Prediction of Winter Wheat Yield Based on Multi-Source Data and Machine Learning in China

Predicting wheat yield from 2001 to 2020 in Hebei Province at county and pixel levels based on synthesized time series images of Landsat and MODIS

To obtain seasonable and precise crop yield information with fine resolution is very important for ensuring the food security. However, the quantity and quality of available images and the selection of prediction variables often limit the performance of yield prediction. In our study, the synthesized images of Landsat and MODIS were used to provide remote sensing (RS) variables, which can fill the missing values of Landsat images well and cover the study area completely. The deep learning (DL) was used to combine different vegetation index (VI) with climate data to build wheat yield prediction model in Hebei Province (HB). The results showed that kernel NDVI (kNDVI) and near-infrared reflectance (NIRv) slightly outperform normalized difference vegetation index (NDVI) in yield prediction. And the regression algorithm had a more prominent effect on yield prediction, while the yield prediction model using Long Short-Term Memory (LSTM) outperformed the yield prediction model using Light Gradient Boosting Machine (LGBM). The model combining LSTM algorithm and NIRv had the best prediction effect and relatively stable performance in single year. The optimal model was then used to generate 30 m resolution wheat yield maps in the past 20 years, with higher overall accuracy. In addition, we can define the optimum prediction time at April, which can consider simultaneously the performance and lead time. In general, we expect that this prediction model can provide important information to understand and ensure food security.

Machine learning ensembles, neural network, hybrid and sparse regression approaches for weather based rainfed cotton yield forecast

Cotton is a major economic crop predominantly cultivated under rainfed situations. The accurate prediction of cotton yield invariably helps farmers, industries, and policy makers. The final cotton yield is mostly determined by the weather patterns that prevail during the crop growing phase. Crop yield prediction with greater accuracy is possible due to the development of innovative technologies which analyses the bigdata with its high-performance computing abilities. Machine learning technologies can make yield prediction reasonable and faster and with greater flexibility than process based complex crop simulation models. The present study demonstrates the usability of ML algorithms for yield forecasting and facilitates the comparison of different models. The cotton yield was simulated by employing the weekly weather indices as inputs and the model performance was assessed by nRMSE, MAPE and EF values. Results show that stacked generalised ensemble model and artificial neural networks predicted the cotton yield with lower nRMSE, MAPE and higher efficiency compared to other models. Variable importance studies in LASSO and ENET model found minimum temperature and relative humidity as the main determinates of cotton yield in all districts. The models were ranked based these performance metrics in the order of Stacked generalised ensemble > ANN > PCA ANN > SMLR ANN > LASSO> ENET > SVM > PCA SMLR > SMLR SVM > SMLR. This study shows that stacked generalised ensembling and ANN method can be used for reliable yield forecasting at district or county level and helps stakeholders in timely decision-making.

Role of modeling and artificial intelligence in process parameter optimization of biochar: A review

Enhancement of crop yield, conservation and quality upgradation of soil, and efficient water management are the main objectives of sustainable agriculture and mitigating climate change's impact on agriculture. In recent days, biochar, obtained via thermochemical alteration of biomass is becoming a powerful agent for soil and water quality improvement, carbon sequestration, greenhouse gas emission reduction, and heavy metal adsorption. The present study predominantly focuses on various process parameters related to biochar preparation through pyrolysis, their impact on biochar production as well as physicochemical characteristics, and the optimization of such process parameters. Different designs of the experiment (DOE) and optimization techniques including traditional and non-traditional optimizations are discussed in the current review, along with their applicability and shortcomings. Since the biochar preparation process is tedious and energy-consuming, the present review will help to understand the importance of optimization in preparing biochar, thereby leading to a better way to prepare biochar.

Mapping annual 10-m maize cropland changes in China during 2017-2021

China contributed nearly one-fifth of the world maize production over the past few years. Mapping the distributions of maize cropland in China is crucial to ensure global food security. Nonetheless, 10 m maize cropland maps in China are still unavailable, restricting the promotion of sustainable agriculture. In this paper, we collect numerous samples to produce annual 10-m maize cropland maps in China from 2017 to 2021 with a machine learning based classification framework. To overcome the temporal variations of plants, the proposed framework takes Sentinel-2 sequence images as input and utilizes deep neural networks and random forest as classifiers to map maize in a zone-specific way. The generated maps have an overall accuracy (OA) spanning from 0.87 to 0.95 and the maize-cultivated areas estimated by the maps are highly consistent with the records in statistical yearbooks (R2 varying from 0.83 to 0.95). To the best of our knowledge, this is the first annual 10-m maize maps across China, which largely facilitates the sustainable agriculture development in China dominated by smallholder farmlands.

Combining biophysical parameters with thermal and RGB indices using machine learning models for predicting yield in yellow rust affected wheat crop

Evaluating crop health and forecasting yields in the early stages are crucial for effective crop and market management during periods of biotic stress for both farmers and policymakers. Field experiments were conducted during 2017-18 and 2018-19 with objective to evaluate the effect of yellow rust on various biophysical parameters of 24 wheat cultivars, with varying levels of resistance to yellow rust and to develop machine learning (ML) models with improved accuracy for predicting yield by integrating thermal and RGB indices with crucial plant biophysical parameters. Results revealed that as the level of rust increased, so did the canopy temperature and there was a significant decrease in crop photosynthesis, transpiration, stomatal conductance, leaf area index, membrane stability index, relative leaf water content, and normalized difference vegetation index due to rust, and the reductions were directly correlated with levels of rust severity. The yield reduction in moderate resistant, low resistant and susceptible cultivars as compared to resistant cultivars, varied from 15.9-16.9%, 28.6-34.4% and 59-61.1%, respectively. The ML models were able to provide relatively accurate early yield estimates, with the accuracy increasing as the harvest approached. The yield prediction performance of the different ML models varied with the stage of the crop growth. Based on the validation output of different ML models, Cubist, PLS, and SpikeSlab models were found to be effective in predicting the wheat yield at an early stage (55-60 days after sowing) of crop growth. The KNN, Cubist, SLR, RF, SpikeSlab, XGB, GPR and PLS models were proved to be more useful in predicting the crop yield at the middle stage (70 days after sowing) of the crop, while RF, SpikeSlab, KNN, Cubist, ELNET, GPR, SLR, XGB and MARS models were found good to predict the crop yield at late stage (80 days after sowing). The study quantified the impact of different levels of rust severity on crop biophysical parameters and demonstrated the usefulness of remote sensing and biophysical parameters data integration using machine-learning models for early yield prediction under biotically stressed conditions.

Artificial Intelligence in Food Safety: A Decade Review and Bibliometric Analysis

Artificial Intelligence (AI) technologies have been powerful solutions used to improve food yield, quality, and nutrition, increase safety and traceability while decreasing resource consumption, and eliminate food waste. Compared with several qualitative reviews on AI in food safety, we conducted an in-depth quantitative and systematic review based on the Core Collection database of WoS (Web of Science). To discover the historical trajectory and identify future trends, we analysed the literature concerning AI technologies in food safety from 2012 to 2022 by CiteSpace. In this review, we used bibliometric methods to describe the development of AI in food safety, including performance analysis, science mapping, and network analysis by CiteSpace. Among the 1855 selected articles, China and the United States contributed the most literature, and the Chinese Academy of Sciences released the largest number of relevant articles. Among all the journals in this field, PLoS ONE and Computers and Electronics in Agriculture ranked first and second in terms of annual publications and co-citation frequency. The present character, hot spots, and future research trends of AI technologies in food safety research were determined. Furthermore, based on our analyses, we provide researchers, practitioners, and policymakers with the big picture of research on AI in food safety across the whole process, from precision agriculture to precision nutrition, through 28 enlightening articles.

Precision Agriculture Using Soil Sensor Driven Machine Learning for Smart Strawberry Production

Ubiquitous sensor networks collecting real-time data have been adopted in many industrial settings. This paper describes the second stage of an end-to-end system integrating modern hardware and software tools for precise monitoring and control of soil conditions. In the proposed framework, the data are collected by the sensor network distributed in the soil of a commercial strawberry farm to infer the ultimate physicochemical characteristics of the fruit at the point of harvest around the sensor locations. Empirical and statistical models are jointly investigated in the form of neural networks and Gaussian process regression models to predict the most significant physicochemical qualities of strawberry. Color, for instance, either by itself or when combined with the soluble solids content (sweetness), can be predicted within as little as 9% and 14% of their expected range of values, respectively. This level of accuracy will ultimately enable the implementation of the next phase in controlling the soil conditions where data-driven quality and resource-use trade-offs can be realized for sustainable and high-quality strawberry production.

The growing factors impact the productivity of new soft winter wheat varieties

Purpose. To determine the growing factors impact the economically valuable characteristics of new soft winter wheat varieties. Methods. Field, biochemical methods, ANOVA. Results. The influence of the growing zone, the growing season conditions and the genotype of the soft winter wheat varieties on yield, protein and gluten content were determined. The conditions of the growing zone have the greatest influence on studied varieties yield – 73%. On average, for 2020–2021, the maximum yield was obtained in the Forest-Steppe zone – 6.23–8.39 t/ha. In the Forest zone, the yield of studied soft winter wheat varieties was 5.31–7.02 t/ha. For the Steppe zone, the yield was within 5.16–6.63 t/ha. It was determined that varieties which showed low yield in the Forrest zone were characterized by higher yield in the Steppe zone. It was found that the growing zone (49%) and growing season (42%) conditions have the greatest effect on the protein content in grain of studied varieties. It was determined that variety ‘MANDARIN’ was characterized by the highest protein content in all growing zones (14.5–16.4%). In the Steppe and Forest-Steppe zones, high protein content was identified in grain of wheat varieties ‘Tata Mata’ (13.6–14.3%), ‘Eneida’ (13.6–14.3%) and ‘Novator’ (14.2 and 13.2%), in the Forest zone – in grain of varieties ‘Vahoma’ and ‘Eneida’ – 13.4% each. It was found that the growing zone conditions impacted gluten content in grain of studied wheat varieties by 64%, the influence of the growing season conditions was 28%. The effect of variety genotype on protein and gluten content was 5 and 4%, respectively. The grain of soft winter wheat varieties ‘MANDARIN’ and ‘Eneida’ contained the largest amount of gluten in all growing zones (27.7–31.8% and 27.3–30.3%). For the Steppe and Forest-Steppe zones, a high gluten content was found in grains of varieties ‘Illusion’ (28.4 and 30.8%) and ‘Tata Mata’ (27.8 and 29.8%). Conclusions. It was found that growing zone conditions of soft winter wheat varieties (73%) and the interaction of factors zone × year (21%) impacted the yield significantly. The growing zone conditions (49 and 64%), growing season conditions (21 and 28%) and genotype of variety (5 and 4%, respectively) had a significant influence on protein and gluten content. Therefore, a significant influence of growing zone conditions on the productivity indicators of soft winter wheat determines the need to select varieties in order to obtain a high yield, taking into account the soil and climatic zone

Determination of Non-Digestible Parts in Dairy Cattle Feces Using U-NET and F-CRN Architectures

Deep learning algorithms can now be used to identify, locate, and count items in an image thanks to advancements in image processing technology. The successful application of image processing technology in different fields has attracted much attention in the field of agriculture in recent years. This research was done to ascertain the number of indigestible cereal grains in animal feces using an image processing method. In this study, a regression-based way of object counting was used to predict the number of cereal grains in the feces. For this purpose, we have developed two different neural network architectures based upon Fully Convolutional Regression Networks (FCRN) and U-Net. The images used in the study were obtained from three different dairy cows enterprises operating in Nigde Province. The dataset consists of the 277 distinct dropping images of dairy cows in the farm. According to findings of the study, both models yielded quite acceptable prediction accuracy with U-Net providing slightly better prediction with a MAE value of 16.69 in the best case, compared to 23.65 MAE value of FCRN with the same batch.

Characteristics and influencing factors of carbon fluxes in winter wheat fields under elevated CO2 concentration

Elevated carbon dioxide (ECO₂) concentration has profound impacts on ecosystem carbon fluxes, with consequent changes in carbon sequestration and its feedback to climate change. Agroecosystem plays an essential role in global carbon sequestration. However, it is not well understood how the carbon fluxes of agroecosystem respond to increasing atmospheric CO₂ concentrations. In this study, an in-situ 2-year field experiment was conducted using open-top chamber with treatments including ambient CO₂ concentration (CK) and ambient plus 200 μmol mol^-1 (T) to investigate the characteristics and main factors influencing carbon fluxes during the 2017-2019 winter wheat growing seasons. Results showed that the dynamics of CO₂ fluxes under different treatments had similar seasonal trends, with the peak flux observed at the heading-filling stage. Compared to the CK, T treatment increased the cumulative amount of CO₂ (CAC) by 17.2% and 24.0% in 2017-2018 and 2018-2019 growing seasons, respectively. In addition, the seasonal CAC was highly dependent on treatment and varied with year, while there was no interactive effect of treatment and year (p > 0.05). ECO₂ concentration increased the biomass of wheat by an average of 8.28% over two growing seasons. There was a significant positive correlation between biomass and CAC, with biomass elucidating 52% and 76% of the variations in CAC under CK and T treatments, respectively. A good correlation was found between net ecosystem exchange (NEE) and environmental variables under different treatments. During the pre-milk ripening period, the NEE mainly depended on photosynthetically active radiation (PAR) and air temperature (Ta), while NEE was mainly controlled by PAR and soil water content (SWC) during the post-milk ripening period. Overall, the findings presented here demonstrate that the carbon exchange in wheat fields under different treatments serves as carbon sequestration, while ECO₂ concentration enhances the capacity of winter wheat fields to act as carbon sinks, which may have feedback to the climate system in the future.

Novel combination artificial neural network models could not outperform individual models for weather-based cashew yield prediction

Cashew is an important cash crop which is ecologically sensitive, making it vulnerable to climate change. So, the present study compares the performance of stepwise linear regression (SLR), least absolute shrinkage and selection operator (LASSO), elastic net, and artificial neural network (ANN) individually against the ANN model combined with SLR, LASSO, elastic net, and principal components analysis (PCA) for prediction of cashew yield based on weather parameters. The model performances were evaluated using three approaches: (1) Taylor plot; (2) statistical metrics like coefficient of determination (R²), root mean square error (RMSE), and normalized RMSE (nRMSE); and (3) ranking followed by Kruskal-Wallis and Dunn's post hoc test. The results revealed that during calibration, the R² and RMSE ranged from 0.486 to 0.999 and 2.184 to 88.040 kg ha^-1, respectively, while RMSE and nRMSE varied from 3.561 to 242.704 kg ha^-1 and 0.799 to 89.949%, respectively, during validation. Kruskal-Wallis and Dunn's post hoc test revealed LASSO as the best model which was at par with ELNET, SLR, and ELNET-ANN. So, these models can be used for cashew yield prediction for the study area well in advance.

Google Earth Engine and Artificial Intelligence (AI): A Comprehensive Review

Remote sensing (RS) plays an important role gathering data in many critical domains (e.g., global climate change, risk assessment and vulnerability reduction of natural hazards, resilience of ecosystems, and urban planning). Retrieving, managing, and analyzing large amounts of RS imagery poses substantial challenges. Google Earth Engine (GEE) provides a scalable, cloud-based, geospatial retrieval and processing platform. GEE also provides access to the vast majority of freely available, public, multi-temporal RS data and offers free cloud-based computational power for geospatial data analysis. Artificial intelligence (AI) methods are a critical enabling technology to automating the interpretation of RS imagery, particularly on object-based domains, so the integration of AI methods into GEE represents a promising path towards operationalizing automated RS-based monitoring programs. In this article, we provide a systematic review of relevant literature to identify recent research that incorporates AI methods in GEE. We then discuss some of the major challenges of integrating GEE and AI and identify several priorities for future research. We developed an interactive web application designed to allow readers to intuitively and dynamically review the publications included in this literature review.

A Geographically Weighted Random Forest Approach to Predict Corn Yield in the US Corn Belt

Crop yield prediction before the harvest is crucial for food security, grain trade, and policy making. Previously, several machine learning methods have been applied to predict crop yield using different types of variables. In this study, we propose using the Geographically Weighted Random Forest Regression (GWRFR) approach to improve crop yield prediction at the county level in the US Corn Belt. We trained the GWRFR and five other popular machine learning algorithms (Multiple Linear Regression (MLR), Partial Least Square Regression (PLSR), Support Vector Regression (SVR), Decision Tree Regression (DTR), and Random Forest Regression (RFR)) with the following different sets of features: (1) full length features; (2) vegetation indices; (3) gross primary production (GPP); (4) climate data; and (5) soil data. We compared the results of the GWRFR with those of the other five models. The results show that the GWRFR with full length features (R2 = 0.90 and RMSE = 0.764 MT/ha) outperforms other machine learning algorithms. For individual categories of features such as GPP, vegetation indices, climate, and soil features, the GWRFR also outperforms other models. The Moran’s I value of the residuals generated by GWRFR is smaller than that of other models, which shows that GWRFR can better address the spatial non-stationarity issue. The proposed method in this article can also be potentially used to improve yield prediction for other types of crops in other regions.

Wheat Yield Prediction in India Using Principal Component Analysis-Multivariate Adaptive Regression Splines (PCA-MARS)

Crop yield forecasting is becoming more essential in the current scenario when food security must be assured, despite the problems posed by an increasingly globalized community and other environmental challenges such as climate change and natural disasters. Several factors influence crop yield prediction, which has complex non-linear relationships. Hence, to study these relationships, machine learning methodologies have been increasingly adopted from conventional statistical methods. With wheat being a primary and staple food crop in the Indian community, ensuring the country’s food security is crucial. In this paper, we study the prediction of wheat yield for India overall and the top wheat-producing states with a comparison. To accomplish this, we use Multivariate Adaptive Regression Splines (MARS) after extracting the main features by Principal Component Analysis (PCA) considering the parameters such as area under cultivation and production for the years 1962–2018. The performance is evaluated by error analyses such as RMSE, MAE, and R2. The best-fitted MARS model is chosen using cross-validation and user-defined parameter optimization. We find that the MARS model is well suited to India as a whole and other top wheat-producing states. A comparative result is obtained on yield prediction between India overall and other states, wherein the state of Rajasthan has a better model than other major wheat-producing states. This research will emphasize the importance of improved government decision-making as well as increased knowledge and robust forecasting among Indian farmers in various states.

Improving the Forecasting of Winter Wheat Yields in Northern China with Machine Learning–Dynamical Hybrid Subseasonal-to-Seasonal Ensemble Prediction

Subseasonal-to-seasonal (S2S) prediction of winter wheat yields is crucial for farmers and decision-makers to reduce yield losses and ensure food security. Recently, numerous researchers have utilized machine learning (ML) methods to predict crop yield, using observational climate variables and satellite data. Meanwhile, some studies also illustrated the potential of state-of-the-art dynamical atmospheric prediction in crop yield forecasting. However, the potential of coupling both methods has not been fully explored. Herein, we aimed to establish a skilled ML–dynamical hybrid model for crop yield forecasting (MHCF v1.0), which hybridizes ML and a global dynamical atmospheric prediction system, and applied it to northern China at the S2S time scale. In this study, we adopted three mainstream machining learning algorithms (XGBoost, RF, and SVR) and the multiple linear regression (MLR) model, and three major datasets, including satellite data from MOD13C1, observational climate data from CRU, and S2S atmospheric prediction data from IAP CAS, used to predict winter wheat yield from 2005 to 2014, at the grid level. We found that, among the four models examined in this work, XGBoost reached the highest skill with the S2S prediction as inputs, scoring R2 of 0.85 and RMSE of 0.78 t/ha 3–4 months, leading the winter wheat harvest. Moreover, the results demonstrated that crop yield forecasting with S2S dynamical predictions generally outperforms that with observational climate data. Our findings highlighted that the coupling of ML and S2S dynamical atmospheric prediction provided a useful tool for yield forecasting, which could guide agricultural practices, policy-making and agricultural insurance.

Hyperspectral Monitoring of Powdery Mildew Disease Severity in Wheat Based on Machine Learning

Powdery mildew has a negative impact on wheat growth and restricts yield formation. Therefore, accurate monitoring of the disease is of great significance for the prevention and control of powdery mildew to protect world food security. The canopy spectral reflectance was obtained using a ground feature hyperspectrometer during the flowering and filling periods of wheat, and then the Savitzky-Golay method was used to smooth the measured spectral data, and as original reflectivity (OR). Firstly, the OR was spectrally transformed using the mean centralization (MC), multivariate scattering correction (MSC), and standard normal variate transform (SNV) methods. Secondly, the feature bands of above four transformed spectral data were extracted through a combination of the Competitive Adaptive Reweighted Sampling (CARS) and Successive Projections Algorithm (SPA) algorithms. Finally, partial least square regression (PLSR), support vector regression (SVR), and random forest regression (RFR) were used to construct an optimal monitoring model for wheat powdery mildew disease index (mean disease index, mDI). The results showed that after Pearson correlation, two-band optimization combinations and machine learning method modeling comparisons, the comprehensive performance of the MC spectrum data was the best, and it was a better method for pretreating disease spectrum data. The transformed spectral data combined with the CARS-SPA algorithm was able to extract the characteristic bands more effectively. The number of bands screened was more than the number of bands extracted by the OR data, and the band positions were more evenly distributed. In comparison of different machine learning modeling methods, the RFR model performed the best (coefficient of determination, R ² = 0.741-0.852), while the SVR and PLSR models performed similarly (R ² = 0.733-0.836). Taken together, the estimation accuracy of spectral data transformation using the MC method combined with the RFR model (MC-RFR) was the highest, the model R ² was 0.849-0.852, and the root mean square error (RMSE) and the mean absolute error (MAE) ranged from 2.084 to 2.177 and 1.684 to 1.777, respectively. Compared with the OR combined with the RFR model (OR-RFR), the R ² increased by 14.39%, and the R ² of RMSE and MAE decreased by 23.9 and 27.87%. Also, the monitoring accuracy of flowering stage is better than that of grain filling stage, which is due to the relative stability of canopy structure in flowering stage. It can be seen that without changing the shape of the spectral curve, and that the use of MC to preprocess spectral data, the use of CARS and SPA algorithms to extract characteristic bands, and the use of RFR modeling methods to enhance the synergy between multiple variables, and the established model (MC-CARS-SPA-RFR) can better extract the covariant relationship between the canopy spectrum and the disease, thereby improving the monitoring accuracy of wheat powdery mildew. The research results of this study provide ideas and methods for realizing high-precision remote sensing monitoring of crop disease status.

Prediction of Field-Scale Wheat Yield Using Machine Learning Method and Multi-Spectral UAV Data

Accurate prediction of food crop yield is of great significance for global food security and regional trade stability. Since remote sensing data collected from unmanned aerial vehicle (UAV) platforms have the features of flexibility and high resolution, these data can be used as samples to develop regional regression models for accurate prediction of crop yield at a field scale. The primary objective of this study was to construct regional prediction models for winter wheat yield based on multi-spectral UAV data and machine learning methods. Six machine learning methods including Gaussian process regression (GPR), support vector machine regression (SVR) and random forest regression (RFR) were used for the construction of the yield prediction models. Ten vegetation indices (VIs) extracted from canopy spectral images of winter wheat acquired from a multi-spectral UAV at five key growth stages in Xuzhou City, Jiangsu Province, China in 2021 were selected as the variables of the models. In addition, in situ measurements of wheat yield were obtained in a destructive sampling manner for prediction algorithm modeling and validation. Prediction results of single growth stages showed that the optimal model was GPR constructed from extremely strong correlated VIs (ESCVIs) at the filling stage (R2 = 0.87, RMSE = 49.22 g/m2, MAE = 42.74 g/m2). The results of multiple stages showed GPR achieved the highest accuracy (R2 = 0.88, RMSE = 49.18 g/m2, MAE = 42.57 g/m2) when the ESCVIs of the flowering and filling stages were used. Larger sampling plots were adopted to verify the accuracy of yield prediction; the results indicated that the GPR model has strong adaptability at different scales. These findings suggest that using machine learning methods and multi-spectral UAV data can accurately predict crop yield at the field scale and deliver a valuable application reference for farm-scale field crop management.

Kernel Ridge Regression Hybrid Method for Wheat Yield Prediction with Satellite-Derived Predictors

Wheat dominates the Australian grain production market and accounts for 10–15% of the world’s 100 million tonnes annual global wheat trade. Accurate wheat yield prediction is critical to satisfying local consumption and increasing exports regionally and globally to meet human food security. This paper incorporates remote satellite-based information in a wheat-growing region in South Australia to estimate the yield by integrating the kernel ridge regression (KRR) method coupled with complete ensemble empirical mode decomposition with adaptive noise (CEEMDAN) and the grey wolf optimisation (GWO). The hybrid model, ‘GWO-CEEMDAN-KRR,’ employing an initial pool of 23 different satellite-based predictors, is seen to outperform all the benchmark models and all the feature selection (ant colony, atom search, and particle swarm optimisation) methods that are implemented using a set of carefully screened satellite variables and a feature decomposition or CEEMDAN approach. A suite of statistical metrics and infographics comparing the predicted and measured yield shows a model prediction error that can be reduced by ~20% by employing the proposed GWO-CEEMDAN-KRR model. With the metrics verifying the accuracy of simulations, we also show that it is possible to optimise the wheat yield to achieve agricultural profits by quantifying and including the effects of satellite variables on potential yield. With further improvements in the proposed methodology, the GWO-CEEMDAN-KRR model can be adopted in agricultural yield simulation that requires remote sensing data to establish the relationships between crop health, yield, and other productivity features to support precision agriculture.

Evaluation of Random Forests (RF) for Regional and Local-Scale Wheat Yield Prediction in Southeast Australia

Wheat accounts for more than 50% of Australia's total grain production. The capability to generate accurate in-season yield predictions is important across all components of the agricultural value chain. The literature on wheat yield prediction has motivated the need for more novel works evaluating machine learning techniques such as random forests (RF) at multiple scales. This research applied a Random Forest Regression (RFR) technique to build regional and local-scale yield prediction models at the pixel level for three southeast Australian wheat-growing paddocks, each located in Victoria (VIC), New South Wales (NSW) and South Australia (SA) using 2018 yield maps from data supplied by collaborating farmers. Time-series Normalized Difference Vegetation Index (NDVI) data derived from Planet's high spatio-temporal resolution imagery, meteorological variables and yield data were used to train, test and validate the models at pixel level using Python libraries for (a) regional-scale three-paddock composite and (b) individual paddocks. The composite region-wide RF model prediction for the three paddocks performed well (R² = 0.86, RMSE = 0.18 t ha^-1). RF models for individual paddocks in VIC (R² = 0.89, RMSE = 0.15 t ha^-1) and NSW (R² = 0.87, RMSE = 0.07 t ha^-1) performed well, but moderate performance was seen for SA (R² = 0.45, RMSE = 0.25 t ha^-1). Generally, high values were underpredicted and low values overpredicted. This study demonstrated the feasibility of applying RF modeling on satellite imagery and yielded 'big data' for regional as well as local-scale yield prediction.

Using UAV and Multispectral Images to Estimate Peanut Maturity Variability on Irrigated and Rainfed Fields Applying Linear Models and Artificial Neural Networks

Using UAV and multispectral images has contributed to identifying field variability and improving crop management through different data modeling methods. However, knowledge on application of these tools to manage peanut maturity variability is still lacking. Therefore, the objective of this study was to compare and validate linear and multiple linear regression with models using artificial neural networks (ANN) for estimating peanut maturity under irrigated and rainfed conditions. The models were trained (80% dataset) and tested (20% dataset) using results from the 2018 and 2019 growing seasons from irrigated and rainfed fields. In each field, plant reflectance was collected weekly from 90 days after planting using a UAV-mounted multispectral camera. Images were used to develop vegetation indices (VIs). Peanut pods were collected on the same dates as the UAV flights for maturity assessment using the peanut maturity index (PMI). The precision and accuracy of the linear models to estimate PMI using VIs were, in general, greater in irrigated fields with R2 > 0.40 than in rainfed areas, which had a maximum R2 value of 0.21. Multiple linear regressions combining adjusted growing degree days (aGDD) and VIs resulted in decreased RMSE for both irrigated and rainfed conditions and increased R2 in irrigated areas. However, these models did not perform successfully in the test process. On the other hand, ANN models that included VIs and aGDD showed accuracy of R2 = 0.91 in irrigated areas, regardless of using Multilayer Perceptron (MLP; RMSE = 0.062) or Radial Basis Function (RBF; RMSE = 0.065), as well as low tendency (1:1 line). These results indicated that, regardless of the ANN architecture used to predict complex and non-linear variables, peanut maturity can be estimated accurately through models with multiple inputs using VIs and aGDD. Although the accuracy of the MLP or RBF models for irrigated and rainfed areas separately was high, the overall ANN models using both irrigated and rainfed areas can be used to predict peanut maturity with the same precision.

Monitoring Wheat Powdery Mildew Based on Hyperspectral, Thermal Infrared, and RGB Image Data Fusion

Powdery mildew severely affects wheat growth and yield; therefore, its effective monitoring is essential for the prevention and control of the disease and global food security. In the present study, a spectroradiometer and thermal infrared cameras were used to obtain hyperspectral signature and thermal infrared images data, and thermal infrared temperature parameters (TP) and texture features (TF) were extracted from the thermal infrared images and RGB images of wheat with powdery mildew, during the wheat flowering and filling periods. Based on the ten vegetation indices from the hyperspectral data (VI), TF and TP were integrated, and partial least square regression, random forest regression (RFR), and support vector machine regression (SVR) algorithms were used to construct a prediction model for a wheat powdery mildew disease index. According to the results, the prediction accuracy of RFR was higher than in other models, under both single data source modeling and multi-source data modeling; among the three data sources, VI was the most suitable for powdery mildew monitoring, followed by TP, and finally TF. The RFR model had stable performance in multi-source data fusion modeling (VI&TP&TF), and had the optimal estimation performance with 0.872 and 0.862 of R² for calibration and validation, respectively. The application of multi-source data collaborative modeling could improve the accuracy of remote sensing monitoring of wheat powdery mildew, and facilitate the achievement of high-precision remote sensing monitoring of crop disease status.

Maize Yield Prediction at an Early Developmental Stage Using Multispectral Images and Genotype Data for Preliminary Hybrid Selection

Assessing crop production in the field often requires breeders to wait until the end of the season to collect yield-related measurements, limiting the pace of the breeding cycle. Early prediction of crop performance can reduce this constraint by allowing breeders more time to focus on the highest-performing varieties. Here, we present a multimodal deep learning model for predicting the performance of maize (Zea mays) at an early developmental stage, offering the potential to accelerate crop breeding. We employed multispectral images and eight vegetation indices, collected by an uncrewed aerial vehicle approximately 60 days after sowing, over three consecutive growing cycles (2017, 2018 and 2019). The multimodal deep learning approach was used to integrate field management and genotype information with the multispectral data, providing context to the conditions that the plants experienced during the trial. Model performance was assessed using holdout data, in which the model accurately predicted the yield (RMSE 1.07 t/ha, a relative RMSE of 7.60% of 16 t/ha, and R2 score 0.73) and identified the majority of high-yielding varieties, outperforming previously published models for early yield prediction. The inclusion of vegetation indices was important for model performance, with a normalized difference vegetation index and green with normalized difference vegetation index contributing the most to model performance. The model provides a decision support tool, identifying promising lines early in the field trial.

Predicting Maize Yield at the Plot Scale of Different Fertilizer Systems by Multi-Source Data and Machine Learning Methods

Timely and reliable maize yield prediction is essential for the agricultural supply chain and food security. Previous studies using either climate or satellite data or both to build empirical or statistical models have prevailed for decades. However, to what extent climate and satellite data can improve yield prediction is still unknown. In addition, fertilizer information may also improve crop yield prediction, especially in regions with different fertilizer systems, such as cover crop, mineral fertilizer, or compost. Machine learning (ML) has been widely and successfully applied in crop yield prediction. Here, we attempted to predict maize yield from 1994 to 2007 at the plot scale by integrating multi-source data, including monthly climate data, satellite data (i.e., vegetation indices (VIs)), fertilizer data, and soil data to explore the accuracy of different inputs to yield prediction. The results show that incorporating all of the datasets using random forests (RF) and AB (adaptive boosting) can achieve better performances in yield prediction (R2: 0.85~0.98). In addition, the combination of VIs, climate data, and soil data (VCS) can predict maize yield more effectively than other combinations (e.g., combinations of all data and combinations of VIs and soil data). Furthermore, we also found that including different fertilizer systems had different prediction accuracies. This paper aggregates data from multiple sources and distinguishes the effects of different fertilization scenarios on crop yield predictions. In addition, the effects of different data on crop yield were analyzed in this study. Our study provides a paradigm that can be used to improve yield predictions for other crops and is an important effort that combines multi-source remotely sensed and environmental data for maize yield prediction at the plot scale and develops timely and robust methods for maize yield prediction grown under different fertilizing systems.

The Role of Earth Observation in Achieving Sustainable Agricultural Production in Arid and Semi-Arid Regions of the World

Crop production is a major source of food and livelihood for many people in arid and semi-arid (ASA) regions across the world. However, due to irregular climatic events, ASA regions are affected commonly by frequent droughts that can impact food production. In addition, ASA regions in the Middle East and Africa are often characterised by political instability, which can increase population vulnerability to hunger and ill health. Remote sensing (RS) provides a platform to improve the spatial prediction of crop production and food availability, with the potential to positively impact populations. This paper, firstly, describes some of the important characteristics of agriculture in ASA regions that require monitoring to improve their management. Secondly, it demonstrates how freely available RS data can support decision-making through a cost-effective monitoring system that complements traditional approaches for collecting agricultural data. Thirdly, it illustrates the challenges of employing freely available RS data for mapping and monitoring crop area, crop status and forecasting crop yield in these regions. Finally, existing approaches used in these applications are evaluated, and the challenges associated with their use and possible future improvements are discussed. We demonstrate that agricultural activities can be monitored effectively and both crop area and crop yield can be predicted in advance using RS data. We also discuss the future challenges associated with maintaining food security in ASA regions and explore some recent advances in RS that can be used to monitor cropland and forecast crop production and yield.

Cereal Yield Forecasting with Satellite Drought-Based Indices, Weather Data and Regional Climate Indices Using Machine Learning in Morocco

Accurate seasonal forecasting of cereal yields is an important decision support tool for countries, such as Morocco, that are not self-sufficient in order to predict, as early as possible, importation needs. This study aims to develop an early forecasting model of cereal yields (soft wheat, barley and durum wheat) at the scale of the agricultural province considering the 15 most productive over 2000–2017 (i.e., 15 × 18 = 270 yields values). To this objective, we built on previous works that showed a tight linkage between cereal yields and various datasets including weather data (rainfall and air temperature), regional climate indices (North Atlantic Oscillation in particular), and drought indices derived from satellite observations in different wavelengths. The combination of the latter three data sets is assessed to predict cereal yields using linear (Multiple Linear Regression, MLR) and non-linear (Support Vector Machine, SVM; Random Forest, RF, and eXtreme Gradient Boost, XGBoost) machine learning algorithms. The calibration of the algorithmic parameters of the different approaches are carried out using a 5-fold cross validation technique and a leave-one-out method is implemented for model validation. The statistical metrics of the models are first analyzed as a function of the input datasets that are used, and as a function of the lead times, from 4 months to 2 months before harvest. The results show that combining data from multiple sources outperformed models based on one dataset only. In addition, the satellite drought indices are a major source of information for cereal prediction when the forecasting is carried out close to harvest (2 months before), while weather data and, to a lesser extent, climate indices, are key variables for earlier predictions. The best models can accurately predict yield in January (4 months before harvest) with an R2 = 0.88 and RMSE around 0.22 t. ha−1. The XGBoost method exhibited the best metrics. Finally, training a specific model separately for each group of provinces, instead of one global model, improved the prediction performance by reducing the RMSE by 10% to 35% depending on the provinces. In conclusion, the results of this study pointed out that combining remote sensing drought indices with climate and weather variables using a machine learning technique is a promising approach for cereal yield forecasting.

Wheat Yield Prediction Based on Unmanned Aerial Vehicles-Collected Red–Green–Blue Imagery

Unmanned aerial vehicles-collected (UAVs) digital red–green–blue (RGB) images provided a cost-effective method for precision agriculture applications regarding yield prediction. This study aims to fully explore the potential of UAV-collected RGB images in yield prediction of winter wheat by comparing it to multi-source observations, including thermal, structure, volumetric metrics, and ground-observed leaf area index (LAI) and chlorophyll content under the same level or across different levels of nitrogen fertilization. Color indices are vegetation indices calculated by the vegetation reflectance at visible bands (i.e., red, green, and blue) derived from RGB images. The results showed that some of the color indices collected at the jointing, flowering, and early maturity stages had high correlation (R2 = 0.76–0.93) with wheat grain yield. They gave the highest prediction power (R2 = 0.92–0.93) under four levels of nitrogen fertilization at the flowering stage. In contrast, the other measurements including canopy temperature, volumetric metrics, and ground-observed chlorophyll content showed lower correlation (R2 = 0.52–0.85) to grain yield. In addition, thermal information as well as volumetric metrics generally had little contribution to the improvement of grain yield prediction when combining them with color indices derived from digital images. Especially, LAI had inferior performance to color indices in grain yield prediction within the same level of nitrogen fertilization at the flowering stage (R2 = 0.00–0.40 and R2 = 0.55–0.68), and color indices provided slightly better prediction of yield than LAI at the flowering stage (R2 = 0.93, RMSE = 32.18 g/m2 and R2 = 0.89, RMSE = 39.82 g/m2) under all levels of nitrogen fertilization. This study highlights the capabilities of color indices in wheat yield prediction across genotypes, which also indicates the potential of precision agriculture application using many other flexible, affordable, and easy-to-handle devices such as mobile phones and near surface digital cameras in the future.

Bottlenecks and opportunities in field-based high-throughput phenotyping for heat and drought stress

Flowering and grain-filling stages are highly sensitive to heat and drought stress exposure, leading to significant loss in crop yields. Therefore, phenotyping to enhance resilience to these abiotic stresses is critical for sustaining genetic gains in crop improvement programs. However, traditional methods for screening traits related to these stresses are slow, laborious, and often expensive. Remote sensing provides opportunities to introduce low-cost, less biased, high-throughput phenotyping methods to capture large genetic diversity to facilitate enhancement of stress resilience in crops. This review focuses on four key physiological traits and processes that are critical in understanding crop responses to drought and heat stress during reproductive and grain-filling periods. Specifically, these traits include: (i) time of day of flowering, to escape these stresses during flowering; (ii) optimizing photosynthetic efficiency; (iii) storage and translocation of water-soluble carbohydrates; and (iv) yield and yield components to provide in-season yield estimates. Moreover, we provide an overview of current advances in remote sensing in capturing these traits, and discuss the limitations with existing technology as well as future direction of research to develop high-throughput phenotyping approaches. In the future, phenotyping these complex traits will require sensor advancement, high-quality imagery combined with machine learning methods, and efforts in transdisciplinary science to foster integration across disciplines.

Selection of Independent Variables for Crop Yield Prediction Using Artificial Neural Network Models with Remote Sensing Data

Knowing the expected crop yield in the current growing season provides valuable information for farmers, policy makers, and food processing plants. One of the main benefits of using reliable forecasting tools is generating more income from grown crops. Information on the amount of crop yielding before harvesting helps to guide the adoption of an appropriate strategy for managing agricultural products. The difficulty in creating forecasting models is related to the appropriate selection of independent variables. Their proper selection requires a perfect knowledge of the research object. The following article presents and discusses the most commonly used independent variables in agricultural crop yield prediction modeling based on artificial neural networks (ANNs). Particular attention is paid to environmental variables, such as climatic data, air temperature, total precipitation, insolation, and soil parameters. The possibility of using plant productivity indices and vegetation indices, which are valuable predictors obtained due to the application of remote sensing techniques, are analyzed in detail. The paper emphasizes that the increasingly common use of remote sensing and photogrammetric tools enables the development of precision agriculture. In addition, some limitations in the application of certain input variables are specified, as well as further possibilities for the development of non-linear modeling, using artificial neural networks as a tool supporting the practical use of and improvement in precision farming techniques.

Machine Learning in Agriculture: A Comprehensive Updated Review

The digital transformation of agriculture has evolved various aspects of management into artificial intelligent systems for the sake of making value from the ever-increasing data originated from numerous sources. A subset of artificial intelligence, namely machine learning, has a considerable potential to handle numerous challenges in the establishment of knowledge-based farming systems. The present study aims at shedding light on machine learning in agriculture by thoroughly reviewing the recent scholarly literature based on keywords' combinations of "machine learning" along with "crop management", "water management", "soil management", and "livestock management", and in accordance with PRISMA guidelines. Only journal papers were considered eligible that were published within 2018-2020. The results indicated that this topic pertains to different disciplines that favour convergence research at the international level. Furthermore, crop management was observed to be at the centre of attention. A plethora of machine learning algorithms were used, with those belonging to Artificial Neural Networks being more efficient. In addition, maize and wheat as well as cattle and sheep were the most investigated crops and animals, respectively. Finally, a variety of sensors, attached on satellites and unmanned ground and aerial vehicles, have been utilized as a means of getting reliable input data for the data analyses. It is anticipated that this study will constitute a beneficial guide to all stakeholders towards enhancing awareness of the potential advantages of using machine learning in agriculture and contributing to a more systematic research on this topic.

Crop Yield Estimation Using Deep Learning Based on Climate Big Data and Irrigation Scheduling

Deep learning has already been successfully used in the development of decision support systems in various domains. Therefore, there is an incentive to apply it in other important domains such as agriculture. Fertilizers, electricity, chemicals, human labor, and water are the components of total energy consumption in agriculture. Yield estimates are critical for food security, crop management, irrigation scheduling, and estimating labor requirements for harvesting and storage. Therefore, estimating product yield can reduce energy consumption. Two deep learning models, Long Short-Term Memory and Gated Recurrent Units, have been developed for the analysis of time-series data such as agricultural datasets. In this paper, the capabilities of these models and their extensions, called Bidirectional Long Short-Term Memory and Bidirectional Gated Recurrent Units, to predict end-of-season yields are investigated. The models use historical data, including climate data, irrigation scheduling, and soil water content, to estimate end-of-season yield. The application of this technique was tested for tomato and potato yields at a site in Portugal. The Bidirectional Long Short-Term memory outperformed the Gated Recurrent Units network, the Long Short-Term Memory, and the Bidirectional Gated Recurrent Units network on the validation dataset. The model was able to capture the nonlinear relationship between irrigation amount, climate data, and soil water content and predict yield with an MSE of 0.017 to 0.039. The performance of the Bidirectional Long Short-Term Memory in the test was compared with the most commonly used deep learning method, the Convolutional Neural Network, and machine learning methods including a Multi-Layer Perceptrons model and Random Forest Regression. The Bidirectional Long Short-Term Memory outperformed the other models with an R2 score between 0.97 and 0.99. The results show that analyzing agricultural data with the Long Short-Term Memory model improves the performance of the model in terms of accuracy. The Convolutional Neural Network model achieved the second-best performance. Therefore, the deep learning model has a remarkable ability to predict the yield at the end of the season.

Enhancing Coffee Supply Chain towards Sustainable Growth with Big Data and Modern Agricultural Technologies

Modern agricultural technology management is nowadays crucial in terms of the economy and the global market, while food safety, quality control, and environmentally friendly practices should not be neglected. This review aims to give perspectives on applying big data analytic and modern technologies to increase the efficacy and effectiveness of the coffee supply chain throughout the process. It was revealed that several tools such as wireless sensor networks, cloud computing, Internet of Things (IoT), image processing, convolutional neural networks (CNN), and remote sensing could be implemented in and used to improve the coffee supply chain. Those tools could help in reducing cost as well as time for entrepreneurs and create a reliable service for the customer. It can be summarized that in the long term, these modern technologies will be able to assist coffee business management and ensure the sustainable growth for the coffee industry.

Wheat Yellow Rust Detection Using UAV-Based Hyperspectral Technology

Yellow rust is a worldwide disease that poses a serious threat to the safety of wheat production. Numerous studies on near-surface hyperspectral remote sensing at the leaf scale have achieved good results for disease monitoring. The next step is to monitor the disease at the field scale, which is of great significance for disease control. In our study, an unmanned aerial vehicle (UAV) equipped with a hyperspectral sensor was used to obtain hyperspectral images at the field scale. Vegetation indices (VIs) and texture features (TFs) extracted from the UAV-based hyperspectral images and their combination were used to establish partial least-squares regression (PLSR)-based disease monitoring models in different infection periods. In addition, we resampled the original images with 1.2 cm spatial resolution to images with different spatial resolutions (3 cm, 5 cm, 7 cm, 10 cm, 15 cm, and 20 cm) to evaluate the effect of spatial resolution on disease monitoring accuracy. The findings showed that the VI-based model had the highest monitoring accuracy (R2 = 0.75) in the mid-infection period. The TF-based model could be used to monitor yellow rust at the field scale and obtained the highest R2 in the mid- and late-infection periods (0.65 and 0.82, respectively). The VI-TF-based models had the highest accuracy in each infection period and outperformed the VI-based or TF-based models. The spatial resolution had a negligible influence on the VI-based monitoring accuracy, but significantly influenced the TF-based monitoring accuracy. Furthermore, the optimal spatial resolution for monitoring yellow rust using the VI-TF-based model in each infection period was 10 cm. The findings provide a reference for accurate disease monitoring using UAV hyperspectral images.

Modified Red Blue Vegetation Index for Chlorophyll Estimation and Yield Prediction of Maize from Visible Images Captured by UAV

The vegetation index (VI) has been successfully used to monitor the growth and to predict the yield of agricultural crops. In this paper, a long-term observation was conducted for the yield prediction of maize using an unmanned aerial vehicle (UAV) and estimations of chlorophyll contents using SPAD-502. A new vegetation index termed as modified red blue VI (MRBVI) was developed to monitor the growth and to predict the yields of maize by establishing relationships between MRBVI- and SPAD-502-based chlorophyll contents. The coefficients of determination (R2s) were 0.462 and 0.570 in chlorophyll contents' estimations and yield predictions using MRBVI, and the results were relatively better than the results from the seven other commonly used VI approaches. All VIs during the different growth stages of maize were calculated and compared with the measured values of chlorophyll contents directly, and the relative error (RE) of MRBVI is the lowest at 0.355. Further, machine learning (ML) methods such as the backpropagation neural network model (BP), support vector machine (SVM), random forest (RF), and extreme learning machine (ELM) were adopted for predicting the yields of maize. All VIs calculated for each image captured during important phenological stages of maize were set as independent variables and the corresponding yields of each plot were defined as dependent variables. The ML models used the leave one out method (LOO), where the root mean square errors (RMSEs) were 2.157, 1.099, 1.146, and 1.698 (g/hundred grain weight) for BP, SVM, RF, and ELM. The mean absolute errors (MAEs) were 1.739, 0.886, 0.925, and 1.356 (g/hundred grain weight) for BP, SVM, RF, and ELM, respectively. Thus, the SVM method performed better in predicting the yields of maize than the other ML methods. Therefore, it is strongly suggested that the MRBVI calculated from images acquired at different growth stages integrated with advanced ML methods should be used for agricultural- and ecological-related chlorophyll estimation and yield predictions.

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%.

Quantifying Uncertainty and Bridging the Scaling Gap in the Retrieval of Leaf Area Index by Coupling Sentinel-2 and UAV Observations

Leaf area index (LAI) estimates can inform decision-making in crop management. The European Space Agency’s Sentinel-2 satellite, with observations in the red-edge spectral region, can monitor crops globally at sub-field spatial resolutions (10–20 m). However, satellite LAI estimates require calibration with ground measurements. Calibration is challenged by spatial heterogeneity and scale mismatches between field and satellite measurements. Unmanned Aerial Vehicles (UAVs), generating high-resolution (cm-scale) LAI estimates, provide intermediary observations that we use here to characterise uncertainty and reduce spatial scaling discrepancies between Sentinel-2 observations and field surveys. We use a novel UAV multispectral sensor that matches Sentinel-2 spectral bands, flown in conjunction with LAI ground measurements. UAV and field surveys were conducted on multiple dates—coinciding with different wheat growth stages—that corresponded to Sentinel-2 overpasses. We compared chlorophyll red-edge index (CIred-edge) maps, derived from the Sentinel-2 and UAV platforms. We used Gaussian processes regression machine learning to calibrate a UAV model for LAI, based on ground data. Using the UAV LAI, we evaluated a two-stage calibration approach for generating robust LAI estimates from Sentinel-2. The agreement between Sentinel-2 and UAV CIred-edge values increased with growth stage—R2 ranged from 0.32 (stem elongation) to 0.75 (milk development). The CIred-edge variance between the two platforms was more comparable later in the growing season due to a more homogeneous and closed wheat canopy. The single-stage Sentinel-2 LAI calibration (i.e., direct calibration from ground measurements) performed poorly (mean R2 = 0.29, mean NRMSE = 17%) when compared to the two-stage calibration using the UAV data (mean R2 = 0.88, mean NRMSE = 8%). The two-stage approach reduced both errors and biases by >50%. By upscaling ground measurements and providing more representative model training samples, UAV observations provide an effective and viable means of enhancing Sentinel-2 wheat LAI retrievals. We anticipate that our UAV calibration approach to resolving spatial heterogeneity would enhance the retrieval accuracy of LAI and additional biophysical variables for other arable crop types and a broader range of vegetation cover types.

Winter Wheat Yield Prediction at County Level and Uncertainty Analysis in Main Wheat-Producing Regions of China with Deep Learning Approaches

Timely and accurate forecasting of crop yields is crucial to food security and sustainable development in the agricultural sector. However, winter wheat yield estimation and forecasting on a regional scale still remains challenging. In this study, we established a two-branch deep learning model to predict winter wheat yield in the main producing regions of China at the county level. The first branch of the model was constructed based on the Long Short-Term Memory (LSTM) networks with inputs from meteorological and remote sensing data. Another branch was constructed using Convolution Neural Networks (CNN) to model static soil features. The model was then trained using the detrended statistical yield data during 1982 to 2015 and evaluated by leave-one-year-out-validation. The evaluation results showed a promising performance of the model with the overall R 2 and RMSE of 0.77 and 721 kg/ha, respectively. We further conducted yield prediction and uncertainty analysis based on the two-branch model and obtained the forecast accuracy in one month prior to harvest of 0.75 and 732 kg/ha. Results also showed that while yield detrending could potentially introduce higher uncertainty, it had the advantage of improving the model performance in yield prediction.