The global Land-Potential Knowledge System (LandPKS): Supporting evidence-based, site-specific land use and management through cloud computing, mobile applications, and crowdsourcing

Full text

5A
JAN/FEB 2013—VOL. 68, NO. 1
JOURNAL OF SOIL AND WATER CONSERVATION
Jeffrey E. Herrick , Kevin C. Urama, Jason W. Karl, John Boos, Mari-Vaughn V. Johnson, Keith D. Shepherd, Jon
Hempel, Brandon T. Bestelmeyer, Jonathan Davies, Jorge Larson Guerra, Chris Kosnik, David W. Kimiti, Abra-
ham Losinyen Ekai, Kit Muller, Lee Noreet, Nicholas Ozor, Thomas Reinsch, José Sarukhan, and Larry T. West
The global Land-Potential Knowledge System (LandPKS):
Supporting evidence-based, site-specic land use and management
through cloud computing, mobile applications, and crowdsourcing
doi:10.2489/jswc.68.1.5A
Jeffrey E. Herrick is Research Soil Scientist and
Brandon T. Bestelmeyer and Jason W. Karl are
Research Ecologists at the Jornada Experimental
Range, USDA Agricultural Research Service,
Las Cruces, New Mexico. Kevin C. Urama is
Executive Director and Nicholas Ozor is Senior
Research Ofcer at the African Technology
Studies Network, Nairobi, Kenya. John Boos is
Geospatial Advisor and Chris Kosnik is Team
Leader at the US Agency for International
Development, Washington, DC. Mari-Vaughn
V. Johnson is Agronomist and Lee Noreet is
Modeling Team Leader/Soil Scientist at the
USDA Natural Resource Conservation Service,
Temple, Texas. Keith D. Shepherd is Principal
Soil Scientist at the World Agroforestry Center
(ICRAF), Nairobi, Kenya. Jon Hempel is Director,
and Larry T. West is Soil Survey Research and
Laboratory National Leader at the National Soil
Survey Center, Lincoln, Nebraska. Jonathan
Davies is Coordinator at the IUCN Global
Drylands Initiative, Nairobi, Kenya. Jorge Larson
Guerra is Use of Biodiversity Coordinator and
José Sarukhan is National Coordinator at
National Commission for Knowledge and Use
of Biodiversity (CONABIO), Mexico City, Mexico.
David W. Kimiti is Graduate Research Assistant
at the New Mexico State University, Las
Cruces, New Mexico. Abraham Losinyen Ekai
is Ford Foundation Scholar at Duke University,
Durham, North Carolina. Kit Muller is Strategic
Planner at the US Department of the Interior
Bureau of Land Management, Washington, DC.
Thomas Reinsch is National Leader for World
Soil Resources at the USDA Natural Resource
Conservation Service, Beltsville, Maryland.
IDEAS &
INNOVATIONS
A
gricultural production must
increase significantly to meet
the needs of a growing global
population with increasing per capita con-
sumption of food, fiber, building materials,
and fuel. Consumption already exceeds
net primary production in many parts of
the world (Imhoff et al. 2004).
In addition to reducing consumption,
there are two options to meet these needs:
production intensification and land con-
version. Both strategies present unique
opportunities, challenges, and risks. The
largest gains achievable through agricul-
tural intensification will likely occur on
lands with the largest unrealized produc-
tion potential, or yield gap. These lands
have high potential production and low
current production. Similarly, the highest
returns on investments to be gained by land
conversion should occur on lands with the
highest potential production, assuming
similar infrastructure, per acre conversion
costs, and other market conditions.
The biggest long-term risk for both
strategies is that application of nonsustain-
able land management practices will result
in soil degradation that is often costly, if
not impossible, to reverse. Exploiting these
opportunities and minimizing risks depend
on careful matching of production systems
with the sustainable production potential
of each type of land. Similar analyses can
be applied to biodiversity conservation to
prioritize land conservation and restora-
tion efforts.
The ability to match land use with land
potential is limited by four factors: (1)
current land potential evaluation systems,
while addressing potential productivity
and degradation resistance, do not con-
sider resilience, (2) it is virtually impossible
to identify, access, and interpret all rel-
evant scientific and local knowledge and
information, (3) this information is often
provided in the form of maps at a scale that
is far too coarse for field-scale management,
and (4) by the time land classification sys-
tems are established and maps developed,
the information is often obsolete. Farmers
and scientists are constantly innovating and
adapting, effectively changing land poten-
tial to support different types of production.
This paper describes how a new
cloud-based Land-Potential Knowledge
System (LandPKS; www.landpotential.
org; figure 1) will allow land potential
to be defined explicitly and dynamically
for unique and constantly changing soil
and climate conditions and to be updated
based on new evidence about the success
or failure of new management systems
on different soils. The knowledge engine
(figure 2), together with simple applica-
tions for mobile phones, will also facilitate
more rapid and complete integration and
dissemination of local and scientific knowl-
edge about sustainable land management.
Figure 1
Land-Potential Knowledge System. See landpotential.org for more information and
opportunities to participate.
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation

6A
JOURNAL OF SOIL AND WATER CONSERVATION
JAN/FEB 2013—VOL. 68, NO. 1
This system expands the concept of an
Ecological Knowledge System (Herrick
and Sarukhan 2007) through the use of
mobile technologies and cloud computing
and making more extensive use of crowd-
sourcing both knowledge and information
(Karl and Herrick forthcoming).
The system will include both simple,
mobile phone interfaces and more sophis-
ticated web tools that can be accessed via
personal computers and linked with other
decision tools. Individual producers will be
able to use the system to answer questions
about sustainable land management options
at the field scale, while policymakers will
be able to aggregate data across larger areas
without losing key pieces of information,
such as the presence of small, highly produc-
tive, biodiverse, or vulnerable sites within a
region. It will also provide extension work-
ers with the ability to instantaneously access
the best available information and interpret
it in the context of local socioeconomic
conditions and local values, including crop
preferences, while scientists will have access
to a global georeferenced database for cali-
brating remote sensing imagery and testing
hypotheses globally. Finally, as a social net-
working tool, it will allow individual
producers to easily connect with others
facing similar challenges on similar types
of land.
LAND POTENTIAL
Overview. Land potential includes three
elements: potential production of one
or more ecosystem services, degradation
resistance, and resilience, or the capacity
to recover following degradation. While
some definitions of resilience integrate
both degradation resistance and capacity
to recover, we distinguish them because
different soil, vegetation, and landscape
properties and processes affect resistance
and resilience to different disturbance
types in different ways (figure 3) (Seybold
et al. 1999). For example, a flat, shallow,
loamy soil may be relatively resistant but
not resilient to water erosion. Conversely,
the same soil may have low resistance to
compaction, but recover relatively quickly
and completely (high resilience). Similarly,
sustainable cultivation of steep slopes tends
to be limited by low soil erosion resistance,
while recovery following soil erosion also
Figure 2
Knowledge engine for the Land-Potential Knowledge System.
Figure 3
Changes in productivity in response to an acute disturbance, such as an extreme storm
event on a freshly plowed field, demonstrate the difference in long-term productivity
potentials for resilient and nonresilient land. Resistant land loses little soil in response
to the disturbance. Potential productivity on land that is resilient but not resistant will
be impacted by the disturbance, but will quickly recover and regain the previous poten-
tial productivity levels. Land that is resilient due to its recovery capacity may have
relatively deep soils that change little with depth. Land that is nonresilient in response
to extreme storm events following tillage often has shallow soils, or soils in which the
lower horizons contain more clay than the surface, resulting in reduced infiltration
(adapted from Seybold et al. 1999).
Potential productivity, including variability due to weather
Legend
Time
Productivity
Resistant to degradation
Resilient
Neither resistant nor resilient
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation

7A
JAN/FEB 2013—VOL. 68, NO. 1
JOURNAL OF SOIL AND WATER CONSERVATION
tends to be limited on otherwise produc-
tive shallow soils (figure 4).
A general definition of land potential
that combines these three elements can
be adapted from the definition of sus-
tainability provided in the report “Our
Common Future:” land potential is the
capacity of land to support ecosystem ser-
vices required to meet “the needs of the
present without compromising the ability
of future generations to meet their own
needs” (Brundtland Commission 1987).
Land potential can also be defined in terms
of the capacity of land to support more
specific land use objectives, including its
potential to provide the resources neces-
sary for one or more species to complete
their life cycles and reproduce. The value
of applying the land potential concept to
biodiversity conservation is that it allows
the potential future range of species to be
predicted based on habitat requirements,
rather than relying solely on historic or
existing plant and animal community pat-
terns. This is particularly important where
climate change and invasive species mod-
ify the conditions necessary for species of
interest to survive and reproduce. The fact
that there are so many different land use
objectives means that it can be difficult to
generate land potential evaluations that
address all needs (FAO 2007).
For most purposes, however, land poten-
tial can be evaluated based on knowledge of
basic soil profile characteristics, topography,
and climate. For example, a nonsaline, deep,
well-drained, medium-textured soil with a
slope of less than 2% in a 1,000 mm (39.4
in) summer-dominated precipitation zone
clearly has greater potential to sustainably
support a wide range of ecosystem services
than a steep, shallow saline soil receiving
200 mm (7.9 in) of rain per year. Soil tex-
ture and depth largely determine soil water
and nutrient supplying capacity. Erosion
risk for bare soil can be predicted with the
inclusion of topographic attributes eas-
ily derived from digital elevation models.
Additional information may be required
in regions where the soil parent mate-
rial, age, or hydrology result in unique soil
characteristics, including clay mineralogy,
unusually high or low pH, and high salinity
and sodicity. Feedbacks between vegetation,
soil, and climate, including both hydrology
and nutrient dynamics, are also impor-
tant, particularly over longer time periods
(Peters et al. 2004).
Spatial Variability. Even across areas
that share common rainfall and tempera-
ture patterns, production potentials may
differ considerably, with some areas having
shallow, highly eroded soils and other areas
characterized by deep, relatively fertile soils
that hold water long into droughts. Land
degradation risk and recovery potential also
vary widely: some soils recover quickly fol-
lowing tillage or overgrazing, while others
may require centuries or millennia.
It is more widely recognized that land
potential also varies at global and regional
scales. In Antarctica, climate limits produc-
tion to near zero and resilience is limited by
both low resistance to soil erosion (Tejedo
et al. 2009) and low recovery potential due
to both climate and shallow soil depths.
Conversely, all other continents have both
regions of low land potential and limited
resilience and regions with extremely high
levels of productivity, degradation resis-
tance, and resilience.
Change Over Time. Global circulation
models predict that the currently observed
spatial variability of land potential will be
compounded by increased climate vari-
ability through the 21st century, creating
even more heterogeneity.
CURRENT APPROACHES TO LAND
POTENTIAL EVALUATION
A number of different evaluation sys-
tems have been developed and applied
around the world in attempts to improve
our understanding and management
of land potential. An extensive review
and analysis of these systems is forth-
coming (United Nations Environment
Programme International Resource Panel,
http://www.unep.org/resourcepanel/).
Two that have been widely applied glob-
ally are the USDA’s Land Capability
Classification (LCC) system (Klingebiel
and Montgomery 1961) and the Food and
Agriculture Organization’s Agroecological
Zoning (AEZ) system (FAO 2007). The
Land Capability Classification was devel-
oped in the 1950s and focused on general
biophysical limitations to sustainable crop
production. The AEZ, developed within
the Framework for Land Evaluation
(FAO 1976), is a more holistic approach
Figure 4
Simplified, generalized patterns of potential production (P) and resistance and resil-
ience (R) based on climate, resistance to erosion, soil depth, soil texture, and potential
for soil organic matter accumulation and soil structure development (under natural con-
ditions for water-limited regions). The Land-Potential Knowledge System will improve
and localize these predictions to individual fields.
Flat, shallow
sandy soils (low
water storage)
Flat, loamy
soils (high
water storage)
Steep loamy soils
(high water stor-
age and water
erosion risk)
Flat loamy soils
(high water
storage)
Steep loamy
soils (high
water storage
and water
erosion risk)
Low rainfall
Potential production and resilience
Moderate rainfall
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation

8A
JOURNAL OF SOIL AND WATER CONSERVATION
JAN/FEB 2013—VOL. 68, NO. 1
that addresses both the biophysical and
socioeconomic factors that may affect pro-
duction of a particular crop in a specific
area. Both the LCC and AEZ effectively
regard crop production as the highest and
best use of land. They reflect differences in
degradation resistance but not the capacity
of the land to recover following degrada-
tion (resilience).
The Food and Agriculture
Organization’s Framework for Land
Evaluation was recently updated in rec-
ognition of the need to address climate
change, biodiversity, and desertification,
incorporate new technologies for land
evaluation, and address the benefits of par-
ticipatory approaches (FAO 2007). This
revision is based on a set of eight principles
(FAO 2007):
1. Land suitability should be assessed and
classified with respect to specified kinds
of land use and services.
2. Land evaluation requires a compari-
son of benefits obtained and the inputs
needed on different types of land to
assess the productive potential, envi-
ronmental services, and sustainable
livelihood.
3. Land evaluation requires a multidisci-
plinary and cross-sectoral approach.
4. Land evaluation should take into
account the biophysical, economic,
social, and political contexts, as well as
the environmental concerns.
5. Suitability refers to use of services on
a sustained basis; sustainability should
incorporate productivity, social equity,
and environmental concerns.
6. Land evaluation involves a comparison
of more than one kind of use of service.
7. Land evaluation needs to consider all
stakeholders.
8. The scale and the level of decision
making should be clearly defined prior
to the land evaluation process.
The LandPKS will ultimately support
the application of each of these prin-
ciples for land evaluation at farm field to
national scales. The LandPKS will facilitate
the integration and application of local
and scientific information and knowledge
(Herrick et al. 2010), including existing
land evaluation systems through the adop-
tion of crowdsourcing, mobile phone, and
innovative decision support system tech-
nologies (Karl and Herrick forthcoming).
LAND-POTENTIAL KNOWLEDGE SYSTEM
DEVELOPMENT AND APPLICATION
Overview. Development of the LandPKS
takes advantage of recent advances in
cloud computing, digital soil mapping,
Global Positioning System–enabled cam-
era phones, and mobile applications. These
technologies allow knowledge and infor-
mation about land potential to be gathered,
integrated, and shared globally, while global
databases and generic models make exist-
ing knowledge more accessible and allow
similar sites to be more easily matched (fig-
ure 5) (Bestelmeyer et al. 2011; Herrick
and Sarukhan 2007). Global Positioning
System–enabled mobile phones can be used
to capture and transmit geolocated photo-
graphs of soil, land use, and erosion features.
Applications can be used to record
additional information about a site using
drop-down menus, text-input, and pic-
ture matching. Slope and color can
also be determined with many phones,
and the USDA (Agricultural Research
Service–Jornada and Natural Resources
Conservation Service–Lincoln), in col-
laboration with World Agroforestry Centre
(Shepherd and Walsh 2007), is currently
developing an application that will increase
the quality of color determinations using
an integrated calibration system. Cloud
computing allows completion of relatively
sophisticated analyses requiring access to
large databases, while innovative analysis
approaches allow different types of data
with currently unspecified error rates to
be integrated (e.g., Hubbard 2010). These
approaches facilitate integration of local
knowledge with, for example, the increas-
ing amounts of information being made
available by the Global Soil Map con-
sortium. This initiative was established in
response to the rapidly increasing demand
for soil information (Sanchez et al. 2009).
The consortium includes leading global
soil science institutions and is committed to
producing digital soil maps that will predict
important soil properties at a fine resolution
using state-of-the-art and emerging tech-
nologies for soil mapping (figure 6).
These and other tools and tech-
nologies allow local knowledge to be
crowdsourced and different sources of
knowledge and information to be cross-
referenced, enabling the LandPKS to
generate site-specific interpretations about
land potential which can be immediately
shared with others, including farmers and
scientists, who may have additional insights.
The LandPKS extends earlier efforts to
integrate local and scientific informa-
tion (Barrios et al. 2006; Herrick et al.
2010) by allowing site-specific conclu-
sions to be instantaneously updated based
on input from other locations with similar
soil and climate characteristics. As a result,
the accuracy, precision, and relevance of
the knowledge engine at the core of the
LandPKS will increase with each use.
Phased Development. The LandPKS
is being developed and implemented
through a phased, modular approach (table
1) designed to complement, rather than
replace, new and existing land evaluation,
database development, and soil mapping
(a)
Figure 5
Similar soil profiles occurring in areas with similar climate in the (a) southwest United
States and (b) northwest Namibia. Both are shallow loamy sands on low gradients
underlain by a partially rubblized petrocalcic horizon (inset).
(b)
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation

9A
JAN/FEB 2013—VOL. 68, NO. 1
JOURNAL OF SOIL AND WATER CONSERVATION
initiatives, as well as government extension
efforts and local and international devel-
opment projects. Characteristics unique
to LandPKS include the ability to provide
site-specific information based on simple
soil descriptions, to effectively integrate
local and scientific knowledge through
expert systems that assess multiple sources
of qualitative and quantitative knowl-
edge and information, and to provide
an interactive self-learning platform that
simultaneously collects and shares knowl-
edge and information among a broad
range of users.
Using a phased, modular approach
will (a) allow stakeholders to apply early
versions of the system to make basic
determinations about land potential; (b)
maximize opportunities for them to con-
tribute knowledge and information to
the system, including providing instant
feedback on initial determinations; and
(c) ensure that the system is sufficiently
flexible and dynamic to take advantage
of and contribute to future tools, tech-
nologies, and information and knowledge
sources. In particular, we are encouraged
by current progress and future plans of,
among others, the Food and Agriculture
Organization; Global Soil Map; African
Soil Information System; European
Environment Agency, including Eye on
Earth; and several Consultative Group
on International Agricultural Research
centers; as well as a number of sustain-
able land management knowledge systems
such as World Overview of Conservation
Approaches and Technologies (WOCAT).
Finally, we believe that the growth of the
semantic web (Villa 2007) and new tools
that essentially automate mobile appli-
cation development will significantly
increase the functionality of the system,
while reducing maintenance costs.
Leadership and Participation.
Development of the LandPKS is being
led by the USDA together with a large
and growing group of partners. Funding
from the US Agency for International
Development is supporting initial develop-
ment and pilot implementation in Africa,
and the Africa Technology Policy Studies
Network is providing leadership and coor-
dination throughout the continent. An
open source, open participation format is
being applied to both development and
implementation. In order to facilitate broad
participation, the amount of additional
information provided by the user will be
tiered based on user information needs,
knowledge, input device (mobile phone vs.
computer keyboard), time availability, and
technical capacity (figure 7).
Figure 6
An example of a first generation digital
soil map derived from existing soil map
data (Soil Survey Geographic Database
and US General Soil Map) using spatially
map unit component weighted means
calculation: soil organic carbon in the sur-
face 5cm (Bliss et al. 2009). Red indicates
low and blue indicates high soil organic
carbon values.
Table 1
Phased approach to the development and implementation of the Land-Potential
Knowledge System. This approach will continue to evolve based on input from a
rapidly growing group of partners.
Phase I (2012/13 to 2015) Phase II (2014/15 to 2016) Phase III (2015 to 2018)
Information sources Existing online Existing online Existing online
databases databases + pilot user- databases + user-
  contributedeld contributedeld
observations observations
Knowledgesources Publishedscientic PhaseI+pilotuser- PhaseIIfull
sources contributed local implementation
  knowledge andrenement
Knowledge system Simple decision Phase I + pilot expert Phase II full
(mechanism) support based on system iteratively implementation and
 basicusersoil/site integratinguser renement
 descriptionandexisting responsestosite-specic
land evaluation systems queries designed to
increase accuracy and
relevance of output
Knowledgesystem Basicbiophysicalland PhaseI+identication PhaseII+identication
(output) potentialforsustainable ofspecicSLMpractices ofsuccessfulSLM
agricultural production with emphasis on options from other
(suitability for grazing, practices already in regions being applied
crop production, and use locally on similar soils that are
 identicationofgeneral  potentiallyrelevantin
 typesofSLMpractices)* localsocioeconomic
context
Connections with Develop system so that Initiate connections and Ensure that
other mobile phone/ it is open and has develop capability for independent
Web-based services† capacity to connect and other organizations to use connection with
 identifypotential LandPKSasaplatform LandPKSissupported
connections. for their own products
Implementation Focused on USDA–US Available for pilot Full implementation
Agency for International application by other
Development pilot areas organizations
*Theseorganizationswillbenetfromsystem’saggregationofknowledgeandinformationpotentially
relevanttothespecicsiteandsoildescriptionprovidedbytheobserverthroughthemobileapplication.
†Forexample,thoseprovidingmorespecicinformation(e.g.,cropprices,inputrecommendations).
Note:SLM=sustainablelandmanagement.
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation

10A
JOURNAL OF SOIL AND WATER CONSERVATION
JAN/FEB 2013—VOL. 68, NO. 1
FUNCTIONS AND BENEFITS OF THE
LAND-POTENTIAL KNOWLEDGE SYSTEM
Overview. The LandPKS is very much
a means to multiple ends, rather than an
end in itself. During an extended infor-
mal scoping process in 2012, we reviewed
the project objectives and strategy with
diverse stakeholders, including pastoralists
and farmers in Turkana, Kenya (Losinyen
2012), and northern Namibia; leadership
and technical staff of the USDA Natural
Resources Conservation Service and US
Department of the Interior Bureau of
Land Management; global scientists; and
participants in two African Technology
Policy Studies Network meetings, includ-
ing representatives of a broad range of
government ministries and the African
Union. Based on these conversations, we
identified the following functions: connect
producers with each other; directly support
land management decisions by farm-
ers, ranchers, and pastoralists, including
through extension; inform land planning
by governments and investments in land
management by governments, nongovern-
mental, and overseas development assistant
organizations; improve other decision sup-
port systems and geographic information
system products; and promote innovation
(figure 1).
We recognize that it is impossible to
optimize the LandPKS for all of these
functions. We are committed to ensuring
that other developers can easily connect
with and leverage the LandPKS to create
products that more effectively address one
or more of these functions locally, nation-
ally, or globally.
Connect Producers with Each Other.
The local knowledge database developed
as part of the LandPKS will include all of
the elements necessary to support social
networking. It will allow individual pro-
ducers to easily connect with others facing
similar challenges on similar types of land.
This is particularly important in countries
where extension services are limited. It can
also be used to support extension activi-
ties by allowing facilitators working on
farmer-to-farmer exchanges identify pro-
ducers who are most likely to be able to
benefit from each other.
Support Land Management Decisions.
Integrating existing knowledge and infor-
mation enables fine-scale interpretation
of available knowledge and information
at any location where a user provides
descriptive soils information (figure 7).
This is important because land potential
often varies more at finer scales than it is
possible to map. Accuracy of pixel-level
predictions based on digital soil maps,
such as those being developed by the
Africa Soil Information System, will be
further improved by user-provided pho-
tos, responses to simple questions about
soil texture and depth, land use and cover,
and new diagnostic soil color applications
and protocols for mobile phones. We are
working to leverage the extensive work
that Africa Soil Information System has
done to develop more sophisticated rela-
tionships based on near- and mid-infrared
spectroscopy (Shepherd and Walsh 2007).
Both Africa Soil Information System and
USDA Natural Resources Conservation
Service databases include visible bands that
can be acquired using cameras included
with many mobile devices.
Inform Land Use Planning, Land
Tenure, and Targeted Investments
Designed to Sustainably Increase
Production. In addition to the obvi-
ous benefits for land use planning, the
LandPKS can, as desired by individual
countries, be used to help ensure that
land tenure reform programs result in an
equitable distribution of land based on
its potential, rather than simply distribut-
ing areas of equal size. LandPKS can also
help governments, funders, and nongov-
ernmental organizations identify specific
locations within each region where invest-
ments in specific projects, such as technical
support and drought assistance, are likely
to have the greatest long-term impact and
return on investment, while helping to
select the interventions that are likely to
have the greatest impact.
Improve Other Decision Support
Systems and Geographic Information
System Products. One of the most fre-
quently cited future benefit of the
LandPKS by scientists is improving soil,
Figure 7
Illustration of potential use of the LandPKS to guide land use management decisions in
an area of northern Namibia where grasslands, savannas, and woodlands are being con-
verted to annual crop production. User uploads geolocated soil photos and local knowl-
edge (1); which is then integrated with global knowledge and information and sent back
to the user as coded land management options, together with a request for additional
information (2); which is again entered on the mobile phone (3); resulting in a refined
suite of coded management options (4). The number of iterations and complexity of user
input will be flexible, depending on user technical capacity and time availability.
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation

11A
JAN/FEB 2013—VOL. 68, NO. 1
JOURNAL OF SOIL AND WATER CONSERVATION
land use, and land cover maps by provid-
ing consistent data, including photographs,
at potentially millions of points globally.
User inputs including soil, vegetation,
land use, and land use history informa-
tion will be made available, subject to any
country-specific limitations, in an online
geo-database.
In addition to using the raw data pro-
vided by users, the LandPKS is being
designed to allow other land management–
related applications and other software to
make more soil-specific recommendations
about, for example, improved crop variet-
ies, fertilizer forms and rates, and specific
management interventions. The number
of mobile applications supporting one
or more aspects of land use management
and conservation planning is increas-
ing rapidly, but few explicitly address the
critical importance of soils and, more
generally, land potential, for localizing
recommendations.
One particularly intriguing initia-
tive that is already explicitly integrating
soil information is the development of
innovative models through the informa-
tion systems and strategic research of the
Consultative Group on International
Agricultural Research Program on Water,
Land and Ecosystems using Applied
Information Economics (figure 8)
(Hubbard 2010; Shepherd and Hubbard
2012). Land and water intervention deci-
sions are being modelled in a way that
captures the uncertainties in costs and
impacts on agro-ecosystem productivity,
environment and human well-being, and
trade-off preferences of stakeholders. This
probabilistic modelling approach identifies
and quantifies the value of information
and further research for improving inter-
vention decisions and provides a holistic
business case for intervention programs
that includes environmental sustainabil-
ity and equity concerns (Shepherd and
Hubbard 2012).
Promote Innovation. Perhaps the
most exciting function of the LandPKS
is promoting innovation. Agriculture
has continued to largely rely on the
research-demonstration-extension model
of innovation. The LandPKS can acceler-
ate innovation in at least three ways. The
first is by reducing the failure frequency in
the implementation of new technologies
by reducing the probability that the tech-
nologies will be applied on land that does
not have the potential to respond. The
second is by allowing for virtual instanta-
neous sharing of successes and failures to
all producers with similar land potential.
Scientists, too, can benefit by replacing
preliminary trials with a global search of
all producers who have already tried a
similar approach on similar soils and can
use the same process to validate conclu-
sions across diverse soils. While there are
clearly a number of risks associated with
this crowdsourcing approach, we believe
that they can be minimized by cross-ref-
erencing different types of knowledge and
information, including producer obser-
vations, measurements, photographs, and
physical models. This general approach
to reducing the uncertainty of individual
conclusions has been widely applied in the
business sector (Hubbard 2010).
SOCIOECONOMIC FACTORS: A NOTE
We fully recognize that in many cases the
ability to sustainably increase agricultural
production and biodiversity conservation
and to address other objectives is not lim-
ited by knowledge and information about
biophysical land potential. Access to mar-
kets, prices for agricultural products and
inputs, social and political instability, and
local food preferences frequently con-
strain land use decisions. We acknowledge
this to be true in many and perhaps most
cases. However, we also believe that while
not sufficient, an understanding of land
potential is necessary to select the most
sustainably productive land use or uses
and management systems, and that this
understanding can also be used to select
from a range of options that is already
limited by nonbiophysical factors. By
starting with the biophysical potential of
the land, we provide a foundation for inte-
grating other factors, such as crop prices.
Socioeconomic factors will be integrated
as LandPKS evolves, either by linking to
other programs, or as an integral part of
the system itself.
Finally, by facilitating global con-
nections, the LandPKS increases the
probability that innovative solutions will
Figure 8
An example of how the LandPKS is being designed to both inform and be informed by
other related decision support systems. In this illustration, probabilistic modelling is
applied to the development and selection of land management interventions. The model
will determine the uncertainty of onsite and offsite impacts of interventions, as well
as behavioral factors like the adoption rate of a new practice or how incentives change
behavior. Ultimately, the effects of an intervention and the quantified preferences are
combined into a single monetized value so that interventions of different types and
sizes can be compared (Shepherd and Hubbard 2012).
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation

12A
JOURNAL OF SOIL AND WATER CONSERVATION
JAN/FEB 2013—VOL. 68, NO. 1
be developed and communicated more
rapidly and systematically. For example, an
innovative approach to rehabili tating land-
scapes dissected by gullies developed in
one country can be instantaneously shared
with producers in other countries, who
can then test it under local conditions.
SUMMARY AND CONCLUSIONS
The rapid expansion of internet acces-
sibility through mobile phone networks
together with simple mobile applications
and expert knowledge systems provide
new opportunities to connect farmers,
extension and development workers, and
policymakers with site-specific knowledge
and information. The amount of electroni-
cally available knowledge and information
about land potential, including resilience, is
also rapidly increasing through the efforts
of a number of organizations throughout
the world. The proposed Land-Potential
Knowledge System will leverage these
emerging trends to connect land managers
committed to sustainable land management
with the most relevant and up-to-date
knowledge and information available.
ACKNOWLEDGEMENTS
We thank all of the individuals who participated in the
LandPKS scoping discussions during the past year. In
addition to the coauthors listed here and a large body
of literature (only a small portion of which is cited),
a large number of individuals have also contributed
to conversations that led to the development of the
approach, including but by no means limited to Fee
Busby (Professor, Utah State University, Logan, Utah),
Joel Brown (Scientist, Natural Resource Conservation
Service, Las Cruces, New Mexico), Ericha Courtright
(Research Assistant, USDA-Agricultural Research
Service, Las Cruces, New Mexico), Kris Havstad
(Lead Scientist, USDA-Agricultural Research Service,
Las Cruces, New Mexico), Elisabeth Huber-Sannwald
(Professor, IPICyT, San Luis Potosi, Mexico), Mike
Pellant (Great Basin Restoration Initiative Coordinator,
Bureau of Land Management, Boise, Idaho), Debra
Peters (Research Scientist, USDA-Agricultural
Research Service, Las Cruces, New Mexico) , David
Pyke (Research Ecologist, US Geological Survey,
Corvallis, Oregon), James Reynolds (Professor, Duke
University, Durham, North Carolina), Pat Shaver
(Rangeland Management Specialist, Corvallis,
Oregon), Chloe Stull-Lane (Program Quality Manager
– African Drylands, Mercy Corps, Nairobi, Kenya),
Dennis Thompson (National Rangeland and Pasture
Lead, USDA Natural Resource Conservation Service,
Washington, DC), Jason Taylor (Monitoring National
Implementation Lead, Bureau of Land Management,
Denver, Colorado), Gordon Toevs (National
Monitoring Lead, Bureau of Land Management,
Washington, DC), Justin van Zee (Research Assistant,
USDA-Agricultural REsesearch Service, Las Cruces,
New Mexico), Nick Webb (Post-doctoral Research
Scientist, USDA Agricultural Research Service, Las
Cruces, New Mexico) and Skye Wills (Soil Scientist,
Natural Resource Service, Lincoln, Nebraska). Pat
Shaver (Rangeland Management Specialist, USDA
Natural Resources Conservation Service, Corvallis,
Oregon), Corinna Riginos (Post-doctoral Research
Associate, University of Wyoming, Laramie, Wyoming),
and Carey Farley (Team Leader, CARE International
funded by US Agency for International Development,
Nairobi, Kenya) were instrumental in beta testing
some of the rapid protocols that will be implemented
as part of the approach, as were Uhangatenua Kapi,
Colin Nott, Matthias Metz, Cornelis van der Waal,
and the Agra field crew (all associated with and sup-
ported by the United States Millennium Challenge
Corporation and the Namibia Millennium Challenge
Account, Windhoek, Namibia).
REFERENCES
Barrios, E., R.J. Delve, M. Bekunda, J. Mowo, J.
Agunda, J. Ramisch, M.T. Trejo, and R.J. Thomas.
2006. Indicators of soil quality: A south–south
development of a methodological guide for link-
ing local and technical knowledge. Geoderma
135:248-259.
Bestelmeyer, B.T., H.R. Peinetti, J. Herrick, D.
Steinaker, E. Adema, and K. Havstad. 2011.
State-and-transition model archetypes: a global
taxonomy of rangeland change. IX International,
Rangeland Congress, Rosario, Argentina.
Bliss, N.B., S. W. Waltman, and L. West. 2009. Detailed
Mapping of Soil Organic Carbon Stocks in the
United States Using SSURGO. Abstract B51F-
0367. American Geophysical Union Meeting,
Fall 2009.
Brundtland Commission. 1987. Our Common
Future. New York, NY: Oxford University Press.
FAO (Food and Agriculture Organization). 1976.
A framework for Land Evaluation. FAO Soils
Bulletin 52. Rome, Italy: Food and Agriculture
Organization.
FAO. 2007. Land Evaluation: Towards a Revised
Framework. Land and Water Discussion Paper 6.
Rome, Italy: Food and Agriculture Organization.
Herrick, J.E., and J. Sarukhan. 2007. A strategy for
ecology in an era of globalization. Frontiers in
Ecology and the Environment 5:172-181.
Herrick, J.E., V.E. Lessard, K.E. Spaeth, P.L. Shaver, R.S.
Dayton, D.A. Pyke, L. Jolley, and J.J. Goebel. 2010.
National ecosystem assessments supported by local
and scientific knowledge. Frontiers in Ecology and
the Environment 8:403-408.
Hubbard, D.W. 2010. How to Measure Anything.
Hoboken, NJ: John Wiley & Sons.
Imhoff, M.L., L. Bounoua, T. Ricketts, C. Loucks, R.
Harriss, and W.T. Lawrence. 2004. Global patterns
in consumption of net primary productivity. Nature
429:870-873.
Karl, J., and J.E. Herrick. Forthcoming. A range-
land Wikicology? Implementing collaborative
Internet technologies for rangeland management.
Rangelands.
Klingebiel, A.A., and P.H. Montgomery. 1961. Land
Capability Classification. Agricultural Handbook
No. 210. Washington, DC: USDA.
Losinyen, A. 2012. Land Potential Knowledge System
Scoping Exercise Report. Unpublished report. Las
Cruces, NM: USDA Agricultural Research Service
Jornada Experimental Range.
Peters, D.P.C., R.A. Pielke, B.T. Bestelmeyer, C.D.
Allen, S. Munson-McGee, and K.M. Havstad.
2004. Cross-scale interactions, nonlinearities, and
forecasting catastrophic events. Proceedings of the
National Academy of Sciences of the United States
of America 101:15130–15135.
Sanchez, P.A., S. Ahamed, F. Carre, A.E. Hartemink, J.
Hempel, J. Huising, P. Lagacherie, A.B. McBratney,
N.J. McKenzie, M.L. de Mendonca-Santos, B.
Minasny, L. Montanarella, P. Okoth, C.A. Palm,
J.D. Sachs, K.D. Shepherd, T-G. Vågen, B. Vanlauwe,
M.G. Walsh, L.A. Winowieck, and G-L. Zhang.
2009. Digital soil map of the world. Science
325(5941):680-681.
Seybold, C.A., J.E. Herrick, and J.J. Brejda. 1999. Soil
resilience: A fundamental component of soil quality.
Soil Science 164:224-234.
Shepherd, K.D., and D.W. Hubbard. 2012. The need
for an intervention decision model. Concept Paper.
World Agroforestry Centre (ICRAF) and Hubbard
Decision Research, Nairobi/Chicago.
Shepherd, K.D., and M.G. Walsh. 2007. Infrared
spectroscopy—enabling an evidence-based diag-
nostic surveillance approach to agricultural and
environmental management in developing coun-
tries. Journal of Near Infrared Spectroscopy 15:1-19.
Tejedo, P., A. Justel, J. Benayas, E. Rico, P. Convey, and
A. Quesada. 2009. Soil trampling in an Antarctic
Specially Protected Area: Tools to assess levels of
human impact. Antarctic Science 3:229-236.
Villa, F. 2007. A semantic framework and software design
to enable the transparent integration, reorganization
and discovery of natural systems knowledge. Journal
of Intelligent Information Systems 29:79-96.
Copyright © 2013 Soil and Water Conservation Society. All rights reserved.
www.swcs.org 68(1):5A-12A Journal of Soil and Water Conservation