Role of existing and emerging technologies in advancing climate-smart agriculture through modeling: A review

Crop yield prediction is critical for investigating the yield gap and potential adaptations to environmental and management factors in arid regions. Crop models (CMs) are powerful tools for predicting yield and water use, but they still have some limitations and uncertainties; therefore, combining them with machine learning algorithms (MLs) could improve predictions and reduce uncertainty. To that end, the DSSAT-CERES-maize model was calibrated in one location and validated in others across Egypt with varying agro-climatic zones. Following that, the dynamic model (CERES-Maize) was used for long-term simulation (1990–2020) of maize grain yield (GY) and evapotranspiration (ET) under a wide range of management and environmental factors. Detailed outputs from three growing seasons of field experiments in Egypt, as well as CERES-maize outputs, were used to train and test six machine learning algorithms (linear regression, ridge regression, lasso regression, K-nearest neighbors, random forest, and XGBoost), resulting in more than 1.5 million simulated yield and evapotranspiration scenarios. Seven warming years (i.e., 1991, 1998, 2002, 2005, 2010, 2013, and 2020) were chosen from a 31-year dataset to test MLs, while the remaining 23 years were used to train the models. The Ensemble model (super learner) and XGBoost outperform other models in predicting GY and ET for maize, as evidenced by R2 values greater than 0.82 and RRMSE less than 9%. The broad range of management practices, when averaged across all locations and 31 years of simulation, not only reduced the hazard impact of environmental factors but also increased GY and reduced ET. Moving beyond prediction and interpreting the outputs from Lasso and XGBoost, and using global and local SHAP values, we found that the most important features for predicting GY and ET are maximum temperatures, minimum temperature, available water content, soil organic carbon, irrigation, cultivars, soil texture, solar radiation, and planting date. Determining the most important features is critical for assisting farmers and agronomists in prioritizing such features over other factors in order to increase yield and resource efficiency values. The combination of CMs and ML algorithms is a powerful tool for predicting yield and water use in arid regions, which are particularly vulnerable to climate change and water scarcity.

/pdf/coupling-process-based-models-and-machine-learning-3nawkpx5.pdf

Coupling Process-Based Models and Machine Learning Algorithms for Predicting Yield and Evapotranspiration of Maize in Arid Environments


 Wheat is the major agricultural crop in Iran. Using reliable tools to estimate wheat grain yield may help the regional planners and stakeholders to employ proper cultivation approaches to increase grain yield in different regions. The process-based crop growth models are the essential analytical tools for representing the main interactions between the environment, crop, resources, and yield production. For these reasons, the crop growth models have been used as an important component in productive farming systems for assessing and improving the crop production at field and regional scales. In this study, the parameterized AquaCrop model was employed to simulate the four major elements that affect the crop growth process, such as soil water content, canopy cover, biomass, and grain yield, of three irrigated winter wheat cultivars. The field experiments were conducted in four locations with different climatic conditions in the south and southwest of Iran during three consecutive growing seasons of 2016–2017, 2017–2018, and 2018–2019. Results showed that AquaCrop had reliable simulation of soil water content with normalized root mean square error (NRMSE) and Nash–Sutcliffe coefficient (CNS) ranging from 0.05 to 0.15 and 0.69 to 0.85, respectively, for the four study locations. The NRMSE and CNS ranged from 0.10 to 0.21 and 0.83 to 0.97 for canopy cover, respectively. The NRMSE and CNS values were 0.11–0.26 and 0.80–0.97 for biomass, respectively. These results indicate that the parameterized AquaCrop model has high skill in the simulation of soil water content, canopy cover, biomass, and grain yield of different winter wheat cultivars in the vast regions in the southern part of Iran. The model was then used to simulate wheat grain yield and water productivity (WP) for 324 scenarios, including nine sowing dates, three irrigation levels, and three climatic conditions. AquaCrop showed that it was skillful to simulate grain yield and WP of different irrigated winter wheat cultivars in different regional climatic conditions, deficit irrigation levels, and sowing dates. Overall, the AquaCrop model could be used as a reliable decision-making tool for the field managers and stakeholders.

Regional modeling of winter wheat yield and water productivity under water-saving irrigation scenarios

Crop growth models can be useful tools to evaluate potential scenarios, yet the limited field data is an obstacle for providing significant insights at the local level. This certainly applies to several regions in Africa, where crop models built regional scenarios based on low resolution field data, leading to unsupported agricultural interventions on smallholders' systems. This review provides a synthesis analysis of crop modeling efforts in Africa using Agricultural Production Systems Simulator (APSIM) studies as a case-study. This research aims to highlight the value of standardized protocols to collect, store and deploy field data, and highlights the critical issue of limited data accessibility of published manuscripts and unavailability of a data sharing platform.Investment in local-level data collection and data sharing platforms is critical to guarantee the advancement of science and to provide reliable assessments to address complex challenges of food, nutrition, and climate security in Africa. 

The urgency for investment on local data for advancing food assessments in Africa: A review case study for APSIM crop modeling

Prediction of annual crop yields at a county granularity is important for national food production and price stability. In this paper, towards the goal of better crop yield prediction, leveraging recent graph signal processing (GSP) tools to exploit spatial correlation among neighboring counties, we denoise relevant features via graph spectral filtering that are inputs to a deep learning prediction model. Specifically, we first construct a combinatorial graph with edge weights that encode county-to-county similarities in soil and location features via metric learning. We then denoise features via a maximum a posteriori (MAP) formulation with a graph Laplacian regularizer (GLR). We focus on the challenge to estimate the crucial weight parameter <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mu$</tex> , trading off the fidelity term and GLR, that is a function of noise variance in an unsupervised manner. We first estimate noise variance directly from noise-corrupted graph signals using a graph clique detection (GCD) procedure that discovers locally constant regions. We then compute an optimal <tex xmlns:mml="http://www.w3.org/1998/Math/MathML" xmlns:xlink="http://www.w3.org/1999/xlink">$\mu$</tex> minimizing an approximate mean square error function via bias-variance analysis. Experimental results from collected USDA data show that using denoised features as input, performance of a crop yield prediction model can be improved noticeably. 

/pdf/unsupervised-graph-spectral-feature-denoising-for-crop-yield-3uzkpnxz.pdf

Unsupervised Graph Spectral Feature Denoising for Crop Yield Prediction

Abstract Crop models are useful tools because they can help understand many complex processes by simulating them. They are mainly designed at a specific spatial scale, the field. But with the new spatial data being made available in modern agriculture, they are being more and more applied at multiple and changing scales. These applications range from typically at broader scales, to perform regional or national studies, or at finer scales to develop modern site-specific management approaches. These new approaches to the application of crop models raise new questions concerning the evaluation of their performance, particularly for downscaled applications. This article first reviews the reasons why practitioners decide to spatialize crop models and the main methods they have used to do this, which questions the best place of the spatialization process in the modelling framework. A strong focus is then given to the evaluation of these spatialized crop models. Evaluation metrics, including the consideration of dedicated sensitivity indices are reviewed from the published studies. Using a simple example of a spatialized crop model being used to define management zones in precision viticulture, it is shown that classical model evaluation involving aspatial indices (e.g. the RMSE) is not sufficient to characterize the model performance in this context. A focus is made at the end of the review on potentialities that a complementary evaluation could bring in a precision agriculture context. 

/pdf/a-review-of-methods-to-evaluate-crop-model-performance-at-stc9gu62.pdf

Jonathan Richetti

Papers

A review of methods to evaluate crop model performance at multiple and changing spatial scales

Using remote sensing, process-based crop models, and machine learning to evaluate crop rotations across 20 million hectares in Western Australia