Crop Yield Models Research Articles

SASYA: An integrated framework for crop biophysical parameter retrieval and within-season crop yield prediction with SAR remote sensing data

Food security can be assured with a reasonable crop yield forecast at a national and regional scale. The agencies require advance estimates of production of major crops at a regional scale for taking various policy decisions. Hence, it is necessary to develop operational systems for crop monitoring and yield forecasting. Unlike the traditional annual crop yield estimates based on the statistics-based framework, remote sensing techniques are recognized as an effective approach for yield forecasting. In particular, the microwave remote sensing data have gained importance due to its all-weather imaging capability and sensitivity to crop phenological changes. In this present work, an integrated framework ‘SASYA-Space technology for Agricultural Systems and Yield forecasting Algorithm’ is introduced for crop yield forecasting using Synthetic Aperture Radar (SAR) remote sensing images. SASYA incorporates a model inversion scheme to estimate crop biophysical parameters (Leaf Area Index–LAI and wet biomass) simultaneously using a multi-output support vector regression (MSVR). The framework cross-link the MSVR based inversion module with a yield prediction module, which utilizes the WOrld FOod STudies (WOFOST) crop simulation model. The robustness of the framework is assessed for estimation of winter barley yield using L-band NISAR data simulated using full-pol L-band E-SAR airborne SAR data acquired during AgriSAR 2006 campaign over DEMMIN test site in Germany. The simulated NISAR data in HH-HV-VV, HH-HV, and VV-HV modes are first compared by investigating accuracies to estimate crop biophysical parameters with the inversion of Water Cloud Model. These LAI and wet biomass estimates are then used in SASYA framework to assimilate with WOFOST crop yield model to estimate grain yield. Different polarization modes HH-HV-VV, HH-HV, and VV-HV are also assessed to investigate best performance in grain yield estimation. Also, different test cases are considered to find the suitable acquisition dates based on crop phenological stages, which returns the best potential yield estimation. The SAR images during heading and early fruit development stage are found to be most important for yield prediction with lowest RMSE error in the range of 0.57–1.612 ton/ha. The LAI, biomass and potential yield map indicate promising results, which can be used further for crop risk assessment for that region.

Using Artificial Neural Networks and Remotely Sensed Data to Evaluate the Relative Importance of Variables for Prediction of Within-Field Corn and Soybean Yields

Crop yield prediction prior to harvest is important for crop income and insurance projections, and for evaluating food security. Yet, modeling crop yield is challenging because of the complexity of the relationships between crop growth and predictor variables, especially at the field scale. In this study, an artificial neural network (ANN) method was used: (1) to evaluate the relative importance of predictor variables for the prediction of within-field corn and soybean end-of-season yield and (2) to evaluate the performance of the ANN models with a minimal optimized variable dataset for their capacity to predict corn and soybean yield over multiple years at the within-field level. Several satellite derived vegetation indices (normalized difference vegetation index—NDVI, red edge NDVI and simple ratio—SR) and elevation derived variables (slope, flow accumulation, aspect) were used as crop yield predictor variables, hypothesizing that the different variables reflect different crop and site conditions. The study identified the SR index and the slope as the most important predictor variables for both crop types during two training and testing years (2011, 2012). The dates of the most important SR images, however, were different for the two crop types and corresponded to their critical crop developmental stages (phenology). The relative mean absolute errors were overall smaller for corn compared to soybean: all of the 2011 corn study fields had errors below 10%; 75% of the fields had errors below 10% in 2012. The errors were more variable for soybean. In 2011, 37% of the fields had errors below 10%, while in 2012, 100% of the fields had errors below 20%. The results are promising and can provide yield estimates at the farm level, which could be useful in refining broader scale (e.g., county, region) yield projections.

Mid-season empirical cotton yield forecasts at fine resolutions using large yield mapping datasets and diverse spatial covariates

Forecasts of crop yield are an important tool for a variety of stakeholders, but most studies produce large-scale, late-season yield forecasts that are not appropriate for farmers. Farmers require forecasts of crop yield mid-season and at fine spatial resolutions to guide site-specific adaptive crop management practices. This study created empirical random forest models to forecast cotton yield mid-season at three different spatial resolutions: 30 m, field, and farm aggregation. Large yield mapping datasets from 14 different seasons were utilised to build the model across 68 different fields from six large cotton farms in eastern Australia. A generic conceptual framework is proposed to empirically model crop yield. This topsaw framework represents the spatial and temporal factors that drive yield, including topography (t), organisms (o), plant measurements (p), soil (s), agronomy/management (a), and weather (w). In this study, the specific predictor variables to represent these driving factors of yield included elevation (t), Normalised Difference Vegetation Index, Enhanced Vegetation Index, MODIS Evapotranspiration (p), gamma radiometrics grid, clay content digital soil map (s), rainfall data, and growing day degrees (w). Forecasts of cotton yield increased in accuracy as the spatial resolution of predictions became coarser. Using a leave-one-year-out cross-validation, cotton yield could be forecasted to an accuracy of 0.42 Lin's Concordance Correlation Coefficient (LCCC) and Root Mean Square Error (RMSE) of 2.11 bales ha−1 at 30 m resolution. This improved to a 0.63 LCCC and 1.72 bales ha−1 RMSE at the field resolution, and a 0.65 LCCC and 0.77 bales ha−1 at the aggregation-scale. As the developed approach solely relied on publicly available predictor variables, there is an opportunity to apply this to any cotton field in Australia, and over a much larger area. Overall, this study is an important step in building an operational approach to forecast mid-season cotton yield at fine spatial resolutions, and has the potential to improve production, input use efficiency, and profitability for individual growers and the Australian Cotton Industry as a whole.

Predicting Soybean Yield at the Regional Scale Using Remote Sensing and Climatic Data

Crop yield modeling at the regional level is one of the most important methods to ensure the profitability of the agro-industrial economy and the solving of the food security problem. Due to a lack of information about crop distribution over large agricultural areas, as well as the crop separation problem (based on remote sensing data) caused by the similarity of phenological cycles, a question arises regarding the relevance of using data obtained from the arable land mask of the region to predict the yield of individual crops. This study aimed to develop a regression model for soybean crop yield monitoring in municipalities and was conducted in the Khabarovsk Territory, located in the Russian Far East. Moderate Resolution Imaging Spectroradiometer (MODIS) data, an arable land mask, the meteorological characteristics obtained using the VEGA-Science web service, and crop yield data for 2010–2019 were used. The structure of crop distribution in the Khabarovsk District was reproduced in experimental fields, and Normalized Difference Vegetation Index (NDVI) seasonal variation approximating functions were constructed (both for total district sown area and different crops). It was found that the approximating function graph for the experimental fields corresponds to a similar graph for arable land. The maximum NDVI forecast error on the 30th week in 2019 using the approximation parameters according to 2014–2018 did not exceed 0.5%. The root-mean-square error (RMSE) was 0.054. The maximum value of the NDVI, as well as the indicators characterizing the temperature regime, soil moisture, and photosynthetically active radiation in the region during the period from the 1st to the 30th calendar weeks of the year, were previously considered as parameters of the regression model for predicting soybean yield. As a result of the experiments, the NDVI and the duration of the growing season were included in the regression model as independent variables. According to 2010–2018, the mean absolute percentage error (MAPE) of the regression model was 6.2%, and the soybean yield prediction absolute percentage error (APE) for 2019 was 6.3%, while RMSE was 0.13 t/ha. This approach was evaluated with a leave-one-year-out cross-validation procedure. When the calculated maximum NDVI value was used in the regression equation for early forecasting, MAPE in the 28th–30th weeks was less than 10%.

Crop biophysical parameter retrieval from Sentinel-1 SAR data with a multi-target inversion of Water Cloud Model

ABSTRACT Estimation of bio-and geophysical parameters from Earth observation (EO) data is essential for developing applications on crop growth monitoring. High spatio-temporal resolution and wide spatial coverage provided by EO satellite data are key inputs for operational crop monitoring. In Synthetic Aperture Radar (SAR) applications, a semi-empirical model (viz., Water Cloud Model (WCM)) is often used to estimate vegetation descriptors individually. However, a simultaneous estimation of these vegetation descriptors would be logical given their inherent correlation, which is seldom preserved in the estimation of individual descriptors by separate inversion models. This functional relationship between biophysical parameters is essential for crop yield models, given that their variations often follow different distribution throughout crop development stages. However, estimating individual parameters with independent inversion models presume a simple relationship (potentially linear) between the biophysical parameters. Alternatively, a multi-target inversion approach would be more effective for this aspect of model inversion compared to an individual estimation approach. In the present research, the multi-output support vector regression (MSVR) technique is used for inversion of the WCM from C-band dual-pol Sentinel-1 SAR data. Plant Area Index (PAI, m2 m−2) and wet biomass (W, kg m−2) are used as the vegetation descriptors in the WCM. The performance of the inversion approach is evaluated with in-situ measurements collected over the test site in Manitoba (Canada), which is a super-site in the Joint Experiment for Crop Assessment and Monitoring (JECAM) SAR inter-comparison experiment network. The validation results indicate a good correlation with acceptable error estimates (normalized root mean square error–nRMSE and mean absolute error–MAE) for both PAI and wet biomass for the MSVR approach and a better estimation with MSVR than single-target models (support vector regression–SVR). Furthermore, the correlation between PAI and wet biomass is assessed using the MSVR and SVR model. Contrary to the single output SVR, the correlation between biophysical parameters is adequately taken into account in MSVR based simultaneous inversion technique. Finally, the spatio-temporal maps for PAI and W at different growth stages indicate their variability with crop development over the test site.

The role of topography, soil, and remotely sensed vegetation condition towards predicting crop yield

Foreknowledge of the spatiotemporal drivers of crop yield would provide a valuable source of information to optimize on-farm inputs and maximize profitability. In recent years, an abundance of spatial data providing information on soils, topography, and vegetation condition have become available from both proximal and remote sensing platforms. Given the wide range of data costs (between USD $0−50/ha), it is important to understand where often limited financial resources should be directed to optimize field production. Two key questions arise. First, will these data actually aid in better fine-resolution yield prediction to help optimize crop management and farm economics? Second, what level of priority should stakeholders commit to in order to obtain these data? Before fully addressing these questions a remaining challenge is the complex nature of spatiotemporal yield variation. Here, a methodological framework is presented to separate the spatial and temporal components of crop yield variation at the subfield level. The framework can also be used to quantify the benefits of different data types on the predicted crop yield as well to better understand the connection of that data to underlying mechanisms controlling yield. Here, fine-resolution (10 m) datasets were assembled for eight 64 ha field sites, spanning a range of climatic, topographic, and soil conditions across Nebraska. Using Empirical Orthogonal Function (EOF) analysis, we found the first axis of variation contained 60–85 % of the explained variance from any particular field, thus greatly reducing the dimensionality of the problem. Using Multiple Linear Regression (MLR) and Random Forest (RF) approaches, we quantified that location within the field had the largest relative importance for modeling crop yield patterns. Secondary factors included a combination of vegetation condition, soil water content, and topography. With respect to predicting spatiotemporal crop yield patterns, we found the RF approach (prediction RMSE of 0.2−0.4 Mg/ha for maize) was superior to MLR (0.3−0.8 Mg/ha). While not directly comparable to MLR and RF the EOF approach had relatively low error (0.5–1.7 Mg/ha) and is intriguing as it requires few calibration parameters (2–6 used here) and utilizes the climate-based aridity index, allowing for pragmatic long-term predictions of subfield crop yield.

A greenhouse climate-yield model focussing on additional light, heat harvesting and its validation

A greenhouse climate-crop yield model was adapted to include additional climate modification techniques suitable for enabling sustainable greenhouse management at high latitudes. Additions to the model were supplementary lighting, secondary heating and heat harvesting technologies. The model: 1) included the impact of different light sources on greenhouse air temperature and tomato production 2) included a secondary heating system 3) calculated the amount of harvested heat whilst lighting was used. The crop yield model was not modified but it was validated for growing tomato in a semi-closed greenhouse equipped with HPS lamps (top-lights) and LED (inter-lights) in Norway. The combined climate-yield model was validated with data from a commercial greenhouse in Norway. The results showed that the model was able to predict the air temperature with sufficient accuracy during the validation periods with Relative Root Mean Square Error

Modeling of Quantiles for Probability Distribution of Crop Yield Under Climate Change (On the Example of Corn)

Introduction. In the context of global warming, there is an urgent need to adapt the agrarian sector to climate change, which, in particular, provides for an adequate choice of crop structure. For this purpose it is necessary to determine which crops are most adapted to the new climatic conditions and to scientifically substantiate their placement in the territory of Ukraine. The traditional approach to crop selection, which consists in conducting field trials of crop response to climate change, is time consuming. An alternative to this approach is application of the methods of mathematical modeling of crop yields in new climatic conditions. The article proposes to use a more flexible approach, namely, the quantile regression method, for modeling yield dependence on climatic parameters, which allows to determine any quantile of the yield distribution function, rather than only one value (average), as in the case of standard regression. The crop yield model based on quantile regression is developed on the grounds of V.P. Dmitrenko model "Weather-harvest" [8], [9]. The following data are used as inputs: (1) corn yields in the context of several areas of the Ukrainian Forest-Steppe in recent years; (2) information on average monthly temperatures and rainfall in these areas in recent years; forecasts of average monthly air temperatures and rainfall in Ukraine for the nearest (by 2030) and more distant (2031 - 2050) perspectives, which are obtained by experts of the Ukrainian Hydrometeorological Institute [10] - [12]. The purpose of the paper is to develop a mathematical model for estimating crop yields that takes into account the uncertainty, associated with climate change in the near and distant perspectives. Results. Using the developed model, estimates of the quantiles of the corn yield distribution function for the nearest (up to 2030) and for the more distant (2031 - 2050) perspectives are obtained both at the level of the individual (Central) region of Ukraine and at the level of the individual (Ternopil) region. The simulation results indicate that weather conditions forecast in [10] - [12] over the next 30 years will more likely produce good corn yields.

Crops pattern change and agricultural diversification: A case study of Domar Upazila, Nilphamari

Cropping pattern means the proportions of area under various crops at a point of time. This paper provides a comprehensive assessment of crop pattern, crop productivity using climate, water and crop yield models. The existing studies present that climate change models with higher spatial resolution can be a way forward for future projections. Various types of statistics, other secondary data and primary questionnaire method was applied to measure and analysis the problem. The cropping pattern differs from macro to micro region, both in space and time and is governed largely by the physical, cultural and technological factors. For the purpose of agricultural regionalization and planning,

CGIAR modeling approaches for resource‐constrained scenarios: II. Models for analyzing socioeconomic factors to improve policy recommendations

AbstractInternational crop‐related research as conducted by the CGIAR uses crop modeling for a variety of purposes. By linking crop models with economic models and approaches, crop model outputs can be effectively used as inputs into socioeconomic modeling efforts for priority setting and policy advice using ex‐ante impact assessment of technologies and scenario analysis. This requires interdisciplinary collaboration and very often collaboration across a variety of research organizations. This study highlights the key topics, purposes, and approaches of socioeconomic analysis within the CGIAR related to cropping systems. Although each CGIAR center has a different mission, all CGIAR centers share a common strategy of striving toward a world free of hunger, poverty, and environmental degradation. This means research is mostly focused toward resource‐constrained smallholder farmers. The review covers global modeling efforts using the IMPACT model to farm household bio‐economic models for assessing the potential impact of new technologies on farming systems and livelihoods. Although the CGIAR addresses all aspects of food systems, the focus of this review is on crop commodities and the economic analysis linked to crop‐growth model results. This study, while not a comprehensive review, provides insights into the richness of the socioeconomic modeling endeavors within the CGIAR. The study highlights the need for interdisciplinary approaches to address the challenges this type of modeling faces.

Effective SAR image segmentation and classification of crop areas using MRG and CDNN techniques

ABSTRACT Crop classification is a significant requirement to estimate crop area, structure, and spatial distribution, as well as provide important input parameters for crop yield models. Different techniques were considered in this system and providing betterment in automation. But none of them gave promising results. So here, a Convolutional Deep Neural Network (CDNN) is proposed to identify the crop areas with the help of Synthetic-Aperture Radar (SAR) satellite images as well as the cultivation status of the crop. First, in training phase, the segmented image of the crop is preprocessed using HLS, then feature is extracted using BRIEF, then, they are classified using CDNN. Then after in testing phase, the input SAR image from the database is further processed using MRG algorithm and classified centered on the training results. After classification, the cultivation status of each classified crop can be identified by taking the Euclidean distance (ED) betwixt the standard parameters and resultant parameters of a specific crop. After computing ED, the ED is contrasted with the threshold value and the cultivation status of a particular crop can be identified. The results are analyzed to ascertain the performance shown by the proposed technique with other existent techniques.

Remote Sensing Applications for Insurance: A Predictive Model for Pasture Yield in the Presence of Systemic Weather

A robust predictive model for crop yield is essential for designing a commercially viable index-based insurance policy. Index-based insurance for crops is still at a relatively infant stage, and more research and development is needed in order to address a main limitation, which is referred to as basis risk. Traditionally, ground weather station measurements have been the most common approach used in weather indices, and this approach has often led to high levels of basis risk. Recent advances in satellite-based remote sensing provide new opportunities to use publicly available and transparent “big data,” to potentially make index-based insurance policies more relevant by reducing basis risk. This is the first article to provide a comprehensive comparison of 13 pasture production indices (PPIs), including those developed based on satellite-derived vegetation and biophysical parameter indices using data products from the Moderate Resolution Imaging Spectroradiometer (MODIS). A validation protocol is established, and a unique dataset covering the period 2002 to 2016 from a network of pasture clip sites in the province of Alberta, Canada, is used to demonstrate new applications for insurance based on remote sensing–derived data. The results of the satellite-derived PPIs are compared to PPIs based on ground weather station data as benchmarks, which to date are the most common design for index-based insurance. Overall, the satellite-based indices report higher correlations with the ground truth forage yield data compared to the weather station PPIs. As an example, in 2015 and 2016 the best performing satellite-based PPIs produced correlations close to 90%. When considering the whole sample period, the highest overall average correlation is 62.0% for the biophysical parameter PPI based on FPAR 500-m MODIS, and the lowest overall average correlation is 43.8% for the vegetation PPI based on EVI 250-m MODIS. Comparatively, the highest correlation reported for the weather station indices is for precipitation, at 12.51%. Other predictive models based on principal component analysis and shrinkage methods (such as lasso, ridge regression, and elastic net) are considered to address the issues of variable selection and dimension reduction in insurance design. This research makes an important contribution to the field of actuarial science and insurance, because it highlights potential new opportunities for insurance design and predictive analytics using large and comprehensive satellite datasets that remain relatively unexplored to date in insurance practice. Though forage is used as the example here, the research could be extended to other crops, and other areas in the property and casualty sector, including for fire and flood.

Determination of the water requirement and crop coefficient values of sugarcane by field water balance method in semiarid region

Sugarcane is a major agro-industrial crop in semiarid regions and generally has high evapotranspiration. Standardized reference evapotranspiration (ET) and location specific crop coefficients are used to estimate crop evapotranspiration. However, precise information on crop coefficient (Kc) is a major impediment in semiarid environments. Field studies were conducted during two seasons of 2015 and 2016 in clay soils to determine crop evapotranspiration and crop coefficients (Kc) of sugarcane for semiarid India.The experimental area was cultivated with irrigation applied at 7–10 days interval by a drip irrigation system in addition to rainfall and the irrigation scheduling was based on field water balance approach. The crop evapotranspiration was determined by field water balance, reference evapotranspiration (ETo) by the Penman-Monteith approach while crop coefficient were computed through the standard FAO-56 methodology.On an annual basis, the total reference evapotranspiration (ETo) and crop evapotranspiration (ETc) were 1318−1426 mm and 1291−1388 mm respectively. Two years average sugarcane crop evapotranspiration estimated by field water balance method was 1339 mm year−1. The irrigation water requirement andeffective rainfall was 991 mm year-1 and 424 mm year-1 respectively. Two years results showed that there was a notable symmetry between Kc obtained from field water balance measurements and FAO-56 reported Kc. The determined Kc values for tillering, grand growth and maturity stages of sugarcane were 0.70, 1.20 and 0.78, respectively. The Kc values were found 25.5 %, 4 % and 20.4 % less during the tillering, grand growth and maturity stage respectively over the FAO-56 Kc values. The 2nd order polynomial equation was fitted with crop coefficient as the dependent variables and ratio of days after transplanting to total crop period as the independent variable. The daily values of Kc from equation is very useful towards efficient management of irrigation water in terms of making a Decision Support System, Soil Moisture based Crop Yield Modeling, Crop Water Requirement based Computer programme or Mobile Application, Automation of irrigation system in the major sugarcane growing countries of semi arid regions.

How many, how far? Quantitative models of Neolithic land use for six wetland sites on the northern Alpine forelands between 4300 and 3700 bc

Dendrochronological studies demonstrate a highly dynamic settlement system in prehistoric wetland sites in the northern Alpine forelands. In this article, we apply an agent-based simulation model of the human–environment system to better understand possible causes of these dynamics. Therefore, we formulate a generic quantitative model of land use and calorie supply in Neolithic wetland sites ca. 4300–3700 bc. Archaeological, geographical and palaeoenvironmental data together with information from an agronomic crop yield model (MONICA) are used in an agent-based simulation of Neolithic land use (WELASSIMO model). We fit the generic model to specific conditions at six archaeological sites and their surrounding environments, using local data. In our simulations, annual crop yields fluctuate markedly around a long term mean which starts to decrease after a few years of crop production. Crop plants supply 60–90% of the annual calorie demand. As sources of readily available non-crop calories are needed to compensate potential low crop yields, we argue that Corylus avellana (hazelnuts) were especially important to provide these extra calories; the simulated importance of non-crop calories is 10–40%. Records of human-induced fires are interpreted as being indicative of a strategy to generally open up the woodland canopy and promote the growth of light-demanding hazel. The extent of the different land use methods is quantified and visualized in tiles of 8 km2 around the six study sites. The specific vegetation cover, the importance of hunting and the number of livestock animals have a major effect on the total area required.

An empirical, Bayesian approach to modelling crop yield: Maize in USA

We apply an empirical, data-driven approach for describing crop yield as a function of monthly temperature and precipitation by employing generative probabilistic models with parameters determined through Bayesian inference. Our approach is applied to state-scale maize yield and meteorological data for the US Corn Belt from 1981 to 2014 as an exemplar, but would be readily transferable to other crops, locations and spatial scales. Experimentation with a number of models shows that maize growth rates can be characterised by a two-dimensional Gaussian function of temperature and precipitation with monthly contributions accumulated over the growing period. This approach accounts for non-linear growth responses to the individual meteorological variables, and allows for interactions between them. Our models correctly identify that temperature and precipitation have the largest impact on yield in the six months prior to the harvest, in agreement with the typical growing season for US maize (April to September). Maximal growth rates occur for monthly mean temperature 18 °C–19 °C, corresponding to a daily maximum temperature of 24 °C–25 °C (in broad agreement with previous work) and monthly total precipitation 115 mm. Our approach also provides a self-consistent way of investigating climate change impacts on current US maize varieties in the absence of adaptation measures. Keeping precipitation and growing area fixed, a temperature increase of 2 °C, relative to 1981–2014, results in the mean yield decreasing by 8%, while the yield variance increases by a factor of around 3. We thus provide a flexible, data-driven framework for exploring the impacts of natural climate variability and climate change on globally significant crops based on their observed behaviour. In concert with other approaches, this can help inform the development of adaptation strategies that will ensure food security under a changing climate.

Statistical emulators of irrigated crop yields and irrigation water requirements

This study provides statistical emulators of global by gridded crop models included in the Inter-Sectoral Impact Model Intercomparison Project Fast Track project to estimate irrigated crop yields and associated irrigation water withdrawals simulated at the grid cell level. An ensemble of crop model simulations is used to build a panel of monthly summer weather variables and corresponding annual yields and irrigation water withdrawals from five gridded crop models. This dataset is then used to estimate crop-specific response functions for each crop model. The average normalized root mean square errors for the response functions range from 3% to 6% for irrigated yields and 2% to 8% for irrigated water withdrawal. Further in- and out-of-sample validation exercises confirm that the statistical emulators are able to replicate the crop models’ spatial patterns of irrigated crop yields and irrigation water withdrawals, both in levels and in terms of changes over time, although accuracy varies by model and by region. The emulators estimated in this study therefore provide a reliable and computationally efficient alternative to global gridded crop yield models.

Soil Erosion Modelling for Sustainable Environmental Management in Sebeya Catchment, Rwanda

Soil erosion models can be understood as a virtual laboratory that brings together data, observations and knowledge from different fields for sustainable environmental management. The present study was carried out on Sebeya catchment which is located in the Western Province of Rwanda. The main objective of this study was to develop a Universal Soil Loss Equation type of erosion model to be used in predicting soil loss and associated crop yields for sustainable agriculture management in Sebeya catchment at the level of parcels. USLE parameters were determined on each parcel in Sebeya catchment using map overlapping techniques as applied in Geographical Information System (GIS). Applying a combination of 0, 1, 2 and 3 soil erosion control measures on each of 259,673 parcels, the simulated annual soil loss for Sebeya catchment was 849.94; 143.27; 88.64 and 28.59 t/ha/yr respectively. Soil Loss and Crop Yield (SOLCY) model has been developed to predict soil loss and crop yields for each main cultivated crop in Sebeya catchment. A combination of 3 soil erosion control measures such as (bench terrace + mulching + drainage channels) has been found to be the most effective in reducing soil erosion on each parcel with slope range of (16 - 60)%. Farmers and agriculture technicians can use SOLCY model. Finally, researchers should develop similar models on other catchments based on SOLCY model design concept.

Modelling crop yield, water consumption, and water use efficiency for sustainable agroecosystem management

Agroecosystems provide food, energy, and essential services, while accounting for 80% of the world’s total water use. Producing more food with limited water resources or improving water use efficiency (WUE) is an urgent task, especially in arid and semi-arid areas with fragile ecosystems and severe water shortages. This study integrated crop type distribution, irrigation data, remote sensing imagery, and meteorological factors to simulate the crop yield, water consumption, and WUE in the Heihe River Basin of China during 2007–2012. We found that variations in water consumption had little impact on crop yield because the value far exceeded the actual water requirement. From 2007 to 2012, the water consumption of corn increased by 11.24%, while its yield decreased by 0.63%. In contrast, the water consumptions of wheat, oilseed rape, and other crops decreased by 15.45%, 10.32%, and 4.61%, while their yields increased by 2.41%, 8.91%, and 1.13%, respectively. Consequently, the WUE of corn decreased by 10.68%, while that of oilseed rape, wheat, barley, and other crops increased by 21.44%, 21.13%, 4.96%, and 6.02%, respectively. The expansion of the area planted with corn led to a 3.36% increase in the WUE of the basin. However, this agricultural development is clearly unsustainable because it increased agricultural water consumption and decreased the ecological and domestic water supply. On-demand irrigation is therefore necessary to reduce water wastage and improve WUE. This would have reduced agricultural water consumption in 2007 and 2012 by 32.58% and 30.13%, and the WUEs would have increased by 48.33% and 43.12%, respectively. Sustainable water-efficient management should comprehensively consider the economic benefits and environmental burdens in crop structure adjustment, irrigation practices, and field management. Additionally, a knowledge-exchange system among experts, resource managers, and farmers is crucial for improving WUE strategies for sustainable agroecosystem development.

Coupling remote sensing with a water balance model for soybean yield predictions over large areas

In this study a new method for predicting soybean yield over large spatial scales, overcoming the difficulties of scalability, is proposed. The method is based on the so-called “simplified triangle” remote sensing technique which is coupled with a crop prediction model of Doorenbos and Kassam 1979 (DK) and the climatological water balance model of Thornthwaite and Mather 1955 (ThM). In the method, surface soil water content (Mo), evapotranspiration (ET), and evaporative fraction (EF) are derived from satellite-derived surface radiant temperature (Ts) and normalized difference vegetation index (NDVI). Use of the proposed method is demonstrated in Brazil’s Parana state for crop years 2002–03 to 2011–12. The soybean crop yield model of DK is evaluated using remotely estimated EF values obtained by a simplified triangle. Predicted crop yield by the satellite measurements and from archived Tropical Rainfall Measuring Mission data (TRMM) and European Centre for Medium-Range Weather Forecasts (ECMWF) data were in good agreement with the measured crop yield. A “d2” index (modified Willmott) between 0.8 and 0.98 and RMSE between 30.8 (kg/ha) to 57.2 (kg/ha) was reported. Crop yield predicted using EF from the triangle were statistically better than the DK and ThM using values of the equivalent of EF obtained from archived surface data when compared with the measured soybean crop data. The proposed method requires no ancillary meteorological or surface data apart from the two satellite images. This makes the technique easy to apply allowing providing spatiotemporal estimates of crop yield in large areas and at different spatial scales requiring little or no surface data.

Probabilistic forecasting of crop yields via quantile random forest and Epanechnikov Kernel function

A reliable and accurate forecasting model for crop yields is of crucial importance for efficient decision-making process in the agricultural sector. However, due to weather extremes and uncertainties, most forecasting models for crop yield are not reliable and accurate. For measuring the uncertainty in crop yield forecast, a probabilistic forecasting model based on quantile random forest and Epanechnikov kernel function (QRF-E) is proposed. The non-linear structure of random forest is applied to build the non-linear quantile regression forecast model and to capture the non-linear relationship betweeen the weather variables and crop yield. . Epanechnikov kernel function and solve-the equation plug-in approach of Sheather and Jones are used in the density estimation. A case study using groundnut and millet yield in Ghana were presented to illustrate the efficiency and robustness of the proposed technique. The values of the prediction interval coverage probability and prediction interval normalized average width for the two crops showed that, the constructed prediction intervals captured the observed yields with high coverage probability. The probability density curves show that QRF-E method has a very high ability to forecast quality prediction intervals with a higher coverage probability. The feature importance gave a score of the importance of each weather variable in building the quantile random forest model. The farmer and other stakeholders are able to realize the specific weather variable that affects the yield of a selected crop through feature importance. The proposed method and its application on crop yield dataset are the first of its kind in literature.