Grilled Cheese

ExploreLog inSign up
Terms of UsePrivacy PolicyCommunity StandardsHelpGet the app

Grilled Cheese is a product of Village Compute

Version devBuilt at: 2026-10-11 02:37:10 EDT

Explore

PostsPeople
LatestRanked
@geosus.bsky.socialOct 8, 2026, 9:06 PM

Trade can shift ecological pressures far beyond where goods are consumed.

Liang et al. (2026) show rising HANPP embodied in Belt and Road trade, increasingly transferring ecological burdens to resource-exporting regions.

doi.org/10.1016/j.ge...

#SustainableTrade #GeoSus

@geosus.bsky.socialSep 28, 2026, 4:37 AM

How is AI transforming ecology?

Wu, Luo, Liu et al. review 30 years of machine learning in ecology and explore the emerging shift toward human–AI collaborative research.

doi.org/10.1016/j.ge...

#AI #Ecology #GeoSus

Integrated Artificial Intelligence (AI) system architecture for future ecological study. The diagram depicts the closed-loop workflow of modern ecological research, integrating automated data acquisition, hybrid process-based AI modeling, and AI-assisted decision-making.
@geosus.bsky.socialSep 21, 2026, 3:51 AM

Can satellite fusion improve wildfire severity mapping?

Nguyen Van et al. (2026) find Landsat-9 is the strongest single sensor, while Sentinel-2 + Landsat-9 provides the most consistent gains across 40 US wildfires.

doi.org/10.1016/j.ge...

#Wildfire #GeoSus

PlanetScope-guided burn severity labeling supported by multisensor differenced Normalized Burn Ratio (dNBR) consensus. (i) Pre- and post-fire images from L8, S2, and L9 were collected to characterize fire-induced spectral changes. (ii) For each sensor, NBR was calculated from the near-infrared (NIR) and shortwave infrared (SWIR) 2 bands before and after the fire, and dNBR was derived as the difference between pre-fire and post-fire NBR. (iii) Sensor-specific dNBR maps were classified into four burn-severity classes: unburned, low, moderate, and high severity. These maps were used as auxiliary burn-severity indicators rather than as ground-truth labels. (iv) Agreement among L8, S2, and L9 dNBR classifications was used to identify conservative, low-uncertainty areas and to avoid pixels with inconsistent sensor responses. (v) Final reference labels were generated primarily through visual interpretation of 3 m PlanetScope imagery.
@geosus.bsky.socialSep 12, 2026, 9:54 PM

Rural settlement fragmentation varies markedly across China.

Huang et al. (2026) find that a 10% settlement expansion was associated with a 23.5% decline in ecological connectivity in coastal regions.

doi.org/10.1016/j.ge...

#LandUse #GeoSus

@geosus.bsky.socialSep 11, 2026, 11:28 PM

Can AI improve water allocation in drylands?

Lu et al. (2026) combine machine learning with multi-objective optimization to balance water use, economic efficiency and ecological needs at basin scale.

doi.org/10.1016/j.ge...

#WaterManagement #GeoAI #GeoSus

@geosus.bsky.socialSep 8, 2026, 2:11 AM

China's wind power expansion is delivering major climate benefits.

Long et al. (2026) estimate 3,672 Mt CO₂ avoided since 2006, with cumulative reductions potentially reaching 50,101 Mt by 2060.

doi.org/10.1016/j.ge...

#WindEnergy #GeoSus