Accurate Yield Estimation Research Articles

Wheat Yield Estimation at High Spatial Resolution through the Assimilation of Sentinel-2 Data into a Crop Growth Model

Monitoring crop growth and estimating crop yield are essential for managing agricultural production, ensuring food security, and maintaining sustainable agricultural development. Combining the mechanistic framework of a crop growth model with remote sensing observations can provide a means of generating realistic and spatially detailed crop growth information that can facilitate accurate crop yield estimates at different scales. The main objective of this study was to develop a robust estimation methodology of within-field winter wheat yield at a high spatial resolution (20 m × 20 m) by combining a light use efficiency-based model and Sentinel-2 data. For this purpose, Sentinel-2 derived leaf area index (LAI) time series were assimilated into the Simple Algorithm for Yield Estimation (SAFY) model using an ensemble Kalman filter (EnKF). The study was conducted on rainfed winter wheat fields in southern Sweden. LAI was estimated using vegetation indices (VIs) derived from Sentinel-2 data with semi-empirical models. The enhanced two-band vegetation index (EVI2) was found to be a useful VI for LAI estimation, with a coefficient of determination (R2) and a root mean square error (RMSE) of 0.80 and 0.65 m2/m2, respectively. Our findings demonstrate that the assimilation of LAI derived from Sentinel-2 into the SAFY model using EnKF enhances the estimation of within-field spatial variability of winter wheat yield by 70% compared to the baseline simulation without the assimilation of remotely sensed data. Additionally, the assimilation of LAI improves the accuracy of winter wheat yield estimation by decreasing the RMSE by 53%. This study demonstrates an approach towards practical applications of freely accessible Sentinel-2 data and a crop growth model through data assimilation for fine-scale mapping of crop yield. Such information is critical for quantifying the yield gap at the field scale, and to aid the optimization of management practices to increase crop production.

Estimation of Soybean Yield by Combining Maturity Group Information and Unmanned Aerial Vehicle Multi-Sensor Data Using Machine Learning

Accurate and rapid estimation of the crop yield is essential to precision agriculture. Critical to crop improvement, yield is a primary index for selecting excellent genotypes in crop breeding. Recently developed unmanned aerial vehicle (UAV) platforms and advanced algorithms can provide powerful tools for plant breeders. Genotype category information such as the maturity group information (M) can significantly influence soybean yield estimation using remote sensing data. The objective of this study was to improve soybean yield prediction by combining M with UAV-based multi-sensor data using machine learning methods. We investigated three types of maturity groups (Early, Median and Late) of soybean, and collected the UAV-based hyperspectral and red–green–blue (RGB) images at three key growth stages. Vegetation indices (VI) and texture features (Te) were extracted and combined with M to predict yield using partial least square regression (PLSR), Gaussian process regression (GPR), random forest regression (RFR) and kernel ridge regression (KRR). The results showed that (1) the method of combining M with remote sensing data could significantly improve the estimation performances of soybean yield. (2) The combinations of three variables (VI, Te and M) gave the best estimation accuracy. Meanwhile, the flowering stage was the optimal single time point for yield estimation (R2 = 0.689, RMSE = 408.099 kg/hm2), while using multiple growth stages produced the best estimation performance (R2 = 0.700, RMSE = 400.946 kg/hm2). (3) By comparing the models constructed by different algorithms for different growth stages, it showed that the models built by GPR showed the best performances. Overall, the results of this study provide insights into soybean yield estimation based on UAV remote sensing data and maturity information.

Spatial Yield Assessment of Sorghum (Sorghum Bicolor) Using DSSAT – CERES –Sorghum Model in Solapur, Maharashtra, India

Accurate estimation of crop yield is crucial for ensuring food security and effective policy making. This study focuses on the estimation of sorghum yield in the Solapur region of Maharashtra (India), employing the Decision Support System for Agrotechnology Transfer (DSSAT) model. Sorghum is the fifth largely produced staple crop of the world which also plays a vital role in the food security produced by India. Maharashtra has the largest area under sorghum crop, and Solapur has the most area under rabi sorghum with an area of 4.6 lakh ha, accounting for 23% of the total area under rabi sorghum in the state. Although productivity is lower in Maharashtra than in other states, these studies will help us to get a preharvest estimate of the crop. Crop Cutting Experiments(CCE) were conducted for rabi sorghum and the model was validated for the simulated yields; which have a range of grain yield from 611 to 1525 kgs ha-1 and showed error with less than 14% and it was evaluated with statistical models such R2, Nash-Sutcliffe efficiency (NSE) and Normalized Root Mean Square Error (NRMSE) and results show as 84%, 0.84 and 0.07. This model can be used further used for the yield gap analysis, and climate change studies for the locations.

A Global Systematic Review of Improving Crop Model Estimations by Assimilating Remote Sensing Data: Implications for Small-Scale Agricultural Systems

There is a growing effort to use access to remote sensing data (RS) in conjunction with crop model simulation capability to improve the accuracy of crop growth and yield estimates. This is critical for sustainable agricultural management and food security, especially in farming communities with limited resources and data. Therefore, the objective of this study was to provide a systematic review of research on data assimilation and summarize how its application varies by country, crop, and farming systems. In addition, we highlight the implications of using process-based crop models (PBCMs) and data assimilation in small-scale farming systems. Using a strict search term, we searched the Scopus and Web of Science databases and found 497 potential publications. After screening for relevance using predefined inclusion and exclusion criteria, 123 publications were included in the final review. Our results show increasing global interest in RS data assimilation approaches; however, 81% of the studies were from countries with relatively high levels of agricultural production, technology, and innovation. There is increasing development of crop models, availability of RS data sources, and characterization of crop parameters assimilated into PBCMs. Most studies used recalibration or updating methods to mainly incorporate remotely sensed leaf area index from MODIS or Landsat into the WOrld FOod STudies (WOFOST) model to improve yield estimates for staple crops in large-scale and irrigated farming systems. However, these methods cannot compensate for the uncertainties in RS data and crop models. We concluded that further research on data assimilation using newly available high-resolution RS datasets, such as Sentinel-2, should be conducted to significantly improve simulations of rare crops and small-scale rainfed farming systems. This is critical for informing local crop management decisions to improve policy and food security assessments.

Precision Detection of Dense Litchi Fruit in UAV Images Based on Improved YOLOv5 Model

The utilization of unmanned aerial vehicles (UAVs) for the precise and convenient detection of litchi fruits, in order to estimate yields and perform statistical analysis, holds significant value in the complex and variable litchi orchard environment. Currently, litchi yield estimation relies predominantly on manual rough counts, which often result in discrepancies between the estimated values and the actual production figures. This study proposes a large-scene and high-density litchi fruit recognition method based on the improved You Only Look Once version 5 (YOLOv5) model. The main objective is to enhance the accuracy and efficiency of yield estimation in natural orchards. First, the PANet in the original YOLOv5 model is replaced with the improved Bi-directional Feature Pyramid Network (BiFPN) to enhance the model’s cross-scale feature fusion. Second, the P2 feature layer is fused into the BiFPN to enhance the learning capability of the model for high-resolution features. After that, the Normalized Gaussian Wasserstein Distance (NWD) metric is introduced into the regression loss function to enhance the learning ability of the model for litchi tiny targets. Finally, the Slicing Aided Hyper Inference (SAHI) is used to enhance the detection of tiny targets without increasing the model’s parameters or computational memory. The experimental results show that the overall AP value of the improved YOLOv5 model has been effectively increased by 22%, compared to the original YOLOv5 model’s AP value of 50.6%. Specifically, the APs value for detecting small targets has increased from 27.8% to 57.3%. The model size is only 3.6% larger than the original YOLOv5 model. Through ablation and comparative experiments, our method has successfully improved accuracy without compromising the model size and inference speed. Therefore, the proposed method in this paper holds practical applicability for detecting litchi fruits in orchards. It can serve as a valuable tool for providing guidance and suggestions for litchi yield estimation and subsequent harvesting processes. In future research, optimization can be continued for the small target detection problem, while it can be extended to study the small target tracking problem in dense scenarios, which is of great significance for litchi yield estimation.

Enhanced Integrator with Drift Elimination for Accurate Flux Estimation in Sensorless Controlled Interior PMSM for High-Performance Full Speed Range Hybrid Electric Vehicles Applications

Interior Permanent Magnet Synchronous Motor (IPMSM) motion-sensorless speed control necessitates precise knowledge of rotor flux, speed, and position. Due to numerous non-ideal aspects, such as converter nonlinearities, detection errors, integral initial value, and parameter mismatches, the conventional first-order integrator’s estimated rotor flux experiences a DC offset (Doff). Low-pass filters (LPF) with a constant cut-off frequency yield accurate estimates only in the medium- and high-speed range; however, at the low-speed area, both magnitude and phase estimates are inaccurate. The presented technique resolves the aforementioned issue for a broad speed range. In order to achieve precise flux estimation, this article presents an improved technique of flux estimator with two distinct drift mitigation strategies for the motion-sensorless field-oriented control (FOC) system of IPMSM. Using the orthogonality of the α- and β-axes, the proposed drift elimination system can estimate drift in different situations while maintaining a high level of dynamic performance. The stator flux linkage (SFL) computation in the synchronous coordinate is established from the estimation of the rotating shaft’s permanent magnetic flux linkage orientation and the statistical equations model of the SFL. By comparing the calculated SFL vector to the SFL vector derived from the stator winding voltage and currents integral model with a drift PI compensation loop, a feedback loop is formed to neutralize integral drift, and the rotational speed and position of an IPMSM is estimated utilizing the vector product of the two flux linkages in a phase-locked loop. Theoretical interpretation is presented, and Matlab Simulink simulations, as well as experimental outcomes, consistently demonstrate that the suggested estimation techniques can eliminate the phenomenon of flux drift.

Erosion and Sediment Transport Modeling: A Systematic Review

Soil erosion and sediment transport have significant consequences, including decreased agricultural production, water quality degradation, and modification to stream channels. Understanding these processes and their interactions with contributing factors is crucial for assessing the environmental impacts of erosion. The primary objective of this review is to identify a suitable soil erosion and sediment transport model for catchment-scale application. The study considers various model selection processes, including model capability and the spatial and temporal domains for assessing spatiotemporal distributions. The review acknowledges the limitations, uncertainties, and unrealistic assumptions associated with soil erosion and sediment transport models. Models are usually developed with a particular objective, which demands an assessment of capabilities, spatial, and temporal applicability, and catchment-scale applicability. Distributed models are often preferred for catchment-scale applications, as they can adequately account for spatial variations in erosion potential and sediment yield, aiding in the evaluation of erosion-contributing elements and planning erosion control measures. Based on the findings of this study, the authors encourage utilizing models (such as Soil and Water Assessment Tool (SWAT) or Automated Geospatial Watershed Assessment Tool (AGWA)) that can forecast net erosion as a function of sediment output for catchment erosion and sediment yield modeling. This review helps researchers and practitioners involved in erosion and sediment modeling by guiding the selection of an appropriate model type based on specific modeling purposes and basin scale. By choosing appropriate models, the accuracy and effectiveness of sediment yield estimation and erosion control measures can be improved.

Simulation of Colored BIPV Modules Using Angular-Dependent Spectral Responsivity

An accurate energy yield estimation for building-integrated photovoltaic (PV) applications with colored glass under field conditions is a key factor for architects, building planners, and potential investors. The power output of these installations depends not only on the angle of incidence but also on the incident solar spectrum. The aim of this work is to simulate the short-circuit current of PV single-cell laminates with colored solar glass under field conditions. Therefore, we use angular-dependent spectral responsivity measurement data from a physical setup. Test samples with gold-, blue-, blue-green-, and gray-colored solar glass and a reference with standard solar glass, oriented on a tilted surface, are simulated. The results for one month of field validation indicate a good agreement with a relative root mean square error between 1.9% and 2.5% for the test samples. A model comparison reveals larger errors for the blue, blue-green, and gold test samples when the spectral responsivity at normal incidence and the incident angle modifier for broadband direct irradiation are used instead. It is concluded that short-circuit current simulation using indoor characterization from spectrally resolved measurement setups provides improved accuracy that is particularly suitable for bankable energy yield estimates of colored building-integrated PV modules.

Estimating wheat grain yield by assimilating phenology and LAI with the WheatGrow model based on theoretical uncertainty of remotely sensed observation

Assimilating remotely sensed observation into crop model solves the model parameterization in regional scale by considering the uncertainty of observation and simulation. However, a fixed value of observation uncertainty could not describe the spatio-temporal heterogeneity of observation error. This study developed a method to estimate wheat yield in China by assimilating remotely sensed observation into WheatGrow model. A dynamic background value of WheatGrow and a theoretical uncertainty of remotely sensed observation were proposed, and a strategy assimilating phenology and LAI hierarchically was applied at county and plot scales. At the county scale, phenology was estimated using the time sequence of PlanetScope images, sowing date was estimated using one-class support vector classification, heading and maturity date were estimated using function-fitting method. The revisiting cycle of PlanetScope was used to determine the theoretical uncertainty of remotely sensed observed sowing date (RS-sowing date), while the uncertainty in RS-heading and RS-maturity dates were calculated using field measurements. LAI was estimated using Sentinel-2 images. The theoretical uncertainty of remotely sensed observed LAI (RS-LAI) was developed using a synthetic dataset. At the plot scale, RS-heading date and RS-LAI were estimated using the same method as for the county scale based on images obtained with a RedEdge-MX camera carried by drone. The results indicated that the theoretical uncertainty sufficiently estimated the trend of LAI error with canopy density. The assimilation strategies based on theoretical uncertainty relieved the error propagation caused by poor parameterization of model background value and the RS observation error in dense canopy, outperforming the results based on constant uncertainty. After hierarchical assimilation based on theoretical uncertainty, the root mean square errors of phenology, LAI and yield were 4.3 d, 0.75, and 825 kg/ha at the county scale, and 4.1 d, 0.57 and 969 kg/ha at the plot scale. These results showed that the hierarchical assimilation based on theoretical uncertainty could improve the accuracy of yield estimation.

Large global variations in the carbon dioxide removal potential of seaweed farming due to biophysical constraints

Estimates suggest that over 4 gigatons per year of carbon dioxide (Gt-CO2 year−1) be removed from the atmosphere by 2050 to meet international climate goals. One strategy for carbon dioxide removal is seaweed farming; however its global potential remains highly uncertain. Here, we apply a dynamic seaweed growth model that includes growth-limiting mechanisms, such as nitrate supply, to estimate the global potential yield of four types of seaweed. We estimate that harvesting 1 Gt year−1 of seaweed carbon would require farming over 1 million km2 of the most productive exclusive economic zones, located in the equatorial Pacific; the cultivation area would need to be tripled to attain an additional 1 Gt year−1 of harvested carbon, indicating dramatic reductions in carbon harvest efficiency beyond the most productive waters. Improving the accuracy of annual harvest yield estimates requires better understanding of biophysical constraints such as seaweed loss rates (e.g., infestation, disease, grazing, wave erosion).

Analysis of unit level models for small area estimation in crop statistics assisted with satellite auxiliary information

Crop statistics for a small area, such as the community development block, are an increasingly important topic in agricultural statistics. Under normality assumptions, the classic Empirical Best Linear Unbiased Prediction (EBLUP) technique is effective for predicting small area means, however the Small Area Estimation (SAE) model can be heavily affected by the incidence of outliers or deviations from the expected distribution. The purpose of this study was to estimate variance, predict block-level wheat crop yield in the Hisar and Sirsa district of Haryana by classical SAE method and a robust random-effect predictor using a slight generalization of Huber’s Proposal 2. In the case of Sirsa district, the results of classical and robust unit level SAE were very close, but not in the case of Hisar district. This could be due to the influential observation found in the Hisar data set. More accurate EBLUP wheat yield estimates are obtained when the Huber-type M-estimation method is initialized by the least square regression estimator.

Faba Bean (Vicia faba L.) Yield Estimation Based on Dual-Sensor Data

Faba bean is an important member of legumes, which has richer protein levels and great development potential. Yield is an important phenotype character of crops, and early yield estimation can provide a reference for field inputs. To facilitate rapid and accurate estimation of the faba bean yield, the dual-sensor (RGB and multi-spectral) data based on unmanned aerial vehicle (UAV) was collected and analyzed. For this, support vector machine (SVM), ridge regression (RR), partial least squares regression (PLS), and k-nearest neighbor (KNN) were used for yield estimation. Additionally, the fusing data from different growth periods based on UAV was first used for estimating faba bean yield to obtain better estimation accuracy. The results obtained are as follows: for a single-growth period, S2 (12 July 2019) had the best accuracy of the estimation model. For fusion data from the muti-growth period, S2 + S3 (12 August 2019) obtained the best estimation results. Furthermore, the coefficient of determination (R2) values for RF were higher than other machine learning algorithms, followed by PLS, and the estimation effects of fusion data from a dual-sensor were evidently better than from a single sensor. In a word, these results indicated that it was feasible to estimate the faba bean yield with high accuracy through data fusion based on dual-sensor data and different growth periods.

Comparison of PlanetScope, Sentinel-2, and landsat 8 data in soybean yield estimation within-field variability with random forest regression

Accurate timely and early-season crop yield estimation within the field variability is important for precision farming and sustainable management applications. Therefore, the ability to estimate the within-field variability of grain yield is crucial for ensuring food security worldwide, especially under climate change. Several Earth observation systems have thus been developed to monitor crops and predict yields. Despite this, new research is required to combine multiplatform data integration, advancements in satellite technologies, data processing, and the application of this discipline to agricultural practices. This study provides further developments in soybean yield estimation by comparing multisource satellite data from PlanetScope (PS), Sentinel-2 (S2), and Landsat 8 (L8) and introducing topographic and meteorological variables. Herein, a new method of combining soybean yield, global positioning systems, harvester data, climate, topographic variables, and remote sensing images has been demonstrated. Soybean yield shape points were obtained from a combine-harvester-installed GPS and yield monitoring system from seven fields over the 2021 season. The yield estimation models were trained and validated using random forest, and four vegetation indices were tested. The result showed that soybean yield can be accurately predicted at 3-, 10-, and 30-m resolutions with mean absolute error (MAE) value of 0.091 t/ha for PS, 0.118 t/ha for S2, and 0.120 t/ha for L8 data (root mean square error (RMSE) of 0.111, 0.076). The combination of the environmental data with the original bands provided further improvements and an accurate yield estimation model within the soybean yield variability with MAE of 0.082 t/ha for PS, 0.097 t/ha for S2, and 0.109 t/ha for L8 (RMSE of 0.094, 0.069, and 0.108 t/ha). The results showed that the optimal date to predict the soybean yield within the field scale was approximately 60 or 70 days before harvesting periods during the beginning bloom stage. The developed model can be applied for other crops and locations when suitable training yield data, which are critical for precision farming, are available.

Adaboost CNN with Horse Herd Optimization Algorithm to Forecast the Rice Crop Yield

Over three billion people use rice every day, and it occupies about 12% of the nation's arable land. Since, due to the growing population and the latest climate change projections, it is critical for governments and planners to obtain timely and accurate rice yield estimates. The proposed work develops a rice crop yield forecasting model based on soil nutrients. Soil nutrients and crop production statistics are taken as an input for the proposed method. In ensemble learning, there are three categories, they are Boosting, Bagging and Stacking. In the proposed method, Boosting technique called Adaboost with Convolutional Neural Network is used to achieve the High accuracy by converting weak classifiers to strong classifiers. Adaptive data cleaning and imputation using frequent values are used as pre-processing approaches in the projected technique. A novel technique known as Convolutional neural network with adaptive boosting (Adaboost) technique is projected and can precisely handle more imbalanced datasets. The data weights are initialized; also the initial CNN is trained utilizing original weights of data. The weights of the second CNN are then modified utilizing the first CNN. These actions will be performed sequentially for all weak classifiers. An optimization algorithm called Horse Herd (HOA) is passed down in the proposed technique to find the optimal weights of the links in the classifier. The proposed method attains 95% accuracy, 87% precision, 85% recall, 5% error, 96% specificity, 87% F1-Score, 97% NPV and 12% FNR value.Thus the designed model as predicted the crop yield prediction in the effective manner.

Rapid Rice Yield Estimation Using Integrated Remote Sensing and Meteorological Data and Machine Learning

This study developed a rapid rice yield estimation workflow and customized yield prediction model by integrating remote sensing and meteorological data with machine learning (ML). Several issues need to be addressed while developing a crop yield estimation model, including data quality issues, data processing issues, selecting a suitable machine learning model that can learn from few available time-series data, and understanding the non-linear relationship between historical crop yield and remote sensing and meteorological factors. This study applied a series of data processing techniques and a customized ML model to improve the accuracy of crop yield estimation at the district level in Nepal. It was found that remote sensing-derived NDVI product alone was not sufficient for accurate estimation of crop yield. After incorporating other meteorological variables into the ML models, estimation accuracy improved dramatically. Along with NDVI, the meteorological variables of rainfall, soil moisture, and evapotranspiration also exhibited a strong association with rice yield. This study also found that stacking multiple tree-based regression models together could achieve better accuracy than benchmark linear regression or standalone ML models. Due to the unique and distinct physio-geographical setting of each district, a variation in estimation accuracy from district to district could be observed. Our data processing and ML model workflow achieved an average of 92% accuracy of yield estimation with RMSE 328.06 kg/ha and MAE 317.21 kg/ha. This methodological workflow can be replicated in other study areas and the results can help the local authorities and stakeholders understand the factors affecting crop yields as well as estimating crop yield before harvesting season to ensure food security and sustainability.

Natural groundwater recharge estimation using multiple methods combined with an experimental study

Abstract Due to increasing human settlements, agricultural development, and economic activities, the extraction of groundwater is increasing. The over-exploitation of aquifers has become a cause for concern within the last century due to these developmental activities and the accurate estimation of sustainable yield from aquifers is very important. Assessing natural aquifer recharge is one of the key challenges in determining the sustainable yield of aquifers. In this study, an integrated model using the SCS-CN method, remote sensing, and a Geographic Information System (GIS) is developed for estimating natural aquifer recharge for the recharge area of the sub-basin of the Vellar river basin, Tamil Nadu, India. Also, actual infiltration measurements were conducted and an infiltration-based model was developed to compute natural groundwater recharge and compare it with empirical methods. On average, the recharge is 17% of the rainfall for the recharge area.

Developing machine learning models for wheat yield prediction using ground-based data, satellite-based actual evapotranspiration and vegetation indices

Timely and accurate crop yield estimation is important for adjusting agronomic management and enseuring agricultural sustainability. Machine learning (ML) algorithms provide new opportunities to integrate agronomic information with ground-based and satellite data and develop flexible yield predictive models. In particular, satellite-based vegetation indices and evapotranspiration provide robust proxies for crop yield estimations in the absence of measurements; nevertheless, most prior model development efforts have focused on using only vegetation indices due to the simplicity of the process. Additionally, the contribution of input categories (i.e., field, meteorological, and satellite data) and the use of appropriate proxies, aligned with the crop growth stages, in developing yield predictive models have not been adequately investigated. To address these challenges, we employed two ML techniques, Random Forest (RF) and extreme gradient boosting algorithm (XGB), to estimate wheat yield using meteorological variables, satellite-driven actual evapotranspiration (ETa), and vegetation indices (VIs). ETa was separately computed using the surface energy balance concept and the METRIC model. The models were first trained and tested in the study area using three input combinations: i) meteorological variables, ii) satellite data, and iii) an ensemble of meteorological and satellite data. Then, the best-performing model was further evaluated using two independent datasets. We found ETa to be particularly important in improving the accuracy of the model predictions. Among the vegetation indices, EVI, EVI2, and NDVI during May, and among the meteorological data, growing degree days during the grain filling stage plus minimum temperature in the stem elongation stage had the highest contributions to yield predictions. Both ML algorithms generated relatively accurate results, where XGB was marginally more accurate than RF, considering an average mean absolute error of 0.39 t ha-1 for XGB and 0.50 t ha-1 for RF. Normalized root-mean-square errors of the ensemble, satellite-derived and meteorological-derived models in XGB were 0.05, 0.07, and 0.10, respectively. Nevertheless, both algorithms’ performances deteriorated in predicting the yield values beyond the range of the training set, though XGB could handle the extrapolation process more efficiently than RF.

An Improved Approach of Winter Wheat Yield Estimation by Jointly Assimilating Remotely Sensed Leaf Area Index and Soil Moisture into the WOFOST Model

The crop model data assimilation approach has been acknowledged as an effective tool for monitoring crop growth and estimating yield. However, the choice of assimilated variables and the mismatch in scale between remotely sensed observations and crop model-simulated state variables have various effects on the performance of yield estimation. This study aims to examine the accuracy of crop yield estimation through the joint assimilation of leaf area index (LAI) and soil moisture (SM) and to examine the scale effect between remotely sensed data and crop model simulations. To address these issues, we proposed an improved crop data-model assimilation (CDMA) framework, which integrates LAI and SM, as retrieved from remotely sensed data, into the World Food Studies (WOFOST) model using the ensemble Kalman filter (EnKF) approach for winter wheat yield estimation. The results showed that the yield estimation at a 10 m grid size outperformed that at a 500 m grid size, using the same assimilation strategy. Additionally, the winter wheat yield estimation accuracy was higher when using the bivariate data assimilation method (R2 = 0.46, RMSE = 756 kg/ha) compared to the univariate method. In conclusion, our study highlights the advantages of joint assimilating LAI and SM for crop yield estimation and emphasizes the importance of finer spatial resolution in remotely sensed observations for crop yield estimation using the CDMA framework. The proposed approach would help to develop a high-accuracy crop yield monitoring system using optical and SAR retrieved parameters.

Impact of STARFM on Crop Yield Predictions: Fusing MODIS with Landsat 5, 7, and 8 NDVIs in Bavaria Germany

Rapid and accurate yield estimates at both field and regional levels remain the goal of sustainable agriculture and food security. Hereby, the identification of consistent and reliable methodologies providing accurate yield predictions is one of the hot topics in agricultural research. This study investigated the relationship of spatiotemporal fusion modelling using STRAFM on crop yield prediction for winter wheat (WW) and oil-seed rape (OSR) using a semi-empirical light use efficiency (LUE) model for the Free State of Bavaria (70,550 km2), Germany, from 2001 to 2019. A synthetic normalised difference vegetation index (NDVI) time series was generated and validated by fusing the high spatial resolution (30 m, 16 days) Landsat 5 Thematic Mapper (TM) (2001 to 2012), Landsat 7 Enhanced Thematic Mapper Plus (ETM+) (2012), and Landsat 8 Operational Land Imager (OLI) (2013 to 2019) with the coarse resolution of MOD13Q1 (250 m, 16 days) from 2001 to 2019. Except for some temporal periods (i.e., 2001, 2002, and 2012), the study obtained an R2 of more than 0.65 and a RMSE of less than 0.11, which proves that the Landsat 8 OLI fused products are of higher accuracy than the Landsat 5 TM products. Moreover, the accuracies of the NDVI fusion data have been found to correlate with the total number of available Landsat scenes every year (N), with a correlation coefficient (R) of +0.83 (between R2 of yearly synthetic NDVIs and N) and −0.84 (between RMSEs and N). For crop yield prediction, the synthetic NDVI time series and climate elements (such as minimum temperature, maximum temperature, relative humidity, evaporation, transpiration, and solar radiation) are inputted to the LUE model, resulting in an average R2 of 0.75 (WW) and 0.73 (OSR), and RMSEs of 4.33 dt/ha and 2.19 dt/ha. The yield prediction results prove the consistency and stability of the LUE model for yield estimation. Using the LUE model, accurate crop yield predictions were obtained for WW (R2 = 0.88) and OSR (R2 = 0.74). Lastly, the study observed a high positive correlation of R = 0.81 and R = 0.77 between the yearly R2 of synthetic accuracy and modelled yield accuracy for WW and OSR, respectively.

Optimizing Wheat Yield Prediction Integrating Data from Sentinel-1 and Sentinel-2 with CatBoost Algorithm

Accurately estimating wheat yield is crucial for informed decision making in precision agriculture (PA) and improving crop management. In recent years, optical satellite-derived vegetation indices (Vis), such as Sentinel-2 (S2), have become widely used, but the availability of images depends on the weather conditions. For its part, Sentinel-1 (S1) backscatter data are less used in agriculture due to its complicated interpretation and processing, but is not impacted by weather. This study investigates the potential benefits of combining S1 and S2 data and evaluates the performance of the categorical boosting (CatBoost) algorithm in crop yield estimation. The study was conducted utilizing dense yield data from a yield monitor, obtained from 39 wheat (Triticum spp. L.) fields. The study analyzed three S2 images corresponding to different crop growth stages (GS) GS30, GS39-49, and GS69-75, and 13 Vis commonly used for wheat yield estimation were calculated for each image. In addition, three S1 images that were temporally close to the S2 images were acquired, and the vertical-vertical (VV) and vertical-horizontal (VH) backscatter were calculated. The performance of the CatBoost algorithm was compared to that of multiple linear regression (MLR), support vector machine (SVM), and random forest (RF) algorithms in crop yield estimation. The results showed that the combination of S1 and S2 data with the CatBoost algorithm produced a yield prediction with a root mean squared error (RMSE) of 0.24 t ha−1, a relative RMSE (rRMSE) 3.46% and an R2 of 0.95. The result indicates a decrease of 30% in RMSE when compared to using S2 alone. However, when this algorithm was used to estimate the yield of a whole plot, leveraging information from the surrounding plots, the mean absolute error (MAE) was 0.31 t ha−1 which means a mean error of 4.38%. Accurate wheat yield estimation with a spatial resolution of 10 m becomes feasible when utilizing satellite data combined with CatBoost.