Search

Crop mapping using satellite imagery is crucial for agriculture applications. However, a fundamental challenge that hinders crop mapping progress is the scarcity of samples. The latest foundation model, Segment Anything Model (SAM), provides an opportunity to address this issue, yet few studies have been conducted in this area. This study investigated the parcel segmentation performance of SAM on commonly used medium-resolution satellite imagery (i.e., Sentinel-2 and Landsat-8) and proposed a novel automated sample generation framework based on SAM. The framework comprises three steps. First, an image optimization automatically selects high-quality images as the inputs for SAM. Then, potential samples are generated based on the masks produced by SAM. Finally, the potential samples are subsequently subjected to a sample cleaning procedure to acquire the most reliable samples. Experiments were conducted in Henan Province, China, and southern Ontario, Canada, using six proven effective classifiers. The effectiveness of our method is demonstrated through the combination of field-survey-collected samples and differently proportioned generated samples. Our results indicated that directly using SAM for parcel segmentation remains challenging, unless the parcels are large, regular in shape, and have distinct color differences from surroundings. Additionally, the proposed approach significantly improved the performance of classifiers and alleviated the sample scarcity problem. Compared to classifiers trained only by field-survey-collected samples, our method resulted in an average improvement of 16% and 78.5% in Henan and Ontario, respectively. The random forest achieved relatively good performance, with weighted-average F1 of 0.97 and 0.996 obtained using Sentinel-2 imagery in the two study areas, respectively. Our study contributes insights into solutions for sample scarcity in crop mapping and highlights the promising application of foundation models like SAM. [ABSTRACT FROM AUTHOR]

The scientific and reasonable division of geographical units of cultivated land reserve resource areas is an important foundation for comprehensive land consolidation (CLC). Saline–alkali land is an important cultivated land reserve resource, and is significant for ensuring food security. This paper constructs a two-level land consolidation index system for cultivated land reserve resource areas. The Kruskal graph theory clustering algorithm was used to divide the study area into four types of ecological zones. On the basis of ecological zones, the study area was further divided into three types of consolidation units. Finally, the spatial relationship between ecological zoning and land use types was clarified, and the classification distribution of land consolidation potential was determined through an analysis of the CLC of two-level geographical units. Da'an City, Jilin province, China, was selected as the study area. The results of this study show the following: (1) In Da'an City, the conservation priority areas are concentrated in the north, the development priority areas are concentrated in the central and southern areas, and the comprehensive upgrade areas and adjustment rebuilding areas are in the transition zones. (2) The CLC potential trend is low in the north but high in the south in Da'an City. This paper proposes a framework for a geographical unit delineation method for saline–alkali-enriched cultivated land reserve resource zones, and analyzes the spatial layout of ecological protection demands and land consolidation potential in Da'an City. The results and conclusions of this study will provide a reference for CLC in cultivated land reserve resource areas. [ABSTRACT FROM AUTHOR]

The influence of individual meteorological factors on different airborne pollutants has been widely studied. However, few studies have considered the effect of temporal scales on the extracted pollutant–meteorology association. Based on convergent cross mapping (CCM), we compared the influence of major meteorological factors on PM 2.5 , PM 10 and O 3 concentrations in 2020 at the 3 and 24 h scales respectively. In terms of the extracted dominant meteorological factor, the consistence between the analysis at the 3 and 24 h scales was relatively low, suggesting a large difference even from a qualitative perspective. In terms of the mean p value, the effect of temporal scale on PM (PM 2.5 and PM 10) –meteorology association was consistent, yet largely different from the temporal-scale effect on O 3. Temperature was the most important meteorological factor for PM 2.5 , PM 10 and O 3 across China at both the 3 and 24 h scales. For PM 2.5 and PM 10 , the extracted PM–temperature association at the 24 h scale was stronger than that at the 3 h scale. Meanwhile, for summer O 3 , due to strong reactions between precursors, the extracted O 3 –temperature association at the 3 h scale was much stronger. Due to the discrete distribution, the extracted association between all pollutants and precipitation was much weaker at the 3 h scale. Similarly, the extracted PM–wind association was notably weaker at the 3 h scale. Due to precursor transport, summertime O 3 –wind association was stronger at the 3 h scale. For atmospheric pressure, the pollutant–pressure association was weaker at the 3 h scale except for summer, when interactions between atmospheric pressure and other meteorological factors were strong. From the spatial perspective, pollutant–meteorology associations at 3 and 24 h were more consistent in the heavily polluted regions, while extracted dominant meteorological factors for pollutants demonstrated more difference at 3 and 24 h in the less polluted regions. This research suggests that temporal scales should be carefully considered when extracting natural and anthropogenic drivers for airborne pollution. [ABSTRACT FROM AUTHOR]