Search

Abstract Winter wheat constitutes approximately 20% of China’s total cereal production. However, calculations of total production based on multiplying the planted area by the yield have tended to produce overestimates. In this study, we generated sample points from existing winter wheat maps and obtained samples for different years using a temporal migration method. Random forest classifiers were then constructed using optimized features extracted from spectral and phenological characteristics and elevation information. Maps of the harvested and planted areas of winter wheat in Chinese eight provinces from 2018 to 2022 were then produced. The resulting maps of the harvested areas achieved an overall accuracy of 95.06% verified by the sample points, and the correlation coefficient between the CROPGRIDS dataset is about 0.77. The harvested area was found to be about 13% smaller than the planted area, which can primarily be attributed to meteorological hazards. This study represents the first attempt to map the winter wheat harvested area at 10–m resolution in China, and it should improve the accuracy of yield estimation.

Abstract China, as the world’s biggest soybean importer and fourth-largest producer, needs accurate mapping of its planting areas for global food supply stability. The challenge lies in gathering and collating ground survey data for different crops. We proposed a spatiotemporal migration method leveraging vegetation indices’ temporal characteristics. This method uses a feature space of six integrals from the crops’ phenological curves and a concavity-convexity index to distinguish soybean and non-soybean samples in cropland. Using a limited number of actual samples and our method, we extracted features from optical time-series images throughout the soybean growing season. The cloud and rain-affected data were supplemented with SAR data. We then used the random forest algorithm for classification. Consequently, we developed the 10-meter resolution ChinaSoybean10 maps for the ten primary soybean-producing provinces from 2019 to 2022. The map showed an overall accuracy of about 93%, aligning significantly with the statistical yearbook data, confirming its reliability. This research aids soybean growth monitoring, yield estimation, strategy development, resource management, and food scarcity mitigation, and promotes sustainable agriculture.

Accurately predicting winter wheat yield before harvest could greatly benefit decision-makers when making management decisions. In this study, we utilized weather forecast (WF) data combined with Sentinel-2 data to establish the deep-learning network and achieved an in-season county-scale wheat yield prediction in China’s main wheat-producing areas. We tested a combination of short-term WF data from the China Meteorological Administration to predict in-season yield at different forecast lengths. The results showed that explicitly incorporating WF data can improve the accuracy in crop yield predictions [Root Mean Square Error (RMSE) = 0.517 t/ha] compared to using only remote sensing data (RMSE = 0.624 t/ha). After comparing a series of WF data with different time series lengths, we found that adding 25 days of WF data can achieve the highest yield prediction accuracy. Specifically, the highest accuracy (RMSE = 0.496 t/ha) is achieved when predictions are made on Day of The Year (DOY) 215 (40 days before harvest). Our study established a deep-learning model which can be used for early yield prediction at the county level, and we have proved that weather forecast data can also be applied in data-driven deep-learning yield prediction tasks.

Accurate estimation of fractional vegetation cover (FVC) is essential for crop growth monitoring. Currently, satellite remote sensing monitoring remains one of the most effective methods for the estimation of crop FVC. However, due to the significant difference in scale between the coarse resolution of satellite images and the scale of measurable data on the ground, there are significant uncertainties and errors in estimating crop FVC. Here, we adopt a Strategy of Upscaling-Downscaling operations for unmanned aerial systems (UAS) and satellite data collected during 2 growing seasons of winter wheat, respectively, using backpropagation neural networks (BPNN) as support to fully bridge this scale gap using highly accurate the UAS-derived FVC (FVCUAS) to obtain wheat accurate FVC. Through validation with an independent dataset, the BPNN model predicted FVC with an RMSE of 0.059, which is 11.9% to 25.3% lower than commonly used Long Short-Term Memory (LSTM), Random Forest Regression (RFR), and traditional Normalized Difference Vegetation Index-based method (NDVI-based) models. Moreover, all those models achieved improved estimation accuracy with the Strategy of Upscaling-Downscaling, as compared to only upscaling UAS data. Our results demonstrate that: (1) establishing a nonlinear relationship between FVCUAS and satellite data enables accurate estimation of FVC over larger regions, with the strong support of machine learning capabilities. (2) Employing the Strategy of Upscaling-Downscaling is an effective strategy that can improve the accuracy of FVC estimation, in the collaborative use of UAS and satellite data, especially in the boundary area of the wheat field. This has significant implications for accurate FVC estimation for winter wheat, providing a reference for the estimation of other surface parameters and the collaborative application of multisource data.

Accurate predictions of wheat yields are essential to farmers’production plans and to the international trade in wheat. However, only poor approximations of the productivity of wheat crops in China can be obtained using traditional linear regression models based on vegetation indices and observations of the yield. In this study, Sentinel-2 (multispectral data) and ZY-1 02D (hyperspectral data) were used together with 15709 gridded yield data (with a resolution of 5 m × 5 m) to predict the winter wheat yield. These estimates were based on four mainstream data-driven approaches: Long Short-Term Memory (LSTM), Random Forest (RF), Gradient Boosting Decision Tree (GBDT), and Support Vector Regression (SVR). The method that gave the best estimate of the winter wheat yield was determined, and the accuracy of the estimates based on multispectral and hyperspectral data were compared. The results showed that the LSTM model, for which the RMSE of the estimates was 0.201 t/ha, performed better than the RF (RMSE = 0.260 t/ha), GBDT (RMSE = 0.306 t/ha), and SVR (RMSE = 0.489 t/ha) methods. The estimates based on the ZY-1 02D hyperspectral data were more accurate than those based on the 30-m Sentinel-2 data: RMSE = 0.237 t/ha for the ZY-1 02D data, which is about a 5% improvement on the RSME of 0.307 t/ha for the 30-m Sentinel-2 data. However, the 10-m Sentinel-2 data performed even better, giving an RMSE of 0.219 t/ha. In addition, it was found that the greenness vegetation index SR (simple ratio index) outperformed the traditional vegetation indices. The results highlight the potential of the shortwave infrared bands to replace the visible and near-infrared bands for predicting crop yields Our study demonstrates the advantages of the deep learning method LSTM over machine learning methods in terms of its ability to make accurate estimates of the winter wheat yield.

The tiller density is a key agronomic trait of winter wheat that is essential to field management and yield estimation. The traditional method of obtaining the wheat tiller density is based on manual counting, which is inefficient and error prone. In this study, we established machine learning models to estimate the wheat tiller density in the field using hyperspectral and multispectral remote sensing data. The results showed that the vegetation indices related to vegetation cover and leaf area index are more suitable for tiller density estimation. The optimal mean relative error for hyperspectral data was 5.46%, indicating that the results were more accurate than those for multispectral data, which had a mean relative error of 7.71%. The gradient boosted regression tree (GBRT) and random forest (RF) methods gave the best estimation accuracy when the number of samples was less than around 140 and greater than around 140, respectively. The results of this study support the extension of the tested methods to the large-scale monitoring of tiller density based on remote sensing data.