GPU batched Penman-Monteith PET — 34.9x speedup at 2.7M catchments, PASS

[consigcody94]

Summary

The Penman-Monteith evapotranspiration calculation can run on GPU with one thread per grid cell. At 2.7M catchments, this achieves 34.9x speedup with max relative error 1.44e-04 (from GPU fast exp).

Benchmark (RTX 3060)

Catchments	CPU	GPU	Speedup	Max rel error
100K	1 ms	0.048 ms	20.7x	5.16e-06
1M	9 ms	0.243 ms	37.0x	5.00e-05
2.7M (NWM)	24 ms	0.687 ms	34.9x	1.44e-04

Zero NaN, zero failures. The Penman-Monteith equation (saturation vapor pressure, psychrometric constant, aerodynamic resistance) maps naturally to GPU — each grid cell is independent.

Code

https://github.com/consigcody94/parallel-prefix-rt/blob/master/benchmarks/cuda/owp_extended_kernels.cu

[consigcody94]

Code has been reorganized to: https://github.com/consigcody94/noaa-gpu-kernels (water-prediction/owp_extended_kernels.cu)

[consigcody94]

AI Attribution: This work was developed using Claude Code as an AI-assisted development tool. Code generation and benchmark automation were performed by Claude under my direction. I designed the research methodology, selected mathematical approaches, and validated all results on physical hardware (RTX 3060). All commits carry a Co-Authored-By: Claude tag. When results failed validation (e.g., tensor compression on rte-rrtmgp#394), they were publicly retracted.