Earth observation data with high spatiotemporal resolution are critical for dynamic monitoring and prediction in geoscience applications, however, due to some technique and budget limitations, it is not easy to acquire satellite images with both high spatial and high temporal resolutions. Spatiotemporal image fusion techniques provide a feasible and economical solution for generating dense-time data with high spatial resolution, pushing the limits of current satellite observation systems. Among existing various fusion algorithms, deeplearningbased models reveal a promising prospect with higher accuracy and robustness. This paper refined and improved the existing deep convolutional spatiotemporal fusion network (DCSTFN) to further boost model prediction accuracy and enhance image quality. The contributions of this paper are twofold. First, the fusion result is improved considerably with brand-new network architecture and a novel compound loss function. Experiments conducted in two different areas demonstrate these improvements by comparing them with existing algorithms. The enhanced DCSTFN model shows superior performance with higher accuracy, vision quality, and robustness. Second, the advantages and disadvantages of existing deeplearningbased spatiotemporal fusion models are comparatively discussed and a network design guide for spatiotemporal fusion is provided as a reference for future research. Those comparisons and guidelines are summarized based on numbers of actual experiments and have promising potentials to be applied for other image sources with customized spatiotemporal fusion networks.

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.

Mapping nationwide in-season crop-type data is a significant and challenging task in agriculture remote sensing. The existing data product for U.S. crop-type planting, such as the Cropland Data Layer (CDL), falls short in facilitating near-real-time applications. This paper designed a workflow aimed at automating the generation of in-season CDL-like products for the U.S. We methodically extracted trusted pixels as land cover labels from historical CDL datasets, employing Sentinel-2, Landsat 8, and Landsat-9 as sources for spectrum data, using the random forest classifier to conduct nationwide crop-type classifications. These classifications were integrated into the In-Season Crop Data Layer (ICDL) covering the entire Conterminous United States (CONUS). This approach facilitated the efficient generation of ICDLs for May, June, and July 2022, achieving satisfactory accuracy in July. Compared to Nebraska and Iowa ground truth data, ICDL achieved F1 scores of (0.911, 0.845) for corn and (0.959, 0.969) for soybean. Furthermore, ICDL’s regional acreage estimates for major crops (corn, soybean, spring wheat, cotton, winter wheat, and rice) closely align with USDA NASS figures, showing minimal variances as low as (0.01%, -0.68%, 0.19%, -4.39%, -5.78%, -1.28%). Notably, ICDL outperforms CDL in most assessments. This research consistently produces annual ICDLs from May to July that are readily accessible to the public in the iCrop system. Simultaneously, it presents an alternative technique for nationwide, in-season mapping of crop types.

Given the increasing prevalence of droughts, unpredictable rainfall patterns, and limited access to dependable water sources in the United States and worldwide, it has become crucial to implement effective irrigation scheduling strategies. Irrigation is triggered when some variables, such as soil moisture or accumulated water deficit, exceed a given threshold in the most common approaches applied in irrigation scheduling. A High-Resolution Land Data Assimilation System (HRLDAS) was used in this study to generate timely and accurate soil moisture and evapotranspiration (ET) data for irrigation management. By integrating HRLDAS products and the crop growth model (AquaCrop), an automated data-driven irrigation scheduling approach was developed and evaluated. For HRLDAS ET and soil moisture, the ET-water balance (ET-WB)-based method and soil-moisture-based method were applied accordingly. The ET-WB-based method showed a 10.6~33.5% water-saving result in dry and set seasons, whereas the soil moisture-based method saved 7.2~37.4% of irrigation water in different weather conditions. Both of these methods demonstrated good results in saving water (with a varying range of 10~40%) without harming crop yield. The optimized thresholds in the two approaches were partially consistent with the default values from the Food and Agriculture Organization and showed a similar trend in the growing season. Furthermore, the forecasted rainfall was integrated into this model to see its water-saving effect. The results showed that an additional 10% of irrigation water, which is 20~50%, can be saved without harming the crop yield. This study automated the data-driven approach for irrigation scheduling by taking advantage of HRLDAS products, which can be generated in a near-real-time manner. The results indicated the great potential of this automated approach for saving water and irrigation decision making.

In the past few decades, most urban areas in the world have been facing the pressure of an increasing population living in poverty. A recent study has shown that up to 80% of the population of some cities in Africa fall under the poverty line. Other studies have shown that poverty is one of the main contributors to residents’ poor health and social conflict. Reducing the number of people living in poverty and improving their living conditions have become some of the main tasks for many nations and international organizations. On the other hand, urban gentrification has been taking place in the poor neighborhoods of all major cities in the world. Although gentrification can reduce the poverty rate and increase the GDP and tax revenue of cities and potentially bring opportunities for poor communities, it displaces the original residents of the neighborhoods, negatively impacting their living and access to social services. In order to support the sustainable development of cities and communities and improve residents’ welfare, it is essential to identify the location, scale, and dynamics of urban poverty and gentrification, and remote sensing can play a key role in this. This paper reviews, summarizes, and evaluates state-of-the-art approaches for identifying and mapping urban poverty and gentrification with remote sensing, GIS, and machine learning techniques. It also discusses the pros and cons of remote sensing approaches in comparison with traditional approaches. With remote sensing approaches, both spatial and temporal resolutions for the identification of poverty and gentrification have been dramatically increased, while the economic cost is significantly reduced.