Accurate and timely estimation of crop yield at a small scale is of great significance to food security and harvest management. Recent studies have proven remote sensing is an efficient method for yield estimation and machine learning, especially deep learning, can infer a good prediction by integrating multi-source datasets like satellite data, climate data, soil data, and so on. However, there are some bottleneck challenges to improve accuracy. First, the popular remote sensing data used for yield prediction fall into two major groups, time-series data, and constant data. Surprisingly little attention has been devoted to deep learning networks that can integrate the two kinds of data effectively; second, both temporal and spatial features play a role in affecting the yields. But most of the existing approaches employed either CNN (convolutional neural network) or RNN (recurrent neural network). CNN cannot learn temporal patterns, while RNN barely can learn spatial characteristics. This work proposed a novel multi-level deep learning (MLDL) model coupling RNN and CNN to extract both spatial and temporal features. The inputs include both time-series remote sensing data and soil property data and the model outputs yield. We experimented with the model in U.S. Corn Belt states and used it to predict corn yield from 2013 to 2016 at the county-level. The results approve the effectiveness and advantages of the proposed approach over the other methods. In the future, the model will be used on other crops like soybean and winter wheat to assist agriculture decision making.

Time series monitoring of earthquake-stricken areas is significant in evaluating post-disaster reconstruction and recovery. The time series of nighttime light (NTL) data collected by the defense meteorological satellite program-operational linescan system (DMSP/OLS) sensors provides a unique and valuable resource to study changes in human activity (HA) because of the long period of available data. In this paper, the DMSP/OLS NTL images’ digital number (DN) is used as a proxy for the intensity of HA since there is a high correlation between them. The purpose of this study is to develop a methodology to analyze the changes of intensity and distribution of HA in different areas affected by a 2008 earthquake in Wenchuan, China. In order to compare the trends of HA before and after the earthquake, the DMSP/OLS NTL images from 2003 to 2013 were processed and analyzed. However, their analysis capability is greatly limited owing to a lack of in-flight calibration. To improve the continuity and comparability of DMSP/OLS NTL images, this study developed an automatic intercalibration method to systematically correct NTL data. The results reveal that: (1) compared with the HA before the earthquake, the reconstruction and recovery of the Wenchuan earthquake have led to a significant increase of HA in earthquake-stricken areas within three years after the earthquake; (2) the fluctuation of HA in a severely-affected area is greater than that in a less-affected area; (3) recovery efforts increase development in the most affected areas to levels that exceeded the rates in similar areas which experienced less damage; and (4) areas alongside roads and close to reconstruction projects exhibited increased development in regions with otherwise low human activity.

Despite the importance of tidal ecosystems in the global carbon budget, the relationships between environmental drivers and carbon dynamics in these wetlands remain poorly understood. This limited understanding results from the challenges associated with in situ flux studies and their correlation with satellite imagery which can be affected by periodic tidal flooding. Carbon dioxide eddy covariance (EC) towers are installed in only a few wetlands worldwide, and the longest eddy-covariance record from Georgia (GA) wetlands contains only two continuous years of observations. The goals of the present study were to evaluate the performance of existing MODIS Gross Primary Production (GPP) products (MOD17A2) against EC derived GPP and develop a tide-robust Normalized Difference Moisture Index (NDMI) based model to predict GPP within a Spartina alterniflora salt marsh on Sapelo Island, GA. These EC tower-based observations represent a basis to associate CO₂ fluxes with canopy reflectance and thus provide the means to use satellite-based reflectance data for broader scale investigations. We demonstrate that Light Use Efficiency (LUE)-based MOD17A2 does not accurately reflect tidal wetland GPP compared to a simple empirical vegetation index-based model where tidal influence was accounted for. The NDMI-based GPP model was capable of predicting changes in wetland CO₂ fluxes and explained 46% of the variation in flux-estimated GPP within the training data, and a root mean square error of 6.96 g C m⁻² in the validation data. Our investigation is the first to create a MODIS-based wetland GPP estimation procedure that demonstrates the importance of filtering tidal observations from satellite surface reflectance data.

Rapeseed is an important oil product in China. China’s current soybean trade issues with major soybean producing countries have caused a large decline in soybean import since 2017. This may bring the increasing needs of rapeseed import, which would have an impact on domestic production. However, our knowledge on the effects of international rapeseed trade on domestic production remains unknown. It is thus important to understand the pattern of rapeseed in China under this scenario, as it may provide necessary information for all relevant stakeholders. With this goal, this study aims to investigate the spatial and temporal patterns of winter rape in China’s major winter rape production region, the Middle Reaches of the Yangtze River Valley (MYR), during 2003–2015 using time-series Moderate Resolution Imaging Spectroradiometer (MODIS) data. A decision tree according to the difference in enhanced vegetation index (EVI) profiles of land-cover types was built to extract winter rape. The results show that there is an essential decrease in both the number and density of winter rape patches under the opening global rapeseed market. There are significant hotspots of winter rape gain and loss, within which the loss dominated the trend. The significant cost advantage of rapeseed in the international market may largely reduce the domestic cultivation in China through telecoupling effects. Understanding the spatio-temporal dynamics of winter rape on the MYR has significant economic and policy implications and can provide great supports for the agricultural production, policy-making, and oil products trade in the international market.

This article proposes a Gaussian-mixture-model (GMM)-based method with optimal Gaussian components to address the high intra-class spectral variability in urban land-cover mapping using remote sensing images with very high resolution (VHR). GMMs can simulate and approximate any data distribution provided the optimal Gaussian components can be found. Through improving the model parameters in view of the characteristic of VHR remote sensing images, the parameter space of GMM is optimized significantly, and the model can find the optimal Gaussian components that are suitable for remote sensing images with different resolutions. Experimental results of Wuhan urban area using two images with different resolutions have demonstrated the efficiency and effectiveness of the model. The optimized GMM-based method performs at least comparably or superior to the state-of-the-art classifiers such as support vector machines (SVMs), characterizes man-made land-cover types better than conventional methods, fuses spectral and textural features of VHR image properly, and meanwhile has lower computational complexity.

Mapping crop patterns with remote sensing data is of great importance for agricultural production, food security and agricultural sustainability. In this paper, a hierarchical clustering method was proposed to map cropping frequency from time-series Moderate Resolution Imaging Spectroradiometer (MODIS) Enhanced Vegetation Indices (EVI) data and the spatial and temporal patterns of cropping frequency from 2001 to 2015 in Hubei Province of China were analyzed. The results are as follows: (1) The total double crop areas decreased slightly, while total single crop areas decreased significantly during 2001 and 2015; (2) The transfer between double crop and single crop was frequent in Hubei with about 11~15% croplands changed their cropping frequency every 5 years; (3) The crop system has obvious regional differentiation for their change trend at the county level.