跳到正文
返回论文列表
cs.AI提交于 已译

面向气候降尺度的物理约束修正流:投影能证明什么?

Towards Physics-Constrained Rectified Flow for Climate Downscaling: What Projections Can Certify?

Arun Sharma

中文摘要

统计降尺度能恢复洪涝与作物胁迫背后的区域细节,但扩散与流匹配生成器可能产生尖锐场,破坏其粗输入所满足的守恒律。物理约束修正流(PC-RF)是一种条件流匹配降尺度器,其训练惩罚针对风散度、负降水和相对粗网格总量的质量漂移,采样后的投影则精确消除后两者。在一个无训练网络的受污染合成场上,投影将非负性违反率(NVR)从0.236、质量守恒误差(MCE)从0.184降至零,惩罚项使散度误差(DE)降低约30%。在ERA5上,投影使PC-RF与无惩罚的修正流基线RF-base的NVR恰为零,MCE在float32舍入范围内为零。惩罚项未带来可与训练预算或运行间波动区分开的感知增益。在2019-2021年匹配预算下,PC-RF相对RF-base无优势(SSIM 0.939 vs. 0.942,PSNR 31.81 vs. 31.79,CRPS 0.771 vs. 0.772)。未投影的CorrDiff复现在2020年PSNR(28.38)和CRPS(1.092)上均优于两个流模型,但其集合平均降水格点中45%为负。在ERA5上,散度算子的符号错误使该惩罚与DE作用于类形变项。精细参考由ERA5插值而来,且在更长窗口累积降水,因此感知分数无法确立降尺度技巧。NVR与MCE仅需一次前向且无需精细参考,故主张将其作为通过/失败检查先行运行,DE作为诊断指标。

关键要点

  1. 01投影可精确消除负降水与质量漂移,NVR和MCE在ERA5上归零
  2. 02散度惩罚使DE降低约30%,但感知指标无显著增益
  3. 03匹配预算下PC-RF与无惩罚基线RF-base性能相当
  4. 04CorrDiff复现PSNR/CRPS更优,但45%集合平均降水格点为负
  5. 05建议将NVR与MCE作为无需精细参考的通过/失败前置检查

解读

尚无解读。

原始英文摘要

arXiv:2611.22283v1 Announce Type: new Abstract: Statistical downscaling restores the regional detail behind flooding and crop stress, but diffusion and flow-matching generators can produce sharp fields that break the conservation laws of their coarse input. Physics-Constrained Rectified Flow (PC-RF) is a conditional flow-matching downscaler. Its training penalties target wind divergence, negative precipitation, and mass drift from the coarse total, and a projection after sampling removes the last two exactly. On one corrupted synthetic field with no trained network, the projection drives the Non-negativity Violation Rate (NVR) from 0.236 and the Mass-Conservation Error (MCE) from 0.184 to zero, and the penalties lower the Divergence Error (DE) by about 30%. On ERA5 the projection gives PC-RF and RF-base, the penalty-free rectified-flow baseline, an NVR of exactly zero and an MCE of zero up to float32 rounding. The penalties add no perceptual gain that we can separate from training budget or run-to-run variation. At a matched budget on 2019-2021, PC-RF has no edge over RF-base (SSIM 0.939 vs. 0.942, PSNR 31.81 vs. 31.79, CRPS 0.771 vs. 0.772). Our unprojected CorrDiff reimplementation beats both flow models on PSNR (28.38) and CRPS (1.092) on 2020, yet 45% of its ensemble-mean precipitation cells are negative. On ERA5 a sign error in the divergence operator made that penalty and DE act on a deformation-like term. The fine reference is interpolated from ERA5 and accumulates precipitation over a longer window, so the perceptual scores cannot establish downscaling skill. NVR and MCE need one pass and no fine reference, so we argue they should run first as pass-or-fail checks, with DE as a diagnostic.

同方向论文 · cs.AI

查看全部 →