Search

Over the last decades, growing crop production across China has had far-reaching consequences for both the environment and human welfare. One of the emerging questions is "how to meet the growing food demand in China?" In essence, the consensus is that the best way forward would be to increase crop yield rather than further extend the current cropland area. However, assessing progress in crop production is challenging as it is driven by multiple factors. To date, there are no studies to determine how multiple factors affect the crop production increase, considering both intensive farming (using yield and multiple cropping index) and large-scale farming (using mean parcel size and number of parcels). Using the Logarithmic-Mean-Divisia-Index (LMDI) decomposition method combined with statistical data and land cover data (GlobeLand30), we assess the contribution of intensive farming and large-scale farming changes to crop production dynamics at the national and county scale. Despite a negative contribution from MPS (mean parcel size, −25%), national crop production increased due to positive contributions from yield (+77%), MCI (multiple cropping index, +27%), as well as NP (number of parcels, +21%). This allowed China to meet the growing national crop demand. We further find that large differences across regions persist over time. For most counties, the increase in crop production is a consequence of improved yields. However, in the North China Plain, NP is another important factor leading to crop production improvement. On the other hand, regions witnessing a decrease in crop production (e.g., the southeast coastal area of China) were characterized by a remarkable decrease in yield and MCI. Our detailed analyses of crop production provide accurate estimates and therefore can guide policymakers in addressing food security issues. Specifically, besides stabilizing yield and maintaining the total NP, it would be advantageous for crop production to increase the mean parcel size and MCI through land consolidation and financial assistance for land transfer and advanced agricultural infrastructure. [ABSTRACT FROM AUTHOR]

The pairwise separability index (SI) has been demonstrated as an effective indicator for capturing crucial phenological differences between two plant species. However, its application to crop types, which have more obvious phenological characteristics than natural vegetation, has received less attention, and extending the pairwise SI to multiple crops for feature selection still remains a challenge. This paper presented two SI extension approaches (\textSI\mathrm{ave} and \textSI\min) to select the optimal spectro-temporal features for multiple crops, and investigated their classification performance using Heilongjiang Province, China, as a study area. Feature interpretability and classification accuracy of different crops were evaluated for the two approaches. The results showed that the \textSI\mathrm{ave} approach generally has relatively high feature interpretability due to its better description of crucial phenological characteristics of different crops. Those crops with high separability are insensitive to the extension approach and have similar classification accuracy for the two approaches, whereas those crops with poor separability show good performance with the \textSI\min method. Due to the higher temporal autocorrelation, the optimal features for crop classification that are selected by the \textSI\mathrm{ave} approach exhibit greater information redundancy across the time domain than those that are selected by the \textSI\min approach, which largely explains the relatively low classification accuracy achieved using the \textSI\mathrm{ave} approach. These comparison results between \textSI\min and \textSI\mathrm{ave} approaches also indicate that time-series images with high temporal resolution do not necessarily produce high classification accuracy, regardless of their ability to describe the seasonal characteristics of crops. [ABSTRACT FROM PUBLISHER]

Leaf area index (LAI) and canopy chlorophyll content (CCC) are important indicators that describe the growth status and nitrogen deficiencies of crops. Several studies have been performed to estimate LAI and CCC using multispectral cameras onboard an unmanned airborne vehicle (UAV) system. However, the impacts of illuminations during UAV flight and problems of how to invert still need more investigation. UAV flights with a multispectral camera were performed under clear (diffuse ratio 0) and cloudy illumination conditions (diffuse ratio 1) over rapeseed, wheat and sunflower (only clear) fields. One-dimension radiative transfer model PROSAIL was run twice to generate a clear-sky model and a cloudy-sky model, respectively. The LAI and CCC of flights under a clear sky were inverted from the clear-sky model, and the flights under cloudy conditions were inverted from both clear-sky and cloudy-sky models to compare the results. Moreover, three Look-Up-Tables (LUT) were built with same input variables but different distributions of LAI. Results showed that LAI from uniform dense LUT had better correspondence with ground measurements for all crops (R2 = 0.51~0.69). The illumination condition had little impact on small to medium LAI (LAI < 5) and CCC. However, the inversion of imageries during cloudy sky conditions from the clear-sky model led to an overestimation of high LAI values. [ABSTRACT FROM AUTHOR]

Information on crop spatial distribution is essential for agricultural monitoring and food security. Classification with remote-sensing time series images is an effective way to obtain crop distribution maps across time and space. Optimal features are the precondition for crop classification and are critical to the accuracy of crop maps. Although several approaches are available for extracting spectral, temporal, and phenological features for crop identification, these methods depend heavily on domain knowledge and human experiences, adding uncertainty to the final crop classification. This study proposed a novel Genetic Programming (GP) approach to learning high-level features from time series images for crop classification to address this issue. We developed a new representation of GP to extend the GP tree's width and depth to dynamically generate either fixed or flexible informative features without requiring domain knowledge. This new GP approach was wrapped with four classifiers, i.e., K-Nearest Neighbor (KNN), Decision Tree (DT), Naive Bayes (NB), and Support Vector Machine (SVM), and was then used for crop classification based on MODIS time series data in Heilongjiang Province, China. The performance of the GP features was compared with the traditional features of vegetation indices (VIs) and the advanced feature learning method Multilayer Perceptron (MLP) to show GP effectiveness. The experiments indicated that high-level features learned by GP improved the classification accuracies, and the accuracies were higher than those using VIs and MLP. GP was more robust and stable for diverse classifiers, different feature numbers, and various training sample sets compared with classification using VI features and the classifier MLP. The proposed GP approach automatically selects valuable features from the original data and uses them to construct high-level features simultaneously. The learned features are explainable, unlike those of a black-box deep learning model. This study demonstrated the outstanding performance of GP for feature learning in crop classification. GP has the potential of becoming a mainstream method to solve complex remote sensing tasks, such as feature transfer learning, image classification, and change detection. [ABSTRACT FROM AUTHOR]