Wheat yield prediction using machine learning and advanced sensing techniques

Stacked ensemble machine learning for phenological wheat yield prediction from UAV-RGB and multispectral satellite data

Agriculture plays an important role in the growth of a country's economy. Crop area and yield predictions using machine learning are important investigation domains in current research fields. Wheat is the most important food crop in Pakistan which is cultivated in the Rabi season. Weather conditions, Remote Sensing (RS) data, and Machine learning (ML) technologies can be used to forecast wheat yield before actual harvesting to assist the management of wheat production, trade, and storage. In this paper, a supervised ML based framework is proposed that extracts features/Vegetation Indices (VIs) including Enhanced Vegetation Index (EVI), Normalized Difference Vegetation Index (NDVI), Red Edge Normalized Difference Vegetation Index (RENDVI), and Normalized Difference Moisture Index (NDMI) from Sentinel-2 Satellite images and contributes for: estimation of wheat area, and identification of most effective VIs in wheat area estimation, prediction of wheat yield, and identification of most effective VIs and meteorological parameters in wheat yield prediction. In the initial experimental setup, good performance output obtained using the Random Forest (RF) machine learning algorithm therefore in this framework RF machine learning algorithm is focused on wheat area estimation and generation of Land Use Land Cover (LULC) maps which is capable of estimating area with an accuracy of 84%, consumer's accuracy of 81 %, producer's accuracy of 83% and kappa statistics of 0.80. LULC maps are used for wheat yield prediction. Multivariate regression forward stepwise technique is applied for yield prediction and selection of effective VIs and meteorological parameters. The adjusted coefficient of determination (R2) between reported and predicted yield found 0.84 with an error of 46.14 Kg/ha for yield prediction.

Objectives: The objective of this study is to improve the accuracy of crop yield prediction models, specifically focusing on wheat and jowar crops in Maharashtra during the Rabi season, by integrating Landsat and Sentinel satellite data with meteorological parameters. Methods: The study utilizes Landsat 8 and Sentinel satellite datasets covering Maharashtra State. Atmospheric correction is applied to extract surface properties, followed by wavelet-based fusion to combine the images. Normalized Difference Vegetation Index (NDVI) is calculated and combined with meteorological parameters using ensemble learning techniques, including Random Forest and Ada-Boost algorithms. Comparative analysis is conducted against existing models, considering parameters such as Mean Absolute Error (MAE), Mean Squared Error (MSE), and Root Mean Squared Error (RMSE). Findings: Significant findings reveal that the proposed methodology outperforms existing models, achieving lower MAE, MSE, and RMSE values for wheat and jowar yield predictions. Additionally, our research highlights the superiority of wheat production over jowar in the Rabi season, based on comprehensive analysis of crop yield predictions. Novelty: This study introduces a novel approach that integrates multiple data sources and employs ensemble learning techniques to enhance crop yield prediction accuracy. By combining Landsat and Sentinel satellite data with meteorological parameters, our methodology provides a more comprehensive understanding of crop growth dynamics, leading to more reliable predictions compared to existing methods. Keywords: Satellite imagery, Machine learning, Normalized Difference Vegetation Index, Fusion, Ensemble learning

Developing machine learning models with multi-source environmental data to predict wheat yield in China

Accurate estimation of wheat yield using Remote Sensing-based models is critical in determining the effects of agricultural drought and sustainable food planning. In this study, Winter wheat yield was estimated for large fields and producer fields by applying Normalized Difference Vegetation Index (NDVI) based linear models (simple linear regression and multiple linear regression) and Machine Learning (ML) techniques (support vector machine_svm, multilayer perceptron_mlp, random forest_rf). In this study, depending on the ecological zone, crop sampling was carried out from 380 rainfed parcels where wheat was planted. On the basis of crop development periods (CDP), the highest correlation between NDVI and yield occurred during the flowering period. In this period, coefficient of determination (R2) was 63% in TIGEM fields and 50% in producer fields for MODIS data, and 61% and 65% for Landsat data, respectively. In TIGEM fields, the best prediction performance was obtained with the MLP model for MODIS (RMSE:0.23-0.65 t/ha) and Landsat (RMSE: 0.28-0.64 t/ha). On the other hand, the highest forecasting accuracy was acquired with the SVM model in producer fields. The RMSE values ranged from 0.74 to 0.80 t/ha for MODIS and 0.51 to 0.60 t/ha for Landsat 8. The error value obtained with MODIS was approximately 1.4 times higher than the Landsat 8 data in producer fields. For yield estimation, the best estimation can be made 4-6 weeks before the harvest. In regional yield estimations, satellite-based ML techniques outperformed linear models. ML models have shown that it can play an important role in crop yield prediction. In crop yield estimation, it is a priority to consider the impact of climate change and ecological differences on crop development.

Core Ideas Normalized difference vegetative index has a stronger correlation with yield than moisture and temperature indices. Optimal winter wheat model includes normalized difference vegetative index, moisture and temperature variables. All three variables have different periods during which they are important. Model can accurately predict winter wheat yield one month before harvest. Gridded data outperform station‐based data for county‐level yield prediction. This article develops a model for predicting winter wheat (Triticum aestivum L.) yield variations in Oklahoma, based on vegetation, moisture, and temperature conditions. A common model structure is identified using stepwise regression with one vegetation indicator (normalized difference vegetative index, NDVI) during wheat jointing and anthesis stages (March and April), one moisture indicator at emergence period (October and November), and one temperature indicator (temperature index, TI) at emergence, jointing and anthesis stages (October, March, and April). The final model accounts for ∼70% of the variation in winter wheat yield and can be used to forecast yields 1 mo before harvest. Spatially, it performs best in the northern and central portions of the Oklahoma winter wheat belt. Model performance is similar regardless of which moisture index is used. The correctly predicted yield variations in at least 9 of the 14 counties every year, and in the best case it correctly predicted yield variation in all counties. Our results also demonstrate that the gridded meteorological data generally outperforms the station‐based data for yield prediction at county level. The methods used in this study can be applied to identify the most significant variables and growth stages for winter wheat yield prediction in other regions.

Annual 30 m winter wheat yield mapping in the Huang-Huai-Hai plain using crop growth model and long-term satellite images

Wheat (Triticum aestivum L.) is one of the world’s primary food crops, and timely and accurate yield prediction is essential for ensuring food security. There has been a growing use of remote sensing, climate data, and their combination to estimate yields, but the optimal indices and time window for wheat yield prediction in arid regions remain unclear. This study was conducted to (1) assess the performance of widely recognized remote sensing indices to predict wheat yield at different growth stages, (2) evaluate the predictive accuracy of different yield predictive machine learning models, (3) determine the appropriate growth period for wheat yield prediction in arid regions, and (4) evaluate the impact of climate parameters on model accuracy. The vegetation indices, widely recognized due to their proven effectiveness, used in this study include the Normalized Difference Vegetation Index (NDVI), the Enhanced Vegetation Index (EVI), and the Atmospheric Resistance Vegetation Index (ARVI). Moreover, four machine learning models, viz. Decision Trees (DTs), Random Forest (RF), Gradient Boosting (GB), and Bagging Trees (BTs), were evaluated to assess their predictive accuracy for wheat yield in the arid region. The whole wheat growth period was divided into three time windows: tillering to grain filling (December 15–March), stem elongation to grain filling (January 15–March), and heading to grain filling (February–March 15). The model was evaluated and developed in the Google Earth Engine (GEE), combining climate and remote sensing data. The results showed that the RF model with ARVI could accurately predict wheat yield at the grain filling and the maturity stages in arid regions with an R2 > 0.75 and yield error of less than 10%. The grain filling stage was identified as the optimal prediction window for wheat yield in arid regions. While RF with ARVI delivered the best results, GB with EVI showed slightly lower precision but still outperformed other models. It is concluded that combining multisource data and machine learning models is a promising approach for wheat yield prediction in arid regions.

Nitrogen and water are two key factors for wheat production due to their major roles in plant growth and development, photosynthesis, yield, and grain protein content. Plant uptake of water and N is fundamentally interactive. Our objectives were: (a) to analyze the effects of different irrigation (IR) and N rates on spring wheat (Triticum aestivum L.) yield and grain protein, and spectral indices (relative greenness [SPAD] and Normalized Difference Vegetation Index [NDVI]), and (b) to identify the optimum IR and N requirements for wheat grain production in semi‐arid conditions of Montana and Idaho. This article details the results from field experiments conducted at three locations for two growing seasons (6 site‐years). Relative greenness measured by SPAD chlorophyll meter was used to assess plant N status, whereas NDVI was used for both plant N status and estimation of wheat yield. Both SPAD and NDVI values increased as N and IR application rates increased. The SPAD and NDVI values explained 80 and 84% of the variation in wheat yield, respectively. We found that IR at 75% of evapotranspiration (ET) throughout the growing season is adequate to optimize wheat yield and grain protein. Nitrogen rate was not correlated with wheat yield at any of the site‐years. Based on this study's results, approximately 150 kg N ha−1 (total, soil residual N plus N added as fertilizer) may be sufficient to optimize yield and grain protein content of irrigated spring wheat in semi‐arid cropping systems.

Applicability of machine learning techniques in predicting wheat yield based on remote sensing and climate data in Pakistan, South Asia

Accurate regional crop classification, acreage estimation, yield prediction, and crop water requirement assessment are essential for effective agricultural planning and market forecasts. This study uses an integrated geospatial and statistical approach to assess major winter crops wheat and sugarcane cultivation in Ghotki District, Pakistan, from 2017/18 to 2022/23. It combines satellite data from Landsat 8 and Sentinel-2, ground truthing, and crop reporting records to analyze key factors such as cultivation area, crop gradients, vegetation health, normalized difference vegetation index (NDVI)-based wheat and sugarcane yield models, crop water requirements, and total irrigation water consumption. Results showed that wheat cultivation areas ranged from 15% to 19%, with the highest coverage observed in the 2021/22 winter season. Sugarcane cultivation ranged from 6% to 10%, peaking in the 2018/19 season. A strong linear association between NDVI and wheat yield (R2 = 0.86) was observed. Wheat and sugarcane yield predictions utilized linear regression, and robust linear regression models, all of which were validated by the findings. Irrigation water demand for the winter season was calculated at 1887 million cubic meters (MCM) in 2017/18, with 1357 MCM supplied by the Sindh Irrigation Drainage Authority (SIDA). By 2020/21, water demand reached 2023 MCM, while SIDA’s supply was 1357 MCM. These results highlight the significance of integrating geospatial analysis with statistical records to provide timely, reliable estimates for cropped areas, yield forecasting, vegetation dynamics, and irrigation planning. The proposed methodology contributes a scaleable solution for informed decision-making in agricultural and water resource management, applicable across other districts in Pakistan and on a global scale.

Exploring the superiority of solar-induced chlorophyll fluorescence data in predicting wheat yield using machine learning and deep learning methods

A MATLAB toolbox for Self Organizing Maps and supervised neural network learning strategies

In general, wheat and maize yields in the Brod-Posavina County (BPC) were about 15% lower (10-year means 1981–1990) in comparison with their yields in the region. Wheat yield variations in the region among the years were higher in comparison with maize yield. For example, the highest yield of wheat and maize were higher than the lowest yield for 61% and 34%, respectively. Analogic comparison for Slav. Brod and N. Gradiska municipalities were 78% and 41% (wheat), as well as 41% and 24% (maize), respectively. In the last 8-year period, mean wheat yields in the region were for 17% lower and maize yield for 4% higher in comparison with mean yields of 80ies. These differences in level of BPC were 10% lower and 12% higher, for wheat and maize, respectively. We presume that low or absence effects of tile drainage because of their inadequate servicing could be expalantion for wheat yield decreasing. The lower yields of wheat are mainly in connection with oversupplies of water. However, low maize yields are in connection with water shortage and the higher air-temperatures. Low supplies of P and K are additional factors of low yields of field crops in the hydromorhic soils of the BPC.

Quantifying the impacts of pre-occurred ENSO signals on wheat yield variation using machine learning in Australia