Search

In this study, we employ advanced data-driven techniques to investigate the complex relationships between the yields of five major crops and various geographical and spatiotemporal features in Senegal. We analyze how these features influence crop yields by utilizing remotely sensed data. Our methodology incorporates clustering algorithms and correlation matrix analysis to identify significant patterns and dependencies, offering a comprehensive understanding of the factors affecting agricultural productivity in Senegal. To optimize the model's performance and identify the optimal hyperparameters, we implemented a comprehensive grid search across four distinct machine learning regressors: Random Forest, Extreme Gradient Boosting (XGBoost), Categorical Boosting (CatBoost), and Light Gradient-Boosting Machine (LightGBM). Each regressor offers unique functionalities, enhancing our exploration of potential model configurations. The top-performing models were selected based on evaluating multiple performance metrics, ensuring robust and accurate predictive capabilities. The results demonstrated that XGBoost and CatBoost perform better than the other two. We introduce synthetic crop data generated using a Variational Auto Encoder to address the challenges posed by limited agricultural datasets. By achieving high similarity scores with real-world data, our synthetic samples enhance model robustness, mitigate overfitting, and provide a viable solution for small dataset issues in agriculture. Our approach distinguishes itself by creating a flexible model applicable to various crops together. By integrating five crop datasets and generating high-quality synthetic data, we improve model performance, reduce overfitting, and enhance realism. Our findings provide crucial insights for productivity drivers in key cropping systems, enabling robust recommendations and strengthening the decision-making capabilities of policymakers and farmers in data-scarce regions.

This systematic review examines the application of artificial intelligence (AI), including machine learning (ML) and deep learning (DL), for climate change adaptation and mitigation in Iran, Pakistan, and Turkey. These three nations—key Economic Cooperation Organization (ECO) members and a nexus between Europe and South Asia—are experiencing diverse environmental challenges due to varying climatic conditions. Following the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines, we conducted a comprehensive search in the Scopus database, ultimately identifying 76 relevant articles out of an initial 492. Although some articles utilized multiple techniques, classical ML methods appeared in approximately 37.3 % of the studies, neural network paradigms in about 57.5 %, and optimization or meta-heuristic algorithms in around 5.0 %. Regarding thematic focus, about 33.3 % of the articles addressed water resource management, 22.2 % focused on climate prediction, 11.1 % on land and agriculture, 9 % on ecosystem modeling, and 24.2 % on natural disaster preparedness and response. The analysis reveals a growing but uneven body of research utilizing AI across the ECO countries. By highlighting successful applications, identifying key gaps—such as limited cross-border collaboration and inconsistent data availability—and proposing a framework for more integrated research, this review aims to guide future initiatives that leverage AI's potential to improve climate resilience and sustainability in the region.

Abstract The nitrogen regulatory policy (NRP) solution is introduced as a mitigation measure against environmental nitrogen losses and keeps food production in the Safe Operating Space of the Nitrogen Planetary Boundary. Meanwhile, scientific research shows that steps taken to reduce environmental harm can increase the unpredictability of calorie production from crops. This study sought to investigate the impact of NRP solutions on the level of risk of accessibility to calorie sources from domestic production, the variations in calorie sources by livestock and non‐livestock diet components, and the responses of different dietary preferences, namely, poor, medium, and rich livestock protein diets, against NRP solutions in the Zayandeh‐Rud River basin, Iran. We developed the aggregate household food security index (AHFSI) and combined it with outputs of crop simulation model to examine how changes in dietary energy supplies under three NRP scenarios—low, moderate, and high nitrogen fertilizer application—affect the stability of three regional dietary preferences. The comparison of NRP scenarios movements realized that increases (or decreases) in nitrogen fertilizer rates contradicted the stability in AHFSI. Additionally, a one‐unit change in the average calories from non‐livestock sources, such as wheat and potatoes, results in greater fluctuations in the standard deviations of produced calories compared to changes in meat and dairy production. We proposed that in order to prevent adverse effects of NRP solutions on food security, mitigation strategies addressing the NRP solution should be structured based on (i) regional heterogeneities, (ii) type of crops, that is, food and feed crops, (iii) the range of nitrogen rates movement; (iv) and the socioeconomic background related to dietary preferences or economic deciles of food expenditure.

Crop yield is usually affected by impending weather, climate conditions, and human interventions like irrigation. Hence, the prompt detection of regions experiencing water scarcity can aid in implementing effective mitigation strategies. Our study utilized a data-driven approach to compute a Water Demand Index (WDI), which incorporates crucial first-order geophysical variables like ambient temperature, vegetation status, and soil moisture, to identify water-stressed fields in Senegal’s agricultural regions during the millet planting, growing, and harvesting periods. We have also explored various scenarios for enhancing the accuracy of millet yield prediction by incorporating other drought indices, soil characteristics, and a bias correction factor. The novel aspects of this research include the development of regional crop yield predictive models based on sub-seasonal and unaggregated crop information to reflect local variances. To optimize the hyperparameters of machine learning (ML) models, various techniques were utilized. Meanwhile, the performance of these ML models was evaluated using a nested cross-validation approach. The outcomes of the analysis demonstrate that the Random Forest Regressor model exhibits superior predictive performance. The results imply that a holistic approach, encompassing diverse environmental factors and crop growth stages, could result in more precise and dependable millet yield predictions. These results mark a leap forward in crop yield prediction for data-poor regions, paving the way for more accurate and efficient agricultural management. By emphasizing the critical role of detailed, spatio-temporal data, this work substantially enhances model precision, informing targeted strategies to boost yields. This advancement stands to serve producers and consumers alike, fostering sustainable, high-yield agriculture.

Abstract In Senegal, agriculture is an important sector underpinning the socioeconomic fabric of the populace. Notably, the agricultural production in this region exhibits heightened sensitivity to climatic perturbations, particularly droughts and heat waves. This study aims to determine the resilience of different agronomic interventions for farmers practicing mixed farming that produce both crops (i.e., groundnut (Arachis hypogaea L.) and pearl millet (Pennisetum glaucum (L.) R. Br.)) and raise animals in the Groundnut Basin in Senegal, which holds historical and socioeconomic significance. To understand the current situation regarding demographics, economics, consumption behavior, and farm operations for smallholder farmers, data were comprehensively collected from government and nongovernment organizations (NGO) reports, scientific papers, organization databases, and surveys. Additionally, the Agricultural Production Systems sIMulator (APSIM) was used to understand how combinations of three planting dates, three plant densities, and six urea nitrogen (N) fertilizer rates affected the yield of pearl millet, which were used as the alternative scenarios to the baseline in the farm modeling and analyses. All the collected and generated data were used as inputs into the Farm Simulation Model (FARMSIM) to generate economic, nutritional, and risk data associated with mixed farming systems. The generated data were then used to determine the resilience of the alternative scenarios against the baseline. Initially, a multi‐objective optimization was employed to meet nutritional needs while maintaining a healthy diet at the lowest cost. Then, the scenarios that met the population's nutritional requirements were evaluated based on four economic indicators: net cash farm income (NCFI), ending cash reserves (EC), net present value (NPV), and internal rate of return (IRR). Lastly, those that passed the economic feasibility test were ranked based on risk criteria certainty equivalent (CE) and risk premium (RP). The analyses found N fertilizer rates of 0, 20, and 100 kg N ha−1 were generally economically not feasible. Additionally, medium (early‐July to late‐August) and late (late‐July to mid‐September) planting dates generally performed better than early (early‐June to late‐July) planting dates, while plant densities of 3.3 and 6.6 pL m−2 performed better than 1.1. The robust resilience approach introduced in this study is easily transferable to other regions.

AbstractWe conducted a comprehensive review of livestock farming in Bangladesh to unveil current challenges and potential opportunities in this agriculture sector. Six challenges were selected as the major constraints to livestock farming in Bangladesh: climate change and natural hazards, poor veterinary care, breeding and management resources, marketing and international trade, and the SARS-CoV-2 pandemic. Solutions to these challenges (such as hazardous weather shelters, intensified surveillance of biosecurity for farms and markets, and non-government, government, and private organizations working together to educate and assist farmers) must target specific regions where the solutions would have the greatest effect.

Subsurface drainage in humid areas prevents field waterlogging but also transports nutrients to freshwater systems. Controlled drainage (CD) reduces drainage discharge and nutrient transport from fields. Some regions are expected to experience increased precipitation in the future, requiring CD to be evaluated under a changing climate. The objective of this study was to compare the performance of CD under two weir managements for a future period (2030–2059) and historical period (1992-2021) in southeast Michigan, USA. Climate projections were obtained for the shared socioeconomic pathway 245 emission scenario. Aggressive management involved maintaining the weir height at 40 cm during the growing season and 15 cm during the non-growing season, with a longer period of managed flow compared to common management, which maintained the weir height at 50 cm during the growing season and 30 cm during the non-growing season. It was predicted that the 30-year average annual precipitation would not change significantly in the future. The 30-year average mean monthly temperature would increase by 3.0°C in the future compared to the historical period. We performed simulations using the calibrated Root Zone Water Quality Model 2 (RZWQM2). The average drainage discharge in the future indicated a 20% increase in the 30-year average drainage discharge for a field with free drainage. The CD with common and aggressive managements reduced drainage discharge by 59% and 67% for the historical period, whereas the performance of CD was even better for the future period (63% and 72%, respectively). The improved future performance of CD can be attributed to a shift in precipitation patterns, with reduced precipitation during the growing season and increased precipitation during the non-growing season. As a result, the more aggressive weir management during this period created additional opportunities for reducing drainage discharge. In conclusion, aggressive management resulted in a slightly better flow-reducing performance than common management while indicating that both methods would effectively reduce drainage discharge in the likely future scenario.

IntroductionThe lack of a holistic view of agriculture’s social and economic aspects emerges as a limiting factor in policy implementations of sustainable intensification. The aim of this study was to (i) synthesize data from the five domains of sustainable agriculture intensification framework focusing on Senegal as a case study, (ii) harmonize and scale data used to minimize spatial–temporal differences, and (iii) build a digital tool aiming researchers and policymakers to facilitate the data visualization and connection.MethodsWe propose and prototype an interactive digital decision-support tool as a potential solution to integrate environmental, social, economic, human and production domains in agriculture projects. To demonstrate the use of the tool, a case study using data from Senegal was developed to show the benefits of open science and connectivity among domains.Results and discussionThe digital tool presented here allows users to explore an open repository from Senegal serving as a foundational approach to integrate diverse agricultural domains when developing sustainable intensification projects.

Abstract Malawi's food and nutrition security heavily relies on rainfed agricultural production, which is prone to the effects of climate change, such as floods and dry spells. To improve the situation, policies have been established as a guide to assist in the implementation of agricultural technologies in the country. This study aims to evaluate the effect of two national policies on addressing nutrition, economics, and food security by introducing modern/existing technologies in the Phalombe district in Malawi. To gain insight into the representative farm operations, financial, demographic, and consumption behaviors of smallholder farmers, the data were obtained through a combination of focus group discussions, expert opinions, and various published reports, including both government and nongovernment sources. The farm simulation model (FARMSIM) was utilized to analyze district‐level crop and livestock farm operations. In this study, four irrigation technologies (treadle pump system, motorized pump system, solar‐powered system, and river diversion irrigation system) were assessed among the agricultural interventions implemented in Malawi. Also, seven land uses (i.e., corn, onion, bean, rice, tomato, leafy vegetable, and cabbage) were developed and simulated under irrigation scenarios. The irrigation systems and land use scenarios were evaluated based on economics, productivity, nutrition, and risk. The treadle pump and leafy vegetable scenario were generally identified as the most preferred combination. In addition, the productivity and riskiness of the other scenarios can be improved by selecting the best combination of crop and irrigation systems. The study concludes that government and private investment in subsidies for large‐scale irrigation systems, as well as improvements to market prices and infrastructure, are necessary to boost food security and enhance the livelihood of smallholder farmers in rural Malawi.

Understanding the end-use of water is essential to a plethora of critical research in premise plumbing. However, direct access to end-use data through physical sensors is prohibitively expensive for most researchers, building owners, operators, and practitioners. Therefore, machine learning models can alleviate these costs by predicting downstream end-use events (e.g., sink, shower, dishwasher, and washing machine) via an affordable subset of upstream sensors. Choosing which upstream sensors, as well as data preprocessing methods, are best for machine learning has historically been a manual process. This paper proposes a novel approach to systematically configure the machine platform automatically. The optima were determined through a Pareto analysis of the exhaustive combinations of upstream predictors and preprocessing methods. The model was trained and validated with real-world data obtained from a house that has been extensively monitored for over a year. Results from the analysis suggested that downstream events can be effectively predicted with minimum overfitting error for most categories, using as few as two to four upstream sensors. This study automatically implemented highly accurate machine learning models to predict downstream features within premise plumbing systems, significantly lowering the costs of researching residential plumbing best practices such as water conservation. HIGHLIGHTS Physical end-use monitoring is prohibitively expensive for most practitioners.; Machine learning is a viable method of lowering the cost of end-use categorization.; Here we proposed a novel approach to systematically configure the machine platform automatically.; Effective end-use prediction is possible with a small subset of upstream features.;

Study Region: The Lenguazaque River Basin is a 290 km2 headwater catchment of the Fuquene Lake Watershed in the Colombian Andes. Regional stakeholders are optimizing water allocations for all users, given numerous quantity related issues such as its simultaneous use for conservation, agriculture, and coal mining. Study Focus: Developing hydrological models in headwater catchments is challenging, especially in developing countries where technical resources and data are limited. This study addresses these challenges by proposing strategies to develop reliable yet mathematically simple models, requiring fewer inputs than data-intensive alternatives. Three stages are proposed, focused on 1) assessing hydrological data, 2) preparing datasets for modeling, and 3) developing models in daily and sub-daily resolutions. The last stage is grounded on a data-based mechanistic modeling approach and includes a novel combination of baseflow separation with digital filters and multi-objective optimization principles. Having a hydrophysical meaning, obtained models are suitable for simulating alternative scenarios. New hydrological insights for the region: Resulting daily models achieved an acceptable performance, significantly better than a semi-distributed model. The performance of sub-daily models was sub-optimal, yet it can be improved by enhancing the quality of sub-daily datasets and the efficiency of computational algorithms. Due to their mathematical nature, the proposed strategies are expected to be applicable to other headwater catchments where improving water management is an imperative need.

Crop models are widely used to predict plant growth, water input requirements, and yield. However, existing models are very complex and require hundreds of variables to perform accurately. Due to these shortcomings, large-scale applications of crop models are limited. In order to address these limitations, reliable crop models were developed using a deep neural network (DNN) – a new approach for predicting crop yields. In addition, the number of required input variables was reduced using three common variable selection techniques: namely Bayesian variable selection, Spearman's rank correlation, and Principal Component Analysis Feature Extraction. The reduced-variable DNN models were capable of estimating future crop yields for 10,000,000 different weather and irrigation scenarios while maintaining comparable accuracy levels to the original model that used all input variables. To establish clear superiority of the methodology, the results were also compared with a very recent feature selection algorithm called min-redundancy max-relevance (mRMR). The results of this study showed that the Bayesian variable selection was the best method for achieving the aforementioned goals. Specifically, the final Bayesian-based DNN model with a structure of 10 neurons in 5 layers performed very similarly (78.6% accuracy) to the original DNN crop model with 400 neurons in 10 layers, even though the size of the neural network was reduced by 80-fold. This effort can help promote sustainable agricultural intensifications through the large-scale application of crop models.

River sedimentation is an important indicator for ecological and geomorphological assessments of soil erosion within any watershed region. Sediment transport in a river basin is therefore a multifaceted field yet being a dynamic task in nature. It is characterized by high stochasticity, non-linearity, non-stationarity, and feature redundancy. Various artificial intelligence (AI) modeling frameworks have been introduced to solve river sediment problems. The present survey is designed to provide an updated account of the latest and most relevant AI-based applications for modeling the sediment transport in river basin systems. The review is established to capture the subsequent developments in the advanced AI models applied for river sediment transport prediction. Also, several hydrological and environmental aspects are identified and analyzed according to the results produced in those studies. The merits and constraints of the well-established AI models are further discussed in much detail, particularly considering state-of-the art, modeling frameworks and their application-specific appraisal, and some of the key proposed future research directions. Together with the synthesis of such information to drive a new understanding of models and methodologies related to suspended river sediment prediction, this review provides a future research vision for hydrologists, water scientists, water resource engineers, oceanography and environmental planners.

Livestock farming is a livelihood activity and is critically important for the food and nutritional security of the majority of the population in West African countries, including Senegal. Nevertheless, livestock farming operates far below the optimum production potential, mainly due to demographical, biophysical, economic, environmental, and sociopolitical challenges. To address these issues, we conducted this review with an overall objective of characterizing different livestock farming systems and to identify challenges and opportunities to improve livestock production in West Africa through the broader perspectives from the case of Senegal. Pastoral, agropastoral, and off-land systems are the three major livestock production systems in this region, which are unique in terms of agroclimatology and degree of intensification and integration. The major challenges identified in livestock farming systems are lack of pasture and quality feed, scarcity of water resources, climate change, undeveloped breeding and management of livestock, poor marketing and trade, and socioeconomic constraints. Moreover, we contribute to the literature on crop-livestock farming in Senegal and West Africa by proposing plausible interventions to improve the productivity of the farming system to improve food and nutritional security. Concentrated efforts must be taken in co-designing effective management interventions for sustainable intensification of livestock sector in the region, considering site-specific approaches.

Increasing climate variability and extreme weather events impose significant challenges to the crop production systems throughout the world. Alternative agricultural treatment systems have been proposed to manage these challenges. However, these treatments have not been sufficiently studied for their ability to improve climate resilience, especially in terms of profitability and risk management, which are important metrics of resilience that determine farm-level adaptation. Hence, we evaluated the climate resilience of three alternative agricultural treatments for a long-term (27-years) rotation of corn-soybean-wheat, cast in the temperate humid climate of Southwest Michigan, United States. The three alternative treatments include no-till, reduced input, and US Department of Agriculture (USDA) certified organic. These are compared to the conventional treatment along with the same crop rotation. Means and volatility of expected gross margins and risk preferences were used as the evaluation metrics. Results demonstrate that the net revenues from the USDA certified organic are largely expected to exceed the net revenues of conventional treatment. Also, for all commodities, organic treatment may exert greater annual stability in revenues. The no-till treatment dominates conventional and reduced input practices in expected annual net revenues with a relatively lower risk to those revenues in light of climate extremes. Furthermore, the organic and no-till treatments showed appropriateness to cater to a range of farmers with different risk preferences. Therefore, the organic and the no-till treatments were deemed climate-resilient. The conventional and reduced input treatments did not show resilience thus will be vulnerable to the changing climate. Despite the economic support for adopting resilient practices, growers have been slow to adopt new approaches. We suggest future research needs for understanding grower motivations for adopting climate-resilient practices and consider policy implications.

Global demand for livestock products is expected to double by 2050, mainly due to improvement in the worldwide standard of living. Meanwhile, climate change is a threat to livestock production because of the impact on quality of feed crop and forage, water availability, animal and milk production, livestock diseases, animal reproduction, and biodiversity. This study reviews the global impacts of climate change on livestock production, the contribution of livestock production to climate change, and specific climate change adaptation and mitigation strategies in the livestock sector. Livestock production will be limited by climate variability as animal water consumption is expected to increase by a factor of three, demand for agricultural lands increase due to need for 70% growth in production, and food security concern since about one-third of the global cereal harvest is used for livestock feed. Meanwhile, the livestock sector contributes 14.5% of global greenhouse gas (GHG) emissions, driving further climate change. Consequently, the livestock sector will be a key player in the mitigation of GHG emissions and improving global food security. Therefore, in the transition to sustainable livestock production, there is a need for: a) assessments related to the use of adaptation and mitigation measures tailored to the location and livestock production system in use, and b) policies that support and facilitate the implementation of climate change adaptation and mitigation measures.

Study region: The Kalamazoo River Watershed, southwest Michigan, USA. Study focus: Climate change is projected to have significant impacts on agricultural production. Therefore, understanding the regional impacts of climate change on irrigation demand for crop production is important for watershed managers and agricultural producers to understand for effective water resources management. In this study, the Soil and Water Assessment Tool was used to assess the impact of climate change on corn and soybean irrigation demand in the Kalamazoo River Watershed. Bias-corrected statistically downscaled climate change data from ten global climate models and four emissions scenarios were used in SWAT to develop projections of irrigation demand and yields for 2020–2039 and 2060–2079. Six adaptation scenarios were developed to shift the planting dates (planting earlier and later in the growing season) to take advantage of periods with greater rainfall or lower temperature increases. New hydrological insights for the region: Uncertainty in irrigation demand was found to increase moving from 2020–2039 to 2060–2079, with demand generally decreasing moving further into the future for corn and soybean. A shift in timing of peak irrigation demand and increases in temperature lead to corn yield reductions. However, soybean yield increased under these conditions. Finally, the adaptation strategy of planting earlier increased irrigation demand and water available for transpiration, while delaying planting resulted in demand decreases for both crops. Keywords: Irrigation demand, Climate change, SWAT, Crop yield, Adaptation, Uncertainty