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.

Since the 20th century, the global soybean trade has undergone major changes, shaped by rising demand, climate-related risks, and shifting international dynamics. Despite its global importance, important gaps remain in understanding the complex drivers and sustainability challenges of this transformation. This review synthesizes both direct and indirect forces reshaping trade flows, spanning market dynamics, supply chain logistics, policy shifts, and technological innovation. We examine how soybean trade expansion has impacted deforestation, inequality, and food security, and assess the responses of governments and companies to address these challenges. Finally, we provide a forward-looking perspective on the strategic pathways needed to ensure a more resilient and sustainable global soybean system. The integrated insights offered in this review can inform sustainable trade strategies and foster cross-scale policy coordination for a more resilient global agri-food system.

Lossless remote sensing image compression aims to reduce the storage size of images without any information loss, ensuring that the decompressed image is identical to the original one. Most existing methods focus on lossy image compression that reduce the storage cost with certain data loss. It is challenging to perform lossless compression due to the very high-resolution images, long encoding-decoding time and low compression efficiency. In this paper, we propose a lossless compression framework that compresses remote sensing images in a coarse-to-fine manner. Specifically, checkerboard segmentation is applied on each image to generate six sub-images from the main diagonal and counter-diagonal of each channel to maximally preserve the detail and structural information. The sub-images from the main diagonal are initially compressed by a traditional compression method, while the sub-images from the counter-diagonal are compressed channel by channel using our proposed probability prediction network (P2Net) and arithmetic coding with the previously encoded sub-images from both the main diagonal and counter-diagonal as prior knowledge. The proposed P2Net consists of a upsampling module, a feature enhancement module, a downsampling module and a probability prediction module to learn the discrete probability distribution of pixels. Lossless compression is conducted with arithmetic coding on the discrete probability distribution. To the best of our knowledge, this is the first deep learning-based lossless compression framework for 3-channel remote sensing images. Experiments demonstrate that our framework outperforms the state-of-the-art methods and requires about 3.4 seconds to compress a 1024 $\times 1024\times$ 3 image with 2.9% efficiency improvement compared to JPEG XL.

An adaptive fixed-time dynamic surface tracking control scheme is developed in this paper for a class of strict-feedback nonlinear systems, where the control input is subject to hysteresis dynamics. To deal with the input hysteresis, a compensation filter is introduced, reducing the difficulty of design and analysis. Based on the universal approximation theory, the radial basis function neural networks are employed to approximate the unknown functions in the nonlinear dynamics. On this basis, fixed-time adaptive laws are constructed to approximate the unknown parameters. The dynamic surface technique is utilized to handle the complexity explosion problem, where fixed-time performance is ensured. Moreover, the designed controller can avoid singularities and achieve fixed-time convergence of error signals. Simulation results verify the efficacy of the method developed, where a comparison between the scheme developed with existing results is provided.

Located in the interior of Eurasia and to the north of the Qinghai–Tibet Plateau, the Northwest China experiences severe drought conditions as moist air from the ocean is unable to travel the long distance and penetrate the region’s mountain barriers. These special geo-climatic conditions result in Northwest China being highly sensitive to climate change. In this study, the characteristics of the response of the normalized difference vegetation index (NDVI) to the trend towards a warmer and wetter climate in Northern China from 1982 to 2019 were investigated. The results show that there were significant differences between these trends for the periods 1982–2000 and 2000–2019, with overall precipitation decreasing before 2000 but increasing afterwards. After 2000, the rate of temperature increase also slowed down, whereas the NDVI increased at an obviously faster rate. Compared with the period 1982–2000, during the period 2000–2019, the NDVI was more affected by precipitation than by the temperature. The results of a normalized linear regression also show that, for most vegetation types, the temperature played a more dominant role during the period 1982–2000, whereas precipitation had a more significant effect on the NDVI during the period 2000–2019. However, it was also found that, throughout the study period, the precipitation had a greater impact on forest NDVI and the temperature had a greater impact on the NDVI in areas of bare land. In addition, the results show that the strength of the relationship between the NDVI and climate in northwest China changed over time, with the relationship between NDVI and precipitation tending to become stronger and the relationship between NDVI and temperature tending to become weaker. The results will provide a new understanding of the relationship between vegetation and climate in northwest China and help to better cope with the risks brought by climate change.