For the sustainable development of marine fishery resources, it is essential to comprehensively, accurately, and objectively obtain the spatial characteristics and evolution law of fishing intensity. However, previous studies have focused more on the use of single data sources, such as AIS (Automatic Information System) and VBD (VIIRS boat detection), to obtain fishing intensity information and, as such, have encountered some problems, such as insufficient comprehensive data coverage for ships, non-uniform spatial distribution of data signal acquisition, and insufficient accuracy in obtaining fishing intensity information. The development of big data and remote sensing Earth observation technology has provided abundant data sources and technical support for the acquisition of fishing intensity data for marine fisheries. Based on this situation, this paper proposes a framework that integrates the data of fishing vessels from two sources (AIS, with high space-time granularity, and VBD, with short revisit cycle and high sensitivity), in order to obtain such information based on closely matching and fusing the vector point data of ship positions. With the help of this framework and the strategy of indirectly representing fishing intensity by data point density after fusion, the spatial characteristics and rules of fishing intensity in typical seasons (February, April, September, and November) in the northern South China Sea in 2018 were systematically analyzed and investigated. The results revealed the following: (1) Matching and fusing AIS and VBD data can provide a better perspective to produce robust and accurate marine fishery intensity data. The two types of data have a low proximity match rate (approximately 1.89% and 6.73% of their respective inputs) and the matching success for fishing vessels in the data was 49.42%. (2) Single AIS data can be used for nearshore (50 to 70 km) marine fishery analysis research, while VBD data reflect the objective marine fishing in space, showing obvious complementarity with AIS. (3) The fishing intensity grid data obtained from the integrated data show that high-intensity fishing in the study area was concentrated in the coastal area of Maoming City, Guangdong (0–50 km); the coastal area of Guangxi Beihai (10–70 km); around Hainan Island in Zhangzhou (10–30 km); and the Sanya nearshore area (0–50 km). However, it did not decay with increasing offshore distance, such as at the Trans-Vietnamese boundary in the Beibu Gulf, near the China–Vietnam Common Fisheries Area (50 km) and high-intensity fishing areas. (4) The obtained fishing intensity data (AIS, VBD, and AIS + VBD) were quantitatively analyzed, showing that the CV (Coefficient of Variation) of the average for each month (after fusing the two types of data) was 0.995, indicating that the distribution of the combined data was better than that before fusion (before fusion: AIS = 0.879, VBD = 1.642). Therefore, the integration of AIS and VBD can meet the need for a more effective, comprehensive, and accurate fishing intensity analysis in marine fishery resources.

Proglacial lakes modulate lake-terminating glacier dynamics and can substantially amplify ice loss via frontal calving. While critical for accurate glacier projections, proglacial lake-glacier interactions remain poorly understood across the Himalayas. Here, we apply zonal statistics to investigate spatiotemporal characteristics of proglacial lake impacts on lake-terminating glacier dynamics, by comparing multi-source remote sensing-based surface velocities and surface-elevation-change (thinning) rates of 346 lake-terminating versus 462 land-terminating glaciers (2000–2020). Lake-terminating glaciers exhibit on average 32.39% higher surface velocity and 41.65% greater thinning rate. The strongest velocity impacts occur in the eastern Himalayas (41.62%), northern central Himalayas (51.03%), and Nyainqentanglha (48.76%), while thinning increases are notable in the eastern Hindu Kush, central Himalayas, and Hengduan range. Temporally, proglacial lake influence on velocity intensified nearly linearly from 1990 to 2020 (24.80% to 38.6%), whereas impacts on thinning rates showed a linear declining trend (62.4% to 28.6%) across four five-year intervals from 2000 to 2020. Manual terminus delineation from multi-source satellite imagery (Landsat, Sentinel-2) indicates that, during 2000–2020, 87.6% of lake-terminating glaciers retreated at a mean rate of 10.88±2.40 m yr⁻¹. An interaction index between proglacial lakes and lake-terminating glaciers revealed the strongest interaction in the Nyainqentanglha and Karakoram, followed by intensity zones in adjacent eastern/central Himalayan sectors. Our findings underscore the need for comprehensive observations and modeling to better resolve the processes that govern lake-glacier interactions. This study demonstrates the particular strength of multi-source remote sensing in revealing large-scale glacier–lake interactions.