Pixel-based neural networks predict convective storm frequency from environmental variables with SSIM exceeding 0.8 while capturing diurnal cycles and orographic effects without explicit spatial or temporal inputs.
Title resolution pending
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
physics.ao-ph 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Using machine learning to downscale coarse-resolution environmental variables for understanding the spatial frequency of convective storms
Pixel-based neural networks predict convective storm frequency from environmental variables with SSIM exceeding 0.8 while capturing diurnal cycles and orographic effects without explicit spatial or temporal inputs.