Agricultural Decision Support Tools Research Articles

Explaining the use of online agricultural decision support tools with weather or climate information in the Midwestern United States

Agricultural decision support tools (DSTs) with weather or climate information can provide useful information to help stakeholders make operational farming decisions and adapt to increasingly variable weather or climate in the context of climate change. However, many of these DSTs are still not fully utilized. Understanding the use of DSTs can help identify strategies to promote their usage to more end-users. This study surveyed farmers (n = 2,633) and advisors (n = 2,719) across 12 states in the Midwest to draw comparisons on their usage of DSTs and factors influencing the usage. The advisors are more likely to take advantage of free and publicly available sources than farmers. Advisors are also more likely to agree on the usefulness of DSTs, feel social pressure to use DSTs, be concerned and perceive risks from variable weather, believe in climate change, and show positive attitudes towards climate change adaptation than farmers. Concerns about weather or climate, descriptive social norms, greater farm size, and general propensity to adopt a new technology are positively associated with higher adoption rate of DSTs for both farmers and advisors. Higher level of perceived behavioral control to deal with weather-related risks, injunctive social norms, gender (male), and age are positively associated with higher adoption rate of DSTs for only advisors. Positive adaptation attitude towards climate change and higher education level are positively associated with higher adoption rate of DSTs for only farmers. Unlike advisors, age is negatively associated with higher adoption rate of DSTs for farmers. Implications of our findings include DST educators leveraging social networks and reinforcing social norms to promote usage among current non-users, building up both farmers' and advisors’ confidence and knowledge in using DSTs, understanding the role of advisors as “change agents” to promote DST usage among farmers, and connecting networks of “innovators” and “early adopters.” With more and more DSTs developed, future scholarship can draw upon our findings to understand how to encourage DSTs adoption among current non-users and extend to other regions.

A decision support framework assessing management impacts on crop yield, soil carbon changes and nitrogen losses to the environment

AbstractAgricultural management practices have multiple impacts on farming systems, including crop yield, soil fertility parameters such as soil organic carbon (SOC), and environmental quality. Agricultural decision support tools (DSTs) are key in sustainable farm strategies to optimize yield and minimize environmental losses because both the current agroecosystem properties as well as the effectiveness of management practices are highly variable in time and space. Here, we introduce a highly data‐driven framework focusing on the evaluation of agronomic measures to reach agronomic and environmental targets. We demonstrate the potential of this approach by a proof of principle for 81 selected farm types across Europe, focusing on measures with respect to crop rotation, fertilization and soil tillage. Synthesizing data from long‐term experiments and meta‐analytical models, we estimated the impact of these measures on crop yield, SOC and N surpluses, while accounting for site‐specific properties for the current and desired situation. The impacts of these measures on all farm types have been quantified, and optimum sets of agronomic measures have been selected in order to maximize crop yield and SOC levels and minimize N surpluses to reach the critical values for NO3− concentrations in groundwater. Our results, quantifying trade‐offs among sustainability indicators that have traditionally been analyzed separately, illustrate that the suitability of measures varies by soil, climate and crop types within Europe. Our approach is promising for mapping region‐specific management recommendations and evaluating the effectiveness of agronomic measures over multiple environmental goals and targets.Highlights We find a lack of empirical‐based DSTs holistically assessing agronomic practices and indicators. Meta‐analytical models were used to assess impacts of best fertilizer, tillage and crop measures. Our multi‐criteria analysis shows impacts vary with crop, soil and climate in European regions. We demonstrate our developed framework focusing on crop yield, soil carbon and nitrogen losses.

Creating the data basis to adapt agricultural decision support tools to new environments, land management and climate change—A case study of the RiceAdvice App

AbstractIncreasing demand for land to ensure human food security in the future has already impelled agricultural production into marginal areas. The environmental conditions found there have a more pronounced impact on agricultural productivity than in the systems used so far under favourable conditions. In addition to this challenge, climate change is expected to increase the unreliability of weather conditions (through increased variability and occurrence of extremes) for farmers considerably. This unreliability is even more serious in developing countries’ farming system where food security is vulnerable. Current efforts in digitalization offer great possibilities to improve farmers` decision‐making processes. A wide range of online tools and smartphone applications is available to support both agricultural extension services and smallholder farmers alike. These apps are often parameterized and validated to certain environments and are troubled when applied to new geographical locations and different environmental conditions. We have conducted field trials to demonstrate potential methods to close knowledge gaps in the data background for one of these apps, RiceAdvice, concerning three key aspects: shifting of cropping calendar, adjustment of fertilizer management and genotype selection. Sites in Ethiopia, Madagascar and Rwanda were selected to represent altitudinal gradients, with overlapping elevations reflecting differences in temperature to enable cross‐country comparisons. Planting dates were distributed throughout three calendar years, with continuous iterative planting dates taking place in Madagascar, in‐ and off‐season planting dates in Rwanda with different fertilizer applications, and one planting date during each rainy season in Ethiopia with different management options. With these trials, we have been able to identify key data sets needed for the adaptation of agricultural decision support tools to new environments. These include the assessment of climatic constraints on innovations to cropping calendars (e.g. double cropping), informed selection of alternative varieties able to complete crucial parts of their phenological development to avoid temperature‐related stress inducing, for example spikelet sterility in rice in late development stages and the effectivity of potential innovations in fertilizer management strategies.

Advancing a farmer decision support tool for agronomic decisions on rainfed and irrigated wheat cropping in Tasmania

Well-designed agricultural decision support tools (DS) equip farmers with a rapid, easy way to compare multiple scenarios as well as the influence of different management strategies on crop production. One such tool, CropARM (http://www.armonline.com.au) assists users in establishing a framework of risk, with simulations incorporating climate scenarios and management actions, such as fertiliser rates, sowing time, row spacing, and irrigation regimes. When used in conjunction with soil and climate characteristics, biophysical model-based DS tools provide information that complements farmer experience and helps establish a framework for risk management given local climate characteristics. In this study, we used the APSIM model to provide the simulation data necessary to expand CropARM for new management conditions and environments in southern Australia. Prior to this work being undertaken, no CropARM data was available for Tasmania and no sites in CropARM allowed users to compare rainfed and irrigated wheat crops. This study collated data from 27 plots across ten sites in Tasmania, from the period 1981 to 2011, under both rainfed and irrigated conditions. APSIM was parameterised with these field observations and the subsequent scenario simulations were used to populate CropARM. Wheat cultivars used in the parameterisation of APSIM include Brennan, Isis, Mackeller, Revenue, Tennant (winter types) and Kellalac (spring type). The validation showed reliable model parameterisation, with an r2 value of close to 1, which is considered satisfactory. 670,680 simulations were undertaken and incorporated within the CropARM database for wheat cropping systems across Tasmania. With regularly updated climate streams, the free online framework provided by CropARM gives users the ability to assess downside risks associated with several different crop management alternatives, and by simultaneously comparing multiple scenarios, users can select management options that are likely to adhere most closely with their desired management objectives.

Sustainable production: Using simulation modeling to identify the benefits of green information systems

Researchers and practitioners highlight the potential for information systems to promote sustainability in agricultural production, but little is known about the private and social benefits of specific agricultural decision support tools. In this study, we utilize the resource-based view to assess a specific green technology using an agricultural-economics simulation to estimate the quantitative benefits of this technology expressed as dollars saved and reduced greenhouse gas emissions. In particular, we employ a five-step simulation modeling approach within a micro-economic model of crop production to assess the ability of yield monitors to promote liquefied petroleum (LP) gas savings and subsequently reduce production costs, reduce greenhouse gas (GHG) emissions associated with LP gas burning, and generate additional revenue at a market for GHG mitigation credits. We estimate that the total benefits of using the green IS to improve the harvesting decision would have been $82 million in post-harvest cost savings and a significant reduction in greenhouse gas emissions. We present this simulation modeling approach, a common methodology in environmental sciences and economics, as a viable methodology for IS researchers interested in modeling intricate decision-making processes that are impacted by technology.

From grid to field: Assessing quality of gridded weather data for agricultural applications

Abstract High quality measured weather data (MWD) are not available in many agricultural regions across the globe. As a result, many studies that dealt with global climate change, land use, and food security scenarios and emerging agricultural decision support tools have relied on gridded weather data (GWD) to estimate crop phenology and crop yields. An issue is the agreement of GWD with MWD and the degree to which this agreement may influence the utility of GWD for agricultural research. The objectives of this study were: (i) to compare the agreement of two widely used gridded weather databases (GWDs) (Daymet and PRISM) and MWD, (ii) to evaluate their robustness at simulating maize growth and development, and (iii) to examine how GWD compare relative to weather data interpolated from existing meteorological stations for which MWD are available. The U.S. Corn Belt, a region that accounts for 43 and 34% of respective global maize and soybean production, was used as a case of study because of its dense weather station network and high-quality MWD. Historical daily MWD were retrieved from 45 locations across the region, resulting in ca . 1300 site-years. To test the accuracy of GWDs, separate simulations of maize yield and development were performed, separately for the two GWDs and MWD, using a well-validated maize crop model. For both GWDs, small biases were observed for temperature and growing degree-days in relation with MWD. However, accuracy was much lower for relative humidity, precipitation, reference evapotranspiration, and degree of seasonal water deficit. There was close agreement in duration of vegetative and reproductive phases between GWD and MWD, with root mean square error (RMSE) ranging from 3 to 7 days for the different crop phases and GWDs. However, robustness of GWDs to reproduce maize yields simulated using MWD was lower as indicated by the RMSE (18 and 24% of average yield for Daymet and PRISM, respectively). There was also a high proportion of site-years (20 and 32% for Daymet and PRISM, respectively) exhibiting a yield deviation >15% in relation to the yield simulated using MWD. Data interpolation using a dense weather station network resulted in lower RMSE% for simulated phenology and yields relative to GWDs. Findings from this study indicate that GWD cannot replace MWD as a basis for field-scale agricultural applications. While GWD appear to be robust for applications that only require temperature for prediction of crop stages, GWD should not be used for applications that depend on accurate estimation of crop water balance, crop growth, and yield. We propose that the evaluation performed in this study should be taken as a routinary activity for any research or agricultural decision tool that relies on GWD.

Decision support tools for agriculture: Towards effective design and delivery

Decision support tools, usually considered to be software-based, may be an important part of the quest for evidence-based decision-making in agriculture to improve productivity and environmental outputs. These tools can lead users through clear steps and suggest optimal decision paths or may act more as information sources to improve the evidence base for decisions. Yet, despite their availability in a wide range of formats, studies in several countries have shown uptake to be disappointingly low. This paper uses a mixed methods approach to investigate the factors affecting the uptake and use of decision support tools by farmers and advisers in the UK. Through a combination of qualitative interviews and quantitative surveys, we found that fifteen factors are influential in convincing farmers and advisers to use decision support tools, which include usability, cost-effectiveness, performance, relevance to user, and compatibility with compliance demands. This study finds a plethora of agricultural decision support tools in operation in the UK, yet, like other studies, shows that their uptake is low. A better understanding of the fifteen factors identified should lead to more effective design and delivery of tools in the future.

Modeling potato root growth and water uptake under water stress conditions

Potato (Solanum tuberosum L.) is considered a drought sensitive crop. Accurate simulation of root growth is critical for estimating water uptake dynamics. However, data required to build and test advanced potato root simulation approaches is lacking. Previously unpublished data from our soil–plant–atmosphere-research (SPAR) chambers was used to evaluate a new two-dimensional diffusive root growth module linked to the existing potato model SPUDSIM. The root module consisted of diffusive parameters controlling the direction of root growth in horizontal and vertical directions, and an additional convective term for vertical growth within the soil. This modified SPUDSIM was tested against observed SPAR data which consisted of root distribution in the soil profile and organ dry weights (DW) at harvest, plus daily water uptake patterns for each of six different irrigation treatments. The difference between simulated and observed DW data was within two standard errors for most plant organs—root DW was over-predicted for well-watered plants. Spatial and temporal patterns of root distribution and water contents were reproduced well. However, the model tended to over-estimate water uptake from soil layers closer to the surface. Differences in simulated root growth patterns among irrigation treatments were the result of fluctuations in soil water status, bulk density, and root density which, in turn, affected the amount of carbon allocated to the roots in different soil layers and the value of the convective term. These results suggest the new module will provide more reliable predictions of potato water uptake for improved agricultural decision support tools.

Agricultural Advisors: A Receptive Audience for Weather and Climate Information?

Abstract As the climate in the midwestern United States becomes increasingly variable because of global climate change, it is critical to provide tools to the agricultural community to ensure adaptability and profitability of agricultural cropping systems. When used by farmers and their advisors, agricultural decision support tools can reduce uncertainty and risks in the planning, operation, and management decisions of the farm enterprise. Agricultural advisors have historically played a key role in providing information and guidance in these decisions. However, little is known about what these advisors know or think about weather and climate information and their willingness to incorporate this type of information into their advice to farmers. In this exploratory study, a diverse set of professionals who advise corn growers, including government, nonprofit, for-profit, and agricultural extension personnel, were surveyed in four states in the midwestern Corn Belt. Results from the survey indicate that advisors are more influenced by current weather conditions and 1–7-day forecasts than longer-term climate outlooks. Advisors predominantly consider historical weather trends and/or forecasts in their advice to farmers on short-term operational decisions versus longer-term tactical and strategic decisions. The main conclusion from this analysis is that opportunities exist to further engage the advisor community on weather and climate issues and, through them, the farmers who are managing the land.

Leaf growth and senescence rates of three pasture grasses and wheat

A glasshouse study was conducted under ideal conditions to determine leaf appearance, elongation, and senescence rates along with life span and leaf length characteristics of four grass species: wheat (Triticum aestivum L.), brown back wallaby grass (Rytidosperma duttonianum (Cashmore) Connor and Edgar), phalaris (Phalaris aquatica L.), and annual ryegrass (Lolium rigidum Gaud.). This study provided a comprehensive characterisation of leaf turnover rates for the entire life cycle of these grasses, some of which are poorly characterised. Importantly, leaf senescence rate has been captured in the same conditions as the other leaf rates of the life cycle. Leaf position proved to be a significant explanatory variable in each of the leaf turnover rates. The relationships between leaf position and the components of leaf turnover were most commonly represented by non-linear models. Further studies may be necessary to validate these statistical models to field situations. However, this information will be useful to calibrate the senescence algorithms of plant growth models in agricultural decision support tools, which may then be applied to simulation studies including the assessment of grass curing for planning activities such as resource allocation, wildfire suppression, and execution of prescribed burning programs by fire management agencies.

Participatory design of agricultural decision support tools: taking account of the use situations

Abstract Although many agronomic researchers currently focus on designing and developing decision support systems, they rarely discuss the methodological implications of such work. In this paper, with the examples of two decision support systems, we propose methodological elements for conducting the participatory design of such tools. Our proposition aims at building dialogue between designers and users but also between humans, tools and work situations. We focus on two main stages: first, a diagnosis of the uses of decision tools within current working situations and, second, the use of a prototype of the tool under design. The first stage serves to characterize the diversity of uses and user situations in order to determine the tool’s flexibility and to identify new concepts for tools. The second stage involves setting up an arrangement whereby a prototype of the decision support tool, open to amendment, can be used in work-like situations and then discussed during debriefing sessions among designers and users. This stage mediates dialogue between all the participants and allows them to develop cross-learning processes. We discuss how these two stages allow for a coordinated expansion of three spaces: the concept space, the knowledge space and the use space. We then discuss the need for such participatory design processes described as dialogical design processes and their contribution to produce new agronomic knowledge supporting a more sustainable agriculture. Finally, we point out a need to provide more opportunity for scientific discussion on participatory design approaches and on design methodology more broadly within the agronomic community.