Satellite active-fire points are often interpreted as forest fires even though they include agricultural and other managed-vegetation burning. We combined 1,789,248 quality-filtered MODIS active-fire detections over China (2003–2024) with annual MCD12Q1 land cover, RESOLVE ecoregions, MCD64A1 burned area, ERA5-Land climate and MOD13A3 NDVI. Land cover and ecoregions were attributed to a reproducible year–month stratified probability sample of 105,600 detections; survey weights recovered national totals. Cropland accounted for 39.22 ^-1 , compared with 21.60 MW across all detections. An independent-sensor comparison with 3,549,096 filtered Suomi-NPP VIIRS detections for 2013–2024 showed strong agreement in deseasonalized monthly counts ( ρ =0.872 ), 0.5^∘ spatial shares ( ρ =0.914 ) and provincial shares ( ρ =0.949 ), although VIIRS recorded a median 3.55 times as many detections. Monthly MODIS metrics also agreed with MCD64A1 burned area after removal of calendar-month seasonality (all-fire ρ =0.669 ; forest ρ =0.513 ). In season- and trend-adjusted models with HAC uncertainty, lower surface soil moisture was associated with more forest detections and burned area, while forest burned area was also greater under higher wind speed and lower NDVI. National detection counts had no significant monotonic trend, but mean FRP declined. Point-level attribution therefore separates a large agricultural-fire signal from a smaller, radiatively disproportionate forest subset; cropland classification is consistent with agricultural burning but does not by itself identify straw combustion or another ignition source.
Remote sensing has emerged as a powerful and cost-effective alternative, providing synoptic and repeatable observations of Optically Active water quality Parameters (OAPs), such as Chlorophyll-a (Chl-a), Total Suspended Solids (TSS), Secchi Disk Depth (SDD), Coloured Dissolved Organic Matter (CDOM), and turbidity. This paper synthesised, for the first time, recent advances in the application of remote sensing and machine learning for OAP assessment and monitoring, with a specific focus on comparative analysis of their retrieval performance in inland water bodies. We applied the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) protocol to select 127 peer-reviewed studies published between 2010 and June 2025. The findings highlight the predominant use of Sentinel-2 and Landsat-8 multispectral data in 43
Seasonal water-level fluctuations in hydropower reservoirs create drawdown zones whose extent and spatial distribution are difficult to characterize from point-based observations alone. This study mapped surface-water extent and drawdown zones in Thac Ba Reservoir, northern Vietnam, using four Landsat Level-2 observations representing selected low-water and high-water conditions in 2015 and 2025. Surface water was classified with a Random Forest model using multispectral bands, water indices and terrain information and the seasonal classifications were overlaid to delineate drawdown zones. Independent validation for the 2025 observations yielded overall accuracies of 0.833–0.867 and Cohen’s kappa values of 0.658–0.722, exceeding those obtained with NDWI thresholding. Reservoir-wide water area ranged from 13,684 to 16,261 ha in the selected 2015 observations and from 16,502 to 19,500 ha in 2025. The corresponding drawdown zones covered 2,792 and 3,269 ha, respectively, with the largest absolute areas concentrated in the Western and Upper subregions. The larger drawdown extent mapped for the 2025 observation pair reflects the hydrological states represented by the selected scenes and should not be interpreted as a continuous decadal trend. The study shows that combining Landsat classification with seasonal mask overlay and morphology-based subregional analysis can provide a useful representation of spatially heterogeneous shoreline exposure in complex reservoirs.
District-scale hail surveillance remains challenging in northern Thailand because rapidly evolving convection, complex mountainous terrain, sparse surface verification, and limited vertically resolved radar products constrain interpretation. This study presents a sounding-constrained, radar-geometry-aware framework for mapping hydrometeor phase structure during the 15 April 2024 hailstorm. Fifteen-minute C-band dual-polarization radar volumes underwent polarimetric quality control, Z–PHI attenuation correction, and gridding to produce constant-altitude plan-position indicators (CAPPIs) at 2–6 km above ground level (AGL) with 2-km resolution. The 07:00 local-time Chiang Mai radiosonde constrained hydrometeor classification and thermal levels. The environment featured most-unstable convective available potential energy (MUCAPE) of 1.63 kJ kg⁻1, entrainment-adjusted convective available potential energy (ECAPE) of 0.94 kJ kg⁻1, a wet-bulb-zero height near 3.7 km, a 2.85-km-deep − 10 to − 30 °C hail-growth zone, 0–6 km bulk shear of 16.2 m s⁻1, and downdraft convective available potential energy (DCAPE) of 1.37 kJ kg⁻1. Strong reflectivity extended into the subfreezing layer, with near-zero differential reflectivity ( Z_DR ), reduced copolar correlation coefficient ( ρ_HV ), and localized positive specific differential phase ( K_DP ) in the mature core. Solid-phase fractions maximized at 4–5 km AGL and were followed 30–45 min later by increased mixed-phase fractions at 2–3 km AGL in Mae Rim District. Elevated vertically integrated liquid density indicated an ice-rich convective core. Because phase classes are radar-derived proxies rather than direct hail observations, this sequence is interpreted as an event-specific phase-evolution timescale, not a validated warning lead time. The framework provides a basis for future multi-event evaluation of district-scale hail monitoring in tropical mountainous environments.
Monitoring crop residue burning (CRB) during the summer season is essential for identifying fire hotspots and to assess its impact on environmental and human health across North-Central India. In this study, we present a long-term, multi-satellite assessment of summer CRB across eight Indian states using active fire and burned area observations. By integrating polar-orbiting and geostationary satellite observations, we characterize the spatial distribution, temporal trends, burned area extent, and diurnal behaviour of summer CRB. Our analysis shows that Madhya Pradesh consistently emerges as the dominant summer CRB hotspot, contributing 38.4
