Informing climate-smart agriculture in low resource settings for practitioners: A review and analysis of interactive tools

Introduction

Small-scale agricultural producers are at ground zero for experiencing the impacts of climate change. Their livelihoods are based on producing crops that are being directly impacted by climate change, which is expected to continue to follow new and increasingly unpredictable patterns in the coming decades. As temperatures rise, precipitation becomes more variable, and extreme weather events become more frequent, small-scale producers in tropical regions in particular are expected to experience the impacts of these stressors most acutely (Aragón et al., 2021; Morton, 2007). Small-scale producers are considered especially vulnerable as they have generally low incomes and rely predominantly on agriculture for their livelihoods (Rosenzweig & Hillel, 2008).

In addition to being a uniquely vulnerable sector, agriculture is uniquely suited to concurrently advancing both mitigation and resilience goals (Cohn et al., 2017). Indeed, the agricultural sector remains the largest economic sector and employer in low-income countries and is often the largest source of emissions (FAO, 2012). Due to this important role, even though countries are not required specifically report on agriculture in their Nationally Determined Contributions (NDCs), 90% of NDCs mention agriculture and 41% mention food security, and it is a priority sector for climate action (Schulte et al., 2020). At the United Nations 28th annual Conference of the Parties of the UNFCCC in 2023 (commonly known as COP28), the Declaration on Sustainable Agriculture, Resilient Food Systems, and Climate Action recognizes these multiple roles and was signed by 159 countries.

In light of these needs, the international community has called for ramping up climate financing to help agricultural producers to confront climate change mitigation and adaptation, but so far, the funding is a tiny fraction of climate financial flows and has been deemed wildly insufficient compared to the needs, especially in reaching small-scale producers (Chiriac et al., 2023; Macquarie et al., 2020). There are growing climate finance opportunities, but project proponents and funders need data and tools to support the design and prioritization process. Policy and planning tasks for NDCs, national, subnational, and sectoral climate action plans and policies also require support of these resources.

The term “climate-smart agriculture (CSA)” has been adopted by the Food and Agriculture Organization (FAO) and other organizations to describe an approach “that helps guide actions to transform agri-food systems towards green and climate resilient practices… It aims to tackle three main objectives: sustainably increasing agricultural productivity and incomes; adapting and building resilience to climate change; and reducing and/or removing greenhouse gas emissions, where possible” (FAO, 2013; Lipper et al., 2014).

In recent years, there has been a proliferation of data and tools to help address climate-smart agriculture and inform decision-makers on topics at the intersection of climate and production agriculture in low-resource settings. The tools have many intended audiences and purposes, such as resilience or adaptation, mitigation, productivity, sustainability, and more; and many cover combinations of these climate and livelihood related goals. The Nationally Determined Contribution (NDC) Partnership has collected and categorized thousands of tools, guidance documents, platforms, and other resources related to climate action. There are 81 items in the NDC Partnership’s Climate Toolbox labeled as relevant to agriculture (NDC Partnership, 2023).

Agriculture is a highly site- and context-specific sector. Weather, soil, water, local technology, management practices, and knowledge vary significantly from one farm to the next. This makes the design of tools that are useful for a broad range of agricultural practitioners and investors especially challenging. This variability and complexity limit the capability of tools to inform the implementation of projects that benefit a wide range of producers. Nonetheless, collectively these tools can shed light on the relationship between climate and farming systems. Using the right tools can allow practitioners to screen for climate risks and quickly identify potential solutions and where more analysis may be needed.

Previous research efforts have shed light on existing climate and agriculture tools, as well as screening processes that leverage tools alongside qualitative inputs and steps. Brown (2017) documented and catalogued the climate screening processes and tools that major international agricultural funders implement, including the World Bank, United States Agency for International Development (USAID), UK Department for International Development, African Development Bank, the Swiss Agency for Development and Cooperation, and the Green Climate Fund (Brown, 2017). Separately, Douxchamps reviewed tools specifically designed for monitoring and evaluating climate resilience in agricultural development, with a specific focus on indicators and methods of measurement (Douxchamps et al., 2017). Lastly, Neset reviewed map-based tools that support climate change adaptation and are freely accessible on the web; however, this review did not focus on developing country-specific contexts, and it only considered tools related to adaptation to climate change (Neset et al., 2016).

In this review, the objective is to analyze climate-smart agricultural tools and create a fit-for-purpose categorization that can benefit decision-makers focused on small-scale producers in low-income contexts and help the research community better coordinate their efforts and build on existing tools. The focus is on interactive and accessible tools because these are the most available and convenient for a wide variety of users. Each tool addresses different aspects of climate-smart agriculture as it relates farming in low-resource settings. The purpose of this review is to lay out which tools are relevant for which purposes, enabling users to more quickly access tools that are useful to them. The review also seeks to highlight key limitations of available tools, including their underlying data gaps and use cases where the tools alone are insufficient in answering practitioners’ questions.

Methods

Given the many available tools available, and the many disparate uses of each tool, clear and targeted inclusion/exclusion criteria, a search strategy, and comparable classifications were needed to guide the analysis.

Lastly, a binary yes or no classification was used for whether each tool provided historical data, future projections, data source documentation, methodology documentation, or evidence of recent updates (>2020).

For tools that include climate impacts, data sources used by each tool were reviewed. Often, tools use data produced or sourced by the publishing institution, and many tools used various data sources that were not used by other tools characterized in this review. However, where underlying data was used by more than one tool characterized herein, these common data were identified in the results in order to shed light on the most common sources of land use, socioeconomic, and climate data that are used by these tools.

Based on careful consideration of the tools’ apparent intended users (sometimes explicitly cited on the website), or who the tools could provide valuable information to, five user archetypes were created: Implementing Partner, Agricultural Development Funder, Agricultural Policy Maker, Sustainable Sourcing Manager, and Agricultural Researcher. For further description of each of these user groups, see Table 2. Tools were also characterized by how they relate to climate change based on whether they incorporate climate risk (exposure, hazard, vulnerability) or measure mitigation potential (Table 3).

Results

A total of 29 tools were identified that met the criteria (see Table 4). Ten were from the NDC Toolbox, ten from the snowball approach, and nine from expert consultations. Out of 29 tools, 20 considered at least one aspect of risk (i.e., exposure, hazard, vulnerability), 11 were focused on climate mitigation, and three tools included aspects of both mitigation and risk (the CSA Programming and Indicator Tool, the Climate Risk Toolbox, and ClimateWatch).

Climate risk: Within the climate risk category, most of the 20 tools included functionality to assess an agricultural-related exposure (e.g., maize yield) to an explicitly climate-driven hazard (e.g., increased temperature or more variable precipitation). Under this category, map-based “simple” tools were the most common type. This type of tool enables users to explore the relationships between climate-related hazards such as temperature, precipitation, extreme weather events, floods, and water scarcity against agricultural-related exposures such as land use, cropland coverage (often by type of crop), or productivity. Examples of lighter touch, simple tools include Resilience Atlas, the World Bank Climate Change Knowledge Portal, and Agricultural Model Intercomparison and Improvement Project (AgMIP) International Food Policy Research Institute (IFPRI) Impacts Viewer. Of these 20 climate risk tools, 15 also provided functionality to assess vulnerability and adaptive capacity, but only four tools provided any way to assess or compare adaptive solutions. Adaptive capacity and vulnerability data generally took the form of socioeconomic data such as income or poverty levels, food security, education levels, or access to beneficial services, as well as biophysical adaptive capacity, e.g., soil quality. Some example tools are AgriAdapt, Climate Vulnerability Monitor, Resilience Atlas, Food Systems Dashboard, Agriculture Adaptation Atlas, Climate Risk Toolbox, Global Agricultural & Disaster Assessment System, and CSA Programming and Indicator Tool.

Tools addressing adaptive solutions generally are more qualitative and help users understand the relationships between environment, activity, and potential outcomes. The five tools that offered users the ability to explore adaptive solutions were the Agriculture Adaptation Atlas, CSA Programming and Indicator Tool, ClimateWatch, the “Climate Risk Planning & Managing Tools for Development Programmes in Agrifood Systems” (CRISP), and RegioCrop (Table 5). These are relatively new tools that have been designed precisely to offer solutions; they also allow users to download relevant data for more detailed or bespoke analyses.

Climate mitigation: For the 11 tools focused on climate mitigation, most (eight) allowed users to examine the GHG impacts of field-level interventions (Table 5). Other tools had a variety of different foci. For example, one tool each examined emissions at the national level (ClimateWatch); in a commodity supply chain; from a particular food product (ACE Calculator); and related to food loss and waste (FLW Value Calculator). These tools generally require more input data for analysis than simple tools but produce quantifiable impacts on GHG emissions and sometimes also yields, food security metrics, or biodiversity outcomes. For these tools to be informative, they typically require more detailed input data and knowledge of specific geographies, farming practices, and/or business management practices.

No tools attempted to address both risks and mitigation in a robust, holistic, or integrated way. The only tool that addressed both mitigation and adaptation was a CSA programming and indicator toolbox, intended to support monitoring, learning, and evaluation types of activities rather than to support agricultural decision-making directly.

Private sector supply chains and/or value chains: A topic that cut across climate risk and climate mitigation is related to tools designed to address commercial agricultural supply chains for specific crops or livestock products. Most of the tools categorized as climate risk also have the capability of providing insights on the vulnerability of crops in commercial supply chains. Two tools that explicitly address this functionality and provide user stories to showcase this usage are World Resources Institute (WRI)’s Aqueduct and AgriAdapt tools. The former has numerous examples of how companies have used the tool to help them better understand how climate-related water risks may play into sourcing decisions (WRI Aqueduct User Stories, 2023). The latter is calibrated to understand risks to specific supply chains, including Colombian coffee, Indian paddy rice, and Indian cotton.

There are six mitigation focused tools that explicitly address how private sector actors can incorporate GHG emission tradeoffs into their planning. These include the Cool Farm Tool, EX-Ante Carbon-balance Tool for Value Chains (EX-ACT VC), Agro-Chain Greenhouse Gas Emissions (ACE) calculator, Food Loss and Waste (FLW) Value Calculator, Trase Supply Chains, and the Global Livestock Environmental Assessment Model (GLEAM). They each have different commodity and value chain orientation and relevance, but each share the functionality of being able to estimate GHG emissions along a value chain as long as the user provides the right inputs. More detailed tools might require input data such as livestock head, feed type composition, soil or water management practices, fertilizer use, days of cultivation of crops, area of land use or land use change, etc.

Commonly used datasets to spatially explore risk: Tools that use data from externally produced data products rely on a limited number of sources for climate, crop, and socioeconomic data. Any data sources that were referenced in more than one risk tool are highlighted in Table 6. Most of these underlying datasets are trusted sources of widely available data used to spatially depict or model key metrics or variables representing exposure, hazard, or vulnerability. Tools that included external sources primarily used agricultural production data from FAO or IFPRI, socioeconomic data from the World Bank or USAID, and climate and environment data from the National Aeronautics and Space Administration (NASA), the National Oceanic and Atmospheric Administration (NOAA), CLIMate ADAptation (CLIMADA), and the Inter-Sectoral Impact Model Intercomparison Project (ISMIP).

User groups: Illustrative user groups were developed to shed light on the kinds of archetypical users that may want to use select tools and how they may derive use of each tool. Five user groups and illustrative examples for how they might use tools to answer different types of questions are described in Table 7.

Takeaways/discussion

Many valuable, easy-to-use tools exist offering non-climate experts and generalists' opportunities to gain interesting insights into the relationship between climate and small-scale producers and farming systems. Tools allow users to screen for climate risks and quickly identify where more digging or research is needed. These tools provide efficiency benefits, allowing users to streamline tasks and freeing up resources to dig deeper or ask other related questions. The tools also enable generalists to better understand concepts related to climate-smart agriculture, and in particular risk and climate mitigation. Integrated, appropriate use of these tools in development efforts (where relevant) can efficiently improve climate outcomes of projects.

No tool should be used as a standalone resource to comprehensively address climate-smart agriculture as it relates to smallholders. Tools that are meant to be broadly applicable are often limited in context-specific detail and nuance. They can be helpful for enabling users to broadly screen for potential climate-related impacts, assess alternative approaches for climate-relevant goals, and inform sustainability in agricultural supply chains. However, to fully address any program or intervention, a wide range of experts, stakeholders, resources, and customized analyses are needed alongside tools.

Researchers may be able to mine the tools to identify and download relevant datasets to conduct more tailored or bespoke analysis. One of the most important benefits of these tools is that they can point to the right datasets to be used for further exploration. The developers of these tools have spent significant time curating datasets relevant to climate, agriculture, and smallholder farming systems. However, these tools had to make tradeoffs between data coverage and specificity. Some of the tools reviewed here reflect data that is not the most spatially disaggregated (but may be the best available covering a wide geographical area of focus for the tool) or recent; for example, the Aqueduct tool includes MapSPAM crop data from 2010, which is 13 years old at the time of this publication. To maximally benefit from these tools when conducting bespoke analysis, researchers should consider both what data they can mine from these tools and what additional, more spatially and temporally disaggregated datasets (which often exist from individual projects, statistical agency data, or survey data collection) they can combine with the datasets in these tools to jumpstart efforts to produce tailored or bespoke analysis. Researchers should remember that they should not assume the tools’ data is up-to-date or the best-available data source for a given research question.

Tools do not collectively address resilience and adaptation planning very well. There are many tools that do address adaptation planning and resilience, but they tend to be more process oriented (e.g., a series of questions that guide the user through a process). Since adaptation options are highly specific to context—involving multiple factors such as likelihood or severity of hazards, propensity to be negatively impacted (vulnerability), the presence of valued assets (exposure), and details about the social/political/cultural context—they are necessarily less conducive to a stylized model or interactive tool. Nonetheless, the Agricultural Adaptation Atlas (AAA) and CRISP do characterize and display different types of adaptation options, and the AAA and Resilience Atlas incorporate relevant adaptive capacities that may be applicable in modulating climate impacts. The Climate Risk Toolbox includes geospatial data to analyze a target geography’s adaptive capacity. But few other interactive tools attempt to tackle resilience and adaptation in a very actionable way.

Tools could do a better job of estimating the impact of climate on a wide range of specific crops and at a high spatial resolution. Most climate impact tools only model the projected impact of climate on the world’s commonly grown, highly traded staple crops: like wheat, rice, cotton, and soy (e.g., the AgMIP IFPRI Impacts Viewer). Some global staple crops, such as wheat, rice, soy, and cotton, have spatial data at higher resolutions of 30 m × 30 m or general cropland at 10 m × 10 m. Models that use IFPRI’s MapSPAM dataset, such as the Resilience Atlas, GADAS, and Aqueduct, allow the user to explore 42 crops at coarser spatial resolution of 10 km × 10 km. Although the data and methods for estimating the productivity impacts of climate on a wide range of crops is now available and presumably a tool could be built to screen how these crops may become more, or less, impacted or suitable with climate change, such a tool was not identified. Tools estimating GHG emissions tend to have the capacity to be more crop explicit, depending on geography and data available for a specific investment.

Climate-smart agricultural tools generally do not cover a wide range of environmental impacts. Tools tend to focus on the impact of climate on productivity, or the impact of agricultural interventions on GHG emission. Fewer go further to assess how agricultural interventions or climate may impact other environmental outcomes such as biodiversity, soil, and water quality. Cool Farm Tool and the EU Reducing Emissions from Deforestation and Forest Degradation (REDD) Facility’s Land-use Planner include some information regarding biodiversity and environmental impacts. The Aqueduct tool focuses on assessing the impact of agriculture on water related outcomes. Tool outputs, which include biodiversity or environment-specific impacts, tend to require more detailed input data on farm management and geographic information.

Conclusion

There is a plethora of interactive tools that share valuable data relevant to climate change and small-scale farming in low-income countries. Here, a non-exhaustive but thorough search identified 29 interactive and accessible tools out of hundreds that were considered. These tools compile data differently to provide valuable insights and details that can be leveraged in various ways to provide needed information to a wide range of users. The most valuable tools are the ones that not only allow information to be displayed but also provide actionable insights and clear use cases that allow any user to easily access and understand how the tools can inform their actions. While valuable, it is important to remember that these tools are just that - tools. They should always be used in conjunction with other sources of data and information (e.g., additional data sets, expert consultations, literatures reviews, other targeted research) to inform decision-making.