Identifying Thermokarst Lakes Using Deep Learning and High-Resolution Satellite Images

Science of Remote Sensing, Год журнала: 2024, Номер 10, С. 100175 - 100175

Опубликована: Ноя. 2, 2024

Язык: Английский

---

Diversity of lakes and ponds in the forest-tundra ecozone: from limnicity to limnodiversity

GIScience & Remote Sensing, Год журнала: 2024, Номер 61(1)

Опубликована: Ноя. 13, 2024

Arctic and subarctic landscapes have unique hydrological limnological features are now experiencing rapid change due to climate warming permafrost thaw. The highly abundant lakes, ponds, rivers across these play an increasingly important role in global biogeochemical cycles sentinels of environmental changes. However, studying remote waters poses challenges for both situ sampling remote-sensing analysis. Here we developed a synergistic strategy that combined PlanetScope Sentinel-2 satellite data estimate limnicity (water fraction per land surface), limnodensity (density water bodies), limnodiversity (optical diversity bodies) along boreal forest-tundra transect, from the non-permafrost continuous zones western Nunavik (Subarctic Canada). Our analyses show this region hosts 335,281 bodies, around 90% 0.0001 0.01 km2 size range. In bedrock outcrops, large bodies were mostly associated with glacially carved depressions (higher limnicity). contrast, small predominately found sedimentary infills valleys limnodensity). discontinuous zone had highest limnodiversity. This was likely thaw (thermokarst), particularly collapse, subsidence, erosion palsas (organic mounds), resulting ponds black- brown-colored waters, lithalsas (mineral brown, light-brown, sometimes white-colored waters. Some limnodense limnodiverse landscapes, although covering only 2 7% total area study region, contained over one-third (34%) number 97% which <0.01 km2; they accounted proportion black-colored (23%), but high brown- (60%) light (92%) throughout region. research underscores utility optical sensing assessing body types evaluating their individual distinct aquatic responses change. dataset may be used improve modeling carbon fluxes by better categorizing affected organic or mineral soil type settings. is factor dictating responses, effects on albedo, feedbacks, ecosystem dynamics framework here applied elsewhere world densities variable properties.

Язык: Английский

---

MineTinyNet-YOLO: An Efficient Small Object Detection Method for Complex Underground Coal Mine Scenarios

Lecture notes in computer science, Год журнала: 2024, Номер unknown, С. 364 - 378

Опубликована: Дек. 30, 2024

Язык: Английский

---

Evaluation of Deep Learning-Based Water Bodies and Flooded Area Detection with Nanosatellites: The PlanetScope Satellite Imageries and HRNet Model

Korean Journal of Remote Sensing, Год журнала: 2024, Номер 40(5-1), С. 617 - 627

Опубликована: Окт. 31, 2024

Язык: Английский

---