Crain RL,
Martin J, Phillips T, Purcell K. 2012. The current state
of citizen science as a tool for ecological research and public
engagement. Frontiers in Ecology and the Environment 10:
291297.
Dickinson JL, Bonney R, eds. 2012. Citizen science: public
participation in environmental research. Ithaca, NY: Cor-
nell University Press.
Dickinson JL, Zuckerberg B, Bonter DN. 2010. Citizen
science as an ecological research tool: challenges and bene-
fits. Annual Review of Ecology, Evolution, and Systematics
41: 149172.
Dines TD. 1996. Atlas 2000 instruction booklet. Monk’s
Wood: BSBI/BRC.
Dony JG. 1967. Flora of Hertfordshire. Hitchin: Urban
District Council.
Druce GC. 1932. The comital flora of the British Isles.
Arbroath: T. Buncle & Co.
Le Duc MG, Hill MO, Sparks TH. 1992. A method for pre-
dicting the probability of species occurrence using data from
systematic surveys. Watsonia 19: 97105.
Ellis R, Waterton C. 2004. Environmental citizenship in
the making: the participation of volunteer naturalists in
UK biological recording and biodiversity policy. Science and
Public Policy 31: 95105.
Field SA, O’Connor PJ, Tyre AJ, Possingham HP. 2007.
Making monitoring meaningful. Austral Ecology 32: 485
491.
Firbank LG, Ellis NE, Hill MO, Lockwood AJ, Swetnam
RD. 1998. Mapping the distribution of weeds in Great Brit-
ain in relation to national survey data and to soil type.
Weed Research 38: 110.
FOEN.2014. Swiss biodiversity monitoring BDM. Bern: Fed-
eral Office for the Environment.
Fox R, Oliver TH, Harrower C, Parsons MS, Thomas
CD, Roy DB. 2014. Long-term changes to the frequency of
occurrence of British moths are consistent with opposing
and synergistic effects of climate and land-use changes.
Journal of Applied Ecology 51: 949957.
Fuller RM. 1987. The changing extent and conservation
interest of lowland grasslands in England and Wales: a
review of grassland surveys 19301984. Biological Conser-
vation 40: 281300.
Fuller RJ, Smith KW, Grice PV, Currie FA, Quine CP.
2007. Habitat change and woodland birds in Britain: implica-
tions for management and future research. Ibis 149: 261268.
Gelman A, Hill J. 2007. Data analysis using regression and
multilevel/hierarchical models. New York, NY: Cambridge
University Press.
Gerard F, Blank L, Bunce RGH, Carmel Y, Caudullo G,
Clerici N, Deshayes M, Erikstad L, Estreguil C,
Framstad E, Granholm A-H, Halabuk A, Halada L,
Harari-Kremer R, Hazeu GW, Hennekens SM,
Holmgren J, Kikas T, Kuusemets V, Lang M, Levin N,
Luck-Vogel M, Morton D, Mucher CA, Nilsson M,
Nordkvist K, Olsson H, Olsvig-Whittaker L, Raet J,
Roberts W, Roerink GJ, Sepp K, Scholefield PA, Vain
A, Van Calster H, Weissteiner CJ. 2012. Assessing the
role of EO in biodiversity monitoring: options for integrat-
ing in-situ observations with EO within the context of the
EBONE concept. Wallingford, UK: NERC/Centre for Ecol-
ogy & Hydrology. Retrieved February 8, 2015, from http://
nora.nerc.ac.uk/17798/
Gitzen RA, Millspaugh JJ, Cooper AB, Licht DS, eds.
2012. Design and analysis of long-term ecological monitoring
studies. Cambridge: Cambridge University Press.
Gregory RD, van Strien A, Vorisek P, Meyling AWG,
Noble DG, Foppen RPB, Gibbons DW. 2005. Developing
indicators for European birds. Philosophical Transactions
of the Royal Society of London Series B, Biological Sciences
360: 269288.
Groom QJ. 2013. Estimation of vascular plant occupancy
and its change using kriging. New Journal of Botany 3: 33
46.
Harding PT, Sheail J. 1992. The Biological Records Centre
a pioneer in data gathering and retrieval. In: Harding P,
ed. Biological recording of changes in British wildlife. Lon-
don: HMSO, 519.
Harris SJ, Risely K, Massimino D, Newson SE, Eaton
MA, Musgrove AJ, Noble DG, Procter D, Baillie SR.
2014. The Breeding Bird Survey 2013. BTO research report
658. Thetford: British Trust for Ornithology.
Hilborn R. 1987. Living with uncertainty in resource man-
agement. North American Journal of Fisheries Management
7: 15.
Hilborn R, Mangel M. 1997. The ecological detective: con-
fronting models with data. Princeton, NJ: Princeton Uni-
versity Press.
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
518 O. L. PESCOTT ET AL.
Hill MO. 2003. Using data from local floras for ecological
research. Watsonia 24: 321329.
Hill MO. 2012. Local frequency as a key to interpreting spe-
cies occurrence data when recording effort is not known.
Methods in Ecology and Evolution 3: 195205.
Hill MO, Preston CD. 2015. Disappearance of boreal plants
in southern Britain habitat loss or climate change? Biologi-
cal Journal of the Linnean Society 115: 598610.
Hill MO, Preston CD. 2014. Changes in distribution and
abundance, 19602013. In: Blockeel TL, Bosanquet SDS,
Hill MO, Preston CD, eds. Atlas of British and Irish bryo-
phytes,Vol. 1. Newbury: Pisces Publications, 3449.
Hochachka WM, Fink D, Zuckerberg B. 2012. Use of citi-
zen-science monitoring for pattern discovery and biological
inference. In: Gitzen RA, Millspaugh JJ, Cooper AB, Licht
DS, eds. Design and analysis of long-term ecological moni-
toring studies. Cambridge: Cambridge University Press,
460480.
Hooftman DAP, Bullock JM. 2012. Mapping to inform con-
servation: a case study of changes in semi-natural habitats
and their connectivity over 70 years. Biological Conserva-
tion 145: 3038.
Irvine KM, Rodhouse TJ. 2010. Power analysis for trend in
ordinal cover classes: implications for long-term vegetation
monitoring. Journal of Vegetation Science 21: 11521161.
Isaac NJB, van Strien AJ, August TA, de Zeeuw MP,
Roy DB. 2014. Statistics for citizen science: extracting sig-
nals of change from noisy ecological data. Methods in Ecol-
ogy and Evolution 5: 10521060.
JNCC. 2008.The UK terrestrial biodiversity surveillance
strategy. Proposal for a terrestrial biodiversity surveillance
and monitoring strategy. Peterborough: JNCC.
Johnson PCD, Barry SJE, Ferguson HM, M
uller P.
2015. Power analysis for generalized linear mixed models
in ecology and evolution. Methods in Ecology and Evolution
6: 133142.
Jones JPG, Collen B, Atkinson G, Baxter PWJ, Bubb P,
Illian JB, Katzner TE, Keane A, Loh J, McDonald-
Madden E, Nicholson E, Pereira HM, Possingham
HP, Pullin AS, Rodrigues ASL, Ruiz-Gutierrez V,
Sommerville M, Milner-Gulland EJ. 2011. The Why,
What, and How of Global Biodiversity Indicators Beyond
the 2010 Target. Conservation Biology 25: 450457.
doi:10.1111/j.1523-1739.2010.01605.x
Jones JPG. 2013. Monitoring in the real world. In: Collen
B, Pettorelli N, Baillie JEM, Durant SM, eds. Biodiversity
monitoring and conservation: bridging the gap between glo-
bal commitment and local action. Chichester: John Wiley &
Sons, 335347.
Legg CJ, Nagy L. 2006. Why most conservation monitoring
is, but need not be, a waste of time. Journal of Environ-
mental Management 78: 194199.
Levins R. 2003. Whose scientific method? Scientific methods
for a complex world. In: Tickner JA, ed. Precaution,
environmental science, and preventive public policy. Wash-
ington, DC: Island Press, 355368.
Lindenmayer DB, Likens GE, Krebs CJ, Hobbs RJ. 2010.
Improved probability of detection of ecological ‘surprises’.
Proceedings of the National Academy of Sciences of the Uni-
ted States of America 107: 2195721962.
Lindenmayer D. 2012. Making monitoring up-front and
centre in Australian biodiversity conservation. In: Linden-
mayer D, Gibbons P, eds. Biodiversity monitoring in Aus-
tralia. Collingwood, VIC: CSIRO Publishing, 714.
Lindenmayer DB, Likens GE. 2010a. The science and
application of ecological monitoring. Biological Conservation
143: 13171328.
Lindenmayer D, Likens G. 2010b. Effective ecological mon-
itoring. Collingwood, VIC: CSIRO Publishing.
Maes D, Isaac NJB, Harrower CA, Collen B, Roy DB,
van Strien AJ. 2015. The use of opportunistic data for
IUCN Red List assessments. Biological Journal of the Lin-
nean Society 115: 690706.
Manley PN, Zielinski WJ, Schlesinger MD, Mori SR.
2004. Evaluation of a multiple-species approach to monitor-
ing species at the ecoregional scale. Ecological Applications
14: 296310.
McClean CJ, Berg LJ, Ashmore MR, Preston CD.
2011. Atmospheric nitrogen deposition explains patterns
of plant species loss. Global Change Biology 17: 2882
2892.
McDonald-Madden E, Baxter PW, Fuller RA, Martin
TG, Game ET, Montambault J, Possingham HP. 2010.
Monitoring does not always count. Trends in Ecology &
Evolution 25: 547550.
Miller JR. 2005. Biodiversity conservation and the extinction
of experience. Trends in Ecology & Evolution 20: 430434.
Miller KM, Mitchell BR. 2014. A new tool for power
analysis of fixed plot data: using simulations and mixed
effects models to evaluate forest metrics. Ecosphere 5:
art110.
Miller-Rushing A, Primack R, Bonney R. 2012. The his-
tory of public participation in ecological research. Frontiers
in Ecology and the Environment 10: 285290.
Nichols JD, Williams BK. 2006. Monitoring for conserva-
tion. Trends in Ecology & Evolution 21: 668673.
Oswald PH, Preston CD, eds. 2011. John Ray’s Cambridge
Catalogue (1660). London: Ray Society.
Pagel J, Anderson BJ, O’Hara RB, Cramer W, Fox R,
Jeltsch F, Roy DB, Thomas CD, Schurr FM. 2014.
Quantifying range-wide variation in population trends from
local abundance surveys and widespread opportunistic
occurrence records. Methods in Ecology and Evolution 5:
751760.
Palmer MA, Bratton JH, eds. 1995. A sample survey of the
flora of Britain and Ireland: the Botanical Society of the
British Isles monitoring scheme 19871988. Peterborough:
Joint Nature Conservation Committee.
Pearman DA, Preston CD, Roy DB, Stewart A. 1998.
The use of BSBI monitoring scheme data to predict nation-
ally scarce species in Britain. Watsonia 22: 2128.
Pearman PB, Weber D. 2007. Common species determine
richness patterns in biodiversity indicator taxa. Biological
Conservation 138: 109119.
Perring FH. 1970. The last seventy years. In: Perring FH, ed.
The flora of a changing Britain. Classey: Hampton, 128135.
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 519
Perring F. 1992. BSBI distribution maps scheme the first
40 years. In: Harding P, ed. Biological recording of changes
in British wildlife. London: HMSO, 14.
Perring F, Walters SM. 1962. Atlas of the British Flora.
London: Thomas Nelson & Sons.
Perring FH, Farrell L. 1977. British red data books: 1, vas-
cular plants. Lincoln: Royal Society for Nature Conservation.
Pescott OL, Simkin JM, August TA, Dore AJ, Randle Z,
Botham M. 2015. Air pollution and its effects on lichens,
bryophytes, and lichen-feeding Lepidoptera: review and evi-
dence from biological records. Biological Journal of the Lin-
nean Society 115: 611635.
Peterken GF, Backmeroff C. 1988. Long-term monitoring
in unmanaged woodland nature reserves. Peterborough:
Nature Conservancy Council.
Pickett STA, Kolasa J, Jones CG. 2007. Ecological under-
standing: the nature of theory and the theory of nature. Lon-
don: Academic Press.
Platt JR. 1964. Strong inference. Science 146: 347353.
Pocock MJO, Roy HE, Preston CD, Roy DB. 2015a. The
Biological Records Centre: a pioneer of citizen science. Bio-
logical Journal of the Linnean Society 115: 475493.
Pocock MJO, Newson SE, Hendersen IG, Peyton J, Suth-
erland WJ, Noble DG, Ball SG, Beckmann BC, Biggs J,
Brereton T, Bullock DJ, Buckland ST, Edwards M,
Eaton MA, Harvey MC, Hill MO, Horlock M, Hubble
DS, Julian AM, Mackey EC, Mann DJ, Marshall MJ,
Medlock JM, O’Mahony EM, Pacheco M, Porter K, Pre-
ntice S, Procter DA, Roy HE, Southway SE, Shortall
CR, Stewart AJA, Wembridge DE, Wright MA, Roy DB.
2015b. Developing and enhancing biodiversity monitoring
programmes: a collaborative assessment of priorities. Jour-
nal of Applied Ecology. doi:10.1111/1365-2664 12423
Pollard E, Yates TJ. 1993. Monitoring butterflies for ecology
and conservation. The British butterfly monitoring scheme.
Institute of Terrestrial Ecology and Joint Nature Conserva-
tion Committee. London: Chapman and Hall.
Porter K, Leach SJ. 2013. Botanical records and their role
in shaping nature conservation priorities and actions an
English perspective. New Journal of Botany 3: 148155.
Possingham HP, Wintle BA, Fuller RA, Joseph LN.
2012a. The conservation return on investment from ecologi-
cal monitoring. In: Lindenmayer D, Gibbons P, eds. Biodi-
versity monitoring in Australia. Collingwood, VIC: CSIRO
Publishing, 4962.
Possingham HP, Fuller RA, Joseph LN. 2012b. Choosing
among long-term ecological monitoring programs and know-
ing when to stop. In: Gitzen RA, Millspaugh JJ, Cooper AB,
Licht DS, eds. Design and analysis of long-term ecological
monitoring studies. Cambridge: Cambridge University
Press, 498508.
Powney GD, Rapacciuolo G, Preston CD, Purvis A, Roy
DB. 2014. A phylogenetically-informed trait-based analysis
of range change in the vascular plant flora of Britain. Bio-
diversity and Conservation 23: 171185.
Preston CD. 2000. Engulfed by suburbia or destroyed by the
plough: the ecology of extinction in Middlesex and Cam-
bridgeshire. Watsonia 23: 5982.
Preston CD, Telfer MG, Arnold HR, Carey PD,
Cooper JM, Dines TD, Hill MO, Pearman DA, Roy
DB, Smart SM. 2002a. The changing flora of the UK.
London: Defra.
Preston CD. 2003. Perceptions of change in English County
Floras, 16601960. Watsonia 24: 287304.
Preston CD, Telfer MG, Roy DB, Carey PD, Hill MO,
Meek WR, Rothery P, Smart SM, Smith GM, Walker
KJ. 2003. The changing distribution of the flora of the Uni-
ted Kingdom: technical report. Monk’s Wood: CEH.
Preston CD. 2013. Following the BSBI’s lead: the influence
of the Atlas of the British flora, 19622012. New Journal of
Botany 3: 214.
Preston CD, Croft JM, Pearman DA. 2002b. The 1962
Atlas and later recording projects. In: Preston CD, Pearman
DA, Dines TD, eds. New Atlas of the British and Irish
Flora. Oxford: Oxford University Press, 38.
Preston CD, Hill MO. 1997. The geographical relationships
of British and Irish vascular plants. Botanical Journal of
the Linnean Society 124: 1120.
Preston CD, Pearman DA, Dines TD, eds. 2002c. New
Atlas of the British and Irish Flora. Oxford: Oxford Univer-
sity Press.
Pretty JN. 2012. The earth only endures: on reconnecting
with nature and our place in it. London: Earthscan.
Proctor MCF. 2010. Trait correlations in bryophytes: explor-
ing an alternative world. New Phytologist 185: 13.
Pullin AS, Woodell SR. 1987. Response of the fen violet,
Viola persicifolia Schreber, to different management
regimes at Woodwalton Fen National Nature Reserve, Cam-
bridgeshire, England. Biological Conservation 41: 203217.
Rackham O. 2006. Woodlands (new naturalist). London:
Harper Collins.
Ray J. 1660. Catalogus plantarum circa Cantabrigiam na-
scentium. Cambridge.
Ray J. 1670. Catalogus plantarum Angliae, et insularum ad-
jacentium. London.
Rich T. 1996. Changing fortunes the BSBI monitoring
scheme. British Wildlife 8: 711.
Rich T, Donovan P, Harmes P, Knapp A, McFarlane M,
Marrable C, Muggeridge N, Nicholson R, Reader M,
Reader P, Rich E, White P. 1996. Flora of Ashdown for-
est. Sussex Botanical Recording Society.
Rich T. 1998. Squaring the circles bias in distribution
maps. British Wildlife 9: 213219.
Rich TCG, Beesley S, Goodwillie R. 2001. Changes in the
vascular plant flora of Ireland between pre-1960 and 1987
1988: the BSBI monitoring scheme. The Irish Naturalists’
Journal 26: 333350.
Rich TCG, Smith PA. 1996. Botanical recording, distribu-
tion maps and species frequency. Watsonia 21: 155167.
Rich TCG, Woodruff ER. 1990. BSBI monitoring scheme
198788: volume 1 text. Peterborough: Nature Conservancy
Council.
Rich TCG, Woodruff ER. 1992. Recording bias in botanical
surveys. Watsonia 19: 7395.
Rich TCG, Woodruff ER. 1996. Changes in the vascular
plant floras of England and Scotland between 19301960
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
520 O. L. PESCOTT ET AL.
and 19871988: the BSBI monitoring scheme. Biological
Conservation 75: 217229.
Robertson MP, Cumming GS, Erasmus BFN. 2010. Get-
ting the most out of atlas data. Diversity and Distributions
16: 363375.
Roy D, Harding P, Preston C, Roy H, eds. 2014. Celebrat-
ing 50 years of the Biological Records Centre. Wallingford:
Centre for Ecology and Hydrology.
Roy DB, Ploquin EF, Randle Z, Risely K, Botham MS,
Middlebrook I, Noble D, Cruickshanks K, Freeman
SN, Brereton TM. 2015. Comparison of trends in butterfly
populations between monitoring schemes. Journal of Insect
Conservation 19: 313324.
Roy DB, Rothery P, Brereton T. 2007. Reduced-effort
schemes for monitoring butterfly populations. Journal of
Applied Ecology 44: 9931000.
Schmeller DS, Henry PY, Julliard R, Gruber B, Clobert
J, Dziock F, Lengyel S, Nowicki P, Deri E, Budrys E.
2009. Advantages of volunteer-based biodiversity monitor-
ing in Europe. Conservation Biology 23: 307316.
Silvertown J. 2009. A new dawn for citizen science. Trends
in Ecology & Evolution 24: 467471.
Spellerberg IF. 2005. Monitoring ecological change. Cam-
bridge: Cambridge University Press.
Stevens CJ, Dise NB, Mountford JO, Gowing DJ. 2004.
Impact of nitrogen deposition on the species richness of
grasslands. Science 303: 18761879.
Stroh PA, Leach SJ, August TA, Walker KJ, Pearman
DA, Rumsey FJ, Harrower CA, Fay MF, Martin JP,
Pankhurst T, Preston CD, Taylor I. 2014. A Vascular
Plant Red List for England. Bristol: Botanical Society of
Britain and Ireland.
Suter GW II. 1993. Ecological risk assessment. Chelsea, MI:
Lewis Publishers.
Telfer MG, Preston CD, Rothery P. 2002. A general
method for measuring relative change in range size from
biological atlas data. Biological Conservation 107: 99109.
Thomas JA. 2005. Monitoring change in the abundance and
distribution of insects using butterflies and other indicator
groups. Philosophical Transactions of the Royal Society of
London Series B, Biological Sciences 360: 339357.
Tulloch AIT, Possingham HP, Joseph LN, Szabo J,
Martin TG. 2013. Realising the full potential of citizen sci-
ence monitoring programs. Biological Conservation 165:
128138.
Urquhart NS, Kincaid TM. 1999. Designs for detecting
trend from repeated surveys of ecological resources. Journal
of Agricultural, Biological, and Environmental Statistics 4:
404414.
Urquhart NS. 2012. The role of monitoring design in detect-
ing trend in long-term ecological monitoring studies. In:
Gitzen RA, Millspaugh JJ, Cooper AB, Licht DS,
eds. Design and analysis of long-term ecological monitoring
studies. Cambridge: Cambridge University Press, 151173.
Va
ck
a
r D, ten Brink B, Loh J, Baillie JEM, Reyers B.
2012. Review of multispecies indices for monitoring
human impacts on biodiversity. Ecological Indicators 17:
5867.
Waite M, Sack L. 2010. How does moss photosynthesis
relate to leaf and canopy structure? Trait relationships for
10 Hawaiian species of contrasting light habitats. New Phy-
tologist 185: 156172.
Walker KJ. 2003a. Using data from local floras to assess flo-
ristic change. Watsonia 24: 305319.
Walker KJ. 2003b. One species lost every year? An evalua-
tion of plant extinctions in selected British vice-counties
since 1900. Watsonia 24: 359374.
Walker K, Dines T, Hutchinson N, Freeman S. 2010b.
Designing a new plant surveillance scheme for the UK.
Peterborough: JNCC.
Walker K, Pearman D, Ellis B, McIntosh J, Lockton A.
2010a. Recording the British and Irish flora 20102020.
London: BSBI.
Walker KJ, Pescott OL, Harris F, Cheffings C, New H,
Bunch N, Roy DB. 2015. Making plants count: a new
National Plant Monitoring Scheme for the United Kingdom.
British Wildlife 26: 243250.
Walker KJ, Preston CD, Boon CR. 2009. Fifty years of
change in an area of intensive agriculture: plant trait
responses to habitat modification and conservation, Bed-
fordshire, England. Biodiversity and Conservation 18:
35973613.
Warren MS, Hill JK, Thomas JA, Asher J, Fox R, Huntley
B, Roy DB, Telfer MG, Jeffcoate S, Harding P, Jeffco-
ate G, Willis SG, Greatorex-Davies JN, Moss D, Thomas
CD. 2001. Rapid responses of British butterflies to opposing
forces of climate and habitat change. Nature 414: 6569.
doi:10.1038/35102054
Wintle BA, Runge MC, Bekessy SA. 2010. Allocating mon-
itoring effort in the face of unknown unknowns. Ecology
Letters 13: 13251337.
Yoccoz NG, Nichols JD, Boulinier T. 2001. Monitoring of
biological diversity in space and time. Trends in Ecology &
Evolution 16: 446453.
©2015 The Linnean Society of London, Biological Journal of the Linnean Society, 2015, 115, 505–521
ECOLOGICAL MONITORING WITH CITIZEN SCIENCE 521
... In this situation, citizen science, the involvement of volunteers in generating scientific data, is often seen as an attractive option by policy-makers to fill knowledge gaps in biodiversity datasets [25]. Citizen science can provide broad geographic coverage of species' occurrences [26,27] and has been used to monitor trends in the distributions [28,29] and abundances of species [30][31][32]. Biodiversity-based citizen science typically follows two methods of data collection: structured biodiversity recording, i.e. recording that follows some sort of protocol or design, and unstructured recording, which refers to the ad hoc collection of records [28]. ...
... Citizen science can provide broad geographic coverage of species' occurrences [26,27] and has been used to monitor trends in the distributions [28,29] and abundances of species [30][31][32]. Biodiversity-based citizen science typically follows two methods of data collection: structured biodiversity recording, i.e. recording that follows some sort of protocol or design, and unstructured recording, which refers to the ad hoc collection of records [28]. The provision of early warnings of the arrival of new IAS is one area in which this has been successful [33][34][35]. ...
... These prioritised needs could be used to help focus existing resources for data collection. For example, supporting the development of standardised methodologies would enable structured biological recording [28], thus ensuring consistency in the way records are collected in new schemes. This would also support recognition of the importance of mentoring, training and supporting volunteers in existing, or newly establishing, schemes which could result in increased funding for these activities [57]. ...
Article
Full-text available
Biodiversity monitoring plays an essential role in tracking changes in ecosystems, species distributions and abundances across the globe. Data collected through both structured and unstructured biodiversity recording can inform conservation measures designed to reduce, prevent, and reverse declines in valued biodiversity of many types. However, given that resources for biodiversity monitoring are limited, it is important that funding bodies prioritise investments relative to the requirements in any given region. We addressed this prioritisation requirement for a biodiverse Mediterranean island (Cyprus) using a three-stage process of expert-elicitation. This resulted in a structured list of twenty biodiversity monitoring needs; specifically, a hierarchy of three groups of these needs was created using a consensus approach. The most highly prioritised biodiversity monitoring needs were those related to the development of robust survey methodologies, and those ensuring that sufficiently skilled citizens are available to contribute. We discuss ways that the results of our expert-elicitation process could be used to support current and future biodiversity monitoring in Cyprus.
... The monitoring of trends in species' distributions or populations is a fundamental activity within ecology and conservation (Lindenmayer and Likens, 2010). The resulting trends may have different uses depending on the rationale and design of the underlying monitoring program, but much "surveillance"-style monitoring is driven by both policy requirements and the curiosity of invested naturalists (Pescott et al., 2015;Schmeller et al., 2009). This means that feedback on trends to nonscientist stakeholders of various types is often a key program output. ...
... The Botanical Society of Britain and Ireland (BSBI) have a long history of collecting species occurrence data to inform ecological and conservation research (Pescott et al., 2015;Preston, 2013), and this has resulted in two published plant distribution atlases during the past one-hundred years (Perring and Walters, 1962;Preston et al., 2002). Data collection for a third atlas (Walker et al., 2010), to be published in 2023, is now complete. ...
Preprint
Full-text available
Temporal trends in species occupancy or abundance are a fundamental source of information for ecology and conservation. Model-based uncertainty in these trends is often communicated as frequentist confidence or Bayesian credible intervals; however, these are often misinterpreted in various ways, even by scientists. Research from the science of information visualisation indicates that line ensemble approaches that depict multiple outcomes compatible with a fitted model or data may be superior for the clear communication of model-based uncertainty. The discretisation of continuous probability information into frequency bins has also been shown to be useful for communicating with non-specialists. We present a simple and widely applicable approach that combines these two ideas, and which can be used to clearly communicate model-based uncertainty in species trends (or composite indicators) to stakeholders. We also show how broader ontological uncertainty can also be communicated via trend plots using risk-of-bias visualisation approaches developed in other disciplines. The techniques are demonstrated using the example of long-term plant distributional change in Britain, but are applicable to any temporal data consisting of averages and associated uncertainty measures. Our approach supports calls for full transparency in the scientific process by clearly displaying the multiple sources of uncertainty that can be estimated by researchers.
... Britain has a unique and rich history of botanical recording which continues to the present day (Harding & Sheail, 1992;Pescott et al., 2015;Pocock et al., 2015). ...
Article
Full-text available
Biological recording in Britain has increased in accessibility and popularity due to recent technological advances. However, remote locations may still be under-recorded, particularly for aquatic plants and taxonomically challenging groups. We describe a set of notable botanical discoveries made in 2021 at Corrour in the Scottish Highlands (v.c.97 Westerness), including Baldellia repens, Illecebrum verticillatum (new to Scotland) and six British altitudinal records. At the time of writing, there are now four Nationally Rare vascular plant taxa recorded on the estate and 29 Nationally Scarce taxa. These findings demonstrate the value of collaboration between land managers, ecologists, BSBI staff and the local community. Both B. repens and I. verticillatum are well established at Corrour in considerable abundance and with clear evidence of regeneration. B. repens also occurs in the Tay catchment and may have arrived at Corrour via vegetative dispersal by waterfowl. The origin of I. verticillatum is more ambiguous but suggested mechanisms of dispersal include forestry, the railway or hydroelectricity workings. Despite a likely element of accidental human-mediated spread, I. verticillatum should be considered an intriguing addition to the flora of Westerness. Climate change could facilitate further establishment of this taxa in northern parts of Britain, and it is likely that other new records of both B. repens and I. verticillatum await.
... The monitoring of trends in species' distributions or populations is a fundamental activity within ecology and conservation (Lindenmayer and Likens, 2010). The resulting trends may have different uses depending on the rationale and design of the underlying monitoring program, but much "surveillance"-style monitoring is driven by both policy requirements and the curiosity of invested naturalists (Pescott et al., 2015;Schmeller et al., 2009). This means that feedback on trends to non-scientist stakeholders of various types is often a key program output. ...
Article
Full-text available
Temporal trends in species occupancy or abundance are a fundamental source of information for ecology and conservation. Model-based uncertainty in these trends is often communicated as frequentist confidence or Bayesian credible intervals, however, these are often misinterpreted in various ways, even by scientists. Research from the science of information visualisation indicates that line ensemble approaches that depict multiple outcomes compatible with a fitted model or data may be superior for the clear communication of model-based uncertainty. The discretisation of continuous probability information into frequency bins has also been shown to be useful for communicating with non-specialists. We present a simple and widely applicable approach that combines these two ideas, and which can be used to clearly communicate model-based uncertainty in species trends (or composite indicators) to stakeholders. We also show how broader ontological uncertainty can be communicated via trend plots using risk-of-bias visualisation approaches developed in other disciplines. The techniques are demonstrated using the example of long-term plant distributional change in Britain, but are applicable to any temporal data consisting of averages and associated uncertainty measures. Our approach supports calls for full transparency in the scientific process by clearly displaying the multiple sources of uncertainty that can be estimated by researchers.
... Laypersons have been involved in the recording of alien plant species for a long time. For example, since the early 1960s, amateur botanists have been systematically recording plants, including aliens, in the British Isles to create distribution maps of the British flora (Pescott et al. 2015). The potentially crucial role of citizen science for early detection and monitoring of invasive plants has only become apparent over the last decades and with the increasing use of record-taking through the Internet and smartphone applications. ...
... Laypersons have been involved in the recording of alien plant species for a long time. For example, since the early 1960s, amateur botanists have been systematically recording plants, including aliens, in the British Isles to create distribution maps of the British flora (Pescott et al. 2015). The potentially crucial role of citizen science for early detection and monitoring of invasive plants has only become apparent over the last decades and with the increasing use of record-taking through the Internet and smartphone applications. ...
... Laypersons have been involved in the recording of alien plant species for a long time. For example, since the early 1960s, amateur botanists have been systematically recording plants, including aliens, in the British Isles to create distribution maps of the British flora (Pescott et al. 2015). The potentially crucial role of citizen science for early detection and monitoring of invasive plants has only become apparent over the last decades and with the increasing use of record-taking through the Internet and smartphone applications. ...
... Of key importance is that the plant species that volunteers are asked to focus on were derived from the weighted occurrence of species within the NVC system, so as to be representative of each community, while at the same time avoiding species that would be taxonomically challenging for amateur botanists. This is termed the 'wildflower level' of participation to minimise identification errors (Pescott et al., 2015). As data from the NPMS align with NVC communities, it will be straightforward to adapt the distribution modelling methods we describe to make use of this CS resource as its database expands (https://www.npms.org. ...
Article
Full-text available
Species distribution models (SDMs) have been widely used to create maps of expected species incidence, often using citizen science (CS) occurrence data as inputs. Environmental policy is informed by knowledge of community distributions, but there have been fewer attempts to utilise the potential of community distribution models (CDMs) to predict these. Many countries have vegetation community classification systems which include phytosociological information on individual species. Within Great Britain, the National Vegetation Classification (NVC) is the primary standard for vegetation communities, and whilst maps have been produced at regional scales, cost‐effective techniques are required for national scales. Published NVC occurrence records of 22 upland NVC communities in England and Wales were used as observed occurrences (presence‐only data). Predictors for the CDMs were enhanced vegetation index (EVI), elevation, slope, aspect, temperature and rainfall. Five modelling methods were investigated: generalised linear models (GLMs), support vector machines (SVM), random forests (RF), maximum entropy (MaxEnt) and maximum likelihood (MaxLike). Model quality was assessed via bootstrapping via area under the curve (AUC), true skill statistic (TSS) and Kappa index. There were only small differences in the accuracy of the models (median TSS model accuracy 0.742; range 0.280 to 0.873) with RF models the best overall CDM method. Across all NVC communities, summer and winter maximum temperatures and annual rainfall were the most important predictor variables. NVCs with spatially disjunct distributions in both lowlands and uplands, or that responded to localised management or environmental conditions, were poorly predicted. Synthesis and applications. Vegetation communities can be reliably predicted at large spatial scales using CDMs from extant datasets. Management practitioners can use community‐level predictions to design targeted field surveys for individual species typically associated with specific communities. Most existing CS survey schemes focus on species rather than communities. Hence future development of new CS schemes similar to the National Plant Monitoring Scheme (NPMS), that aligns with the NVC, will enable CS data to generate up‐to‐date maps of both communities and species.
... 33 & 36;Panareda Clopés, 2000; Jiménez-Alfaro, 2009, p. 125;Martínez Labarga, 2014; etc.). Pero, la bondad más sobresaliente estriba en que, cuanto más equilibrados, sistemáticos y exhaustivos sean la documentación o/y el levantamiento de la información, más fidedigna será la comparación geográfica de las distribuciones, ya sea entre diferentes taxones dentro de un mismo territorio(Nuet Badia & Panareda Clopés, 1991-1993González Granados, 1997;Panareda Clopés et al., 2005; Gastón González, 2008, p. 11;García-Abad Alonso, 2019), como entre el mismo taxón en territorios diferentes (García-AbadAlonso, 2015Alonso, , 2016García-Abad Alonso et al., 2018).En relación con el tamaño de unidad cartográfica para representar distribuciones de plantas, los estudios corológicos a partir de unidades de pocos kilómetros de lado (1 a 3 km: resoluciones de 1 a 9 km²) favorece el conocimiento de la flora a escala local(Green, 2008;Pescott et al., Boletín de la Asociación de Geógrafos Españoles, (92) ...
Article
Full-text available
La distribución geográfica de las plantas, su frecuencia y abundancia constituyen importantes resortes para conocer la biodiversidad. A estas propiedades conviene unir la temporalidad vegetativa para mejorar el conocimiento de sus ocurrencias geográficas. En este trabajo se analizan estas cuestiones a partir de la comparación de información florística detallada geográficamente en dos cuadrados UTM de 10×10 km (región natural de la Alcarria, centro de España). Se aplica la metodología de índices de ocupación de la flora previamente ensayada en estudios corológicos, que implica diferenciar dos grupos florísticos: de visibilidad permanente y no permanente durante el ciclo anual. Se exponen y comparan los datos generales sobre frecuencia y abundancia relativa de ambos grupos, así como una muestra de la distribución de doce plantas no permanentes en cuatro resoluciones espaciales (1, 4, 25 y 100 km²). Los resultados permiten comprobar que se mejora la confiabilidad corológica al derivar los datos de la resolución mayor a las menores, pese a perder detalle geográfico; que las clasificaciones de las plantas no permanentes con mayor frecuencia y con mayor abundancia relativa son parecidas, pero difieren; y que el uso de cuadrículas de 2 km de lado resulta apropiado para representar sus distribuciones a escala local.
Article
There is ever-greater need for information about changing marine biodiversity, but such information is sparse at large spatial or temporal scales. Records about distributions of species collected by volunteers can fill gaps in knowledge that cannot yet be addressed by more structured sampling. Bayesian occupancy models show great promise for estimating trends in occurrence of species through time. This study uses the Sparta occupancy model with records from the Seasearch programme from coastal waters of Britain and Ireland during the period 2000–2020, focussing on three species of Crustacea ( Cancer pagurus , Homarus gammarus and Palinurus elephas ). Populations of P. elephas crashed in the 1970s, but now appear to be re-establishing in south-west England. The Sparta model provides evidence about recovery that is more robust than anecdotal reports or simple counts of records. Estimates of occupancy are made at different spatial scales and compared among species and areas. Trends in occupancy are compared qualitatively with patterns in fisheries landings data. Occupancy by P. elephas has increased drastically since 2014, a pattern not seen in the other two species. For each species, occupancy varied among areas and in some areas, patterns in estimates of occupancy were similar to trends in landings from fisheries. Citizen science records are increasingly recognized to have value which has not yet been fully exploited. Greater use should be made of the Seasearch dataset in order to provide population trends for benthic marine taxa. Such analyses will broaden our understanding of and taxonomic coverage of changes in biodiversity.
Article
Full-text available
An analysis of the types of uncertainties faced by resource managers is presented. Uncertainties are classified by the frequency of occurrence. Managers develop ways for dealing with frequently occurring uncertainties that do not commonly present extraordinary problems. Uncertainties that occur infrequently require an adaptive learning approach to management where we must learn about the true states of nature by careful monitoring, evaluation, and experimentation. In an undesirable situation, the ability to respond rapidly is most important. Uncertainties that occur rarely, called surprise, are very difficult to deal with. Suggested responses include holding some resources in reserve to cope with the unexpected and developing broadly based monitoring systems to detect surprises as early as possible.
Book
Long-term monitoring programs are fundamental to understanding the natural environment and effectively tackling major environmental problems. Yet they are often done very poorly and ineffectively. Effective Ecological Monitoring describes what makes successful and unsuccessful long-term monitoring programs. Short and to the point, it illustrates key aspects with case studies and examples. It is based on the collective experience of running long-term research and monitoring programs of the two authors – experience which spans more than 70 years. The book first outlines why long-term monitoring is important, then discusses why long-term monitoring programs often fail. The authors then highlight what makes good and effective monitoring. These good and bad aspects of long-term monitoring programs are further illustrated in the fourth chapter of the book. The final chapter sums up the future of long-term monitoring programs and how to make them better, more effective and better targeted.
Book
Long-term monitoring programs are fundamental to understanding the natural environment and managing major environmental problems. Yet they are often done very poorly and ineffectively. This second edition of the highly acclaimed Effective Ecological Monitoring describes what makes monitoring programs successful and how to ensure that long-term monitoring studies persist. The book has been fully revised and updated but remains concise, illustrating key aspects of effective monitoring with case studies and examples. It includes new sections comparing surveillance-based and question-based monitoring, analysing environmental observation networks, and provides examples of adaptive monitoring. Based on the authors’ 80 years of collective experience in running long-term research and monitoring programs, Effective Ecological Monitoring is a valuable resource for the natural resource management, ecological and environmental science and policy communities.
Code
R package for Data Analysis using multilevel/hierarchical model
Book
As the impacts of anthropogenic activities increase in both magnitude and extent, biodiversity is coming under increasing pressure. Scientists and policy makers are frequently hampered by a lack of information on biological systems, particularly information relating to long-term trends. Such information is crucial to developing an understanding as to how biodiversity may respond to global environmental change. Knowledge gaps make it very difficult to develop effective policies and legislation to reduce and reverse biodiversity loss. This book explores the gap between global commitments to biodiversity conservation, and local action to track biodiversity change and implement conservation action. High profile international political commitments to improve biodiversity conservation, such as the targets set by the Convention on Biological Diversity, require innovative and rapid responses from both science and policy. This multi-disciplinary perspective highlights barriers to conservation and offers novel solutions to evaluating trends in biodiversity at multiple scales.
Article
IUCN Red Lists are recognized worldwide as powerful instruments for the conservation of species. Quantitative criteria to standardize approaches for estimating population trends, geographic ranges and population sizes have been developed at global and sub-global levels. Little attention has been given to the data needed to estimate species trends and range sizes for IUCN Red List assessments. Few regions collect monitoring data in a structured way and usually only for a limited number of taxa. Therefore, opportunistic data are increasingly used for estimating trends and geographic range sizes. Trend calculations use a range of proxies: (i) monitoring sentinel populations, (ii) estimating changes in available habitat, or (iii) statistical models of change based on opportunistic records. Geographic ranges have been determined using: (i) marginal occurrences, (ii) habitat distributions, (iii) range-wide occurrences, (iv) species distribution modelling (including site-occupancy models), and (v) process-based modelling. Red List assessments differ strongly among regions (Europe, Britain and Flanders, north Belgium). Across different taxonomic groups, in European Red Lists IUCN criteria B and D resulted in the highest level of threat. In Britain, this was the case for criterion D and criterion A, while in Flanders criterion B and criterion A resulted in the highest threat level. Among taxonomic groups, however, large differences in the use of IUCN criteria were revealed. We give examples from Europe, Britain and Flemish Red List assessments using opportunistic data and give recommendations for a more uniform use of IUCN criteria among regions and among taxonomic groups.
Article
The bryophyte flora of the British Isles comprises four native hornworts, 284 liverworts and 716 mosses. These species are about two thirds of the European total. Past phytogeographical studies have concentrated particularly on Atlantic bryophytes, especially liverworts. The theory that these species can be divided into two distinct categories, one of Holarctic origin and the other of tropical and Southern Hemisphere origin, is confirmed for the British Isles. A system of elements established by us for British and Irish vascular plants is based on distribution in northern and western Eurasia, dividing the flora up first into latitudinal categories and secondly into longitudinal ones. For bryophytes, two extra elements have been added, Hyperoceanic Temperate and Hyperoceanic Southern-temperate. About 40% of the flora belongs to arctic and boreal elements and 20% to southern elements. The remaining 40% belongs to elements of the temperate broadleaved forest zone including those that extend into the boreal zone. The composition of each element is considered in terms of world distribution, habitats and distribution in the British Isles. Most species have enormous world ranges; only about 5% are endemic to Europe. Very few species are convincingly continental in that they are commoner in eastern Europe than the west. Several southern species, however, are more common in semi-arid continental interiors than in the relatively humid regions of northwest Europe. Only six liverworts and 13 mosses are known in the British Isles as established introductions; all except four originate from temperate or subtropical regions outside Europe.
Chapter
Introduction Long-term ecological monitoring is generally considered an essential tool for the effective management of biodiversity (Strayer et al. 1986, Lindenmayer and Likens 2010a). There is a large body of scientific literature describing why long-term ecological monitoring is important and describing sophisticated and useful approaches for conducting it. In this chapter, however, we ask two slightly different and somewhat controversial questions. First, how do we identify which long-term ecological monitoring programs are more worthwhile than others? There are limited resources to fund monitoring and choices must be made. Second, how do we know when we should stop monitoring, if ever? So few examples of successful long-term monitoring exist that the idea of stopping monitoring sounds like heresy, but if the benefits are diminishing and the effort could be better used elsewhere, why not? To answer these questions we need to be able to classify and quantify the benefits of long-term monitoring and hence we pull together many of the ideas on this issue from the literature and other chapters in this book. Monitoring has well-known benefits and is likely to play an increasingly important role as we try to determine how large-scale anthropogenic changes, in the context of a rapidly changing climate, alter the way in which we should manage ecosystems (Balmford et al. 2005, Field et al. 2007). Effective monitoring is a critical link in the cycle of adaptive management that aims to iteratively improve conservation actions over time (Holling 1978, Walters 1986) and is important for influencing government policy and investment in environmental programs (GAO 2006, OECD 2006).