From Drops to Grid: Noise-Aware Spatio-Temporal Neural Process for Rainfall Estimation

Fuente: arXiv
Saved in:
Bibliographic Details
Main Authors: Sarabia, Rafael Pablos, Nyborg, Joachim, Birk, Morten, Assent, Ira
Format: Preprint
Published: 2026
Subjects:
Online Access:
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1866917467037630464
author Sarabia, Rafael Pablos
Nyborg, Joachim
Birk, Morten
Assent, Ira
author_facet Sarabia, Rafael Pablos
Nyborg, Joachim
Birk, Morten
Assent, Ira
contents High-resolution rainfall observations are crucial for weather forecasting, water management, and hazard mitigation. Traditional operational measurements are often biased and low-resolution, limiting their ability to capture local rainfall. Accurate high-resolution rainfall maps require integrating sparse surface observations, yet existing deep learning densification methods are hindered by rainfall's skewed, localized nature, noise, and limited spatio-temporal fusion. We present DropsToGrid, a Neural Process-based method that generates dense rainfall fields by fusing temporal sequences from noisy, irregularly distributed private weather stations with spatial context from radar. Leveraging multi-scale feature extraction, temporal attention, and multi-modal fusion, the model produces stochastic, continuous rainfall estimates and explicitly quantifies uncertainty. Evaluations on real-world datasets demonstrate that DropsToGrid outperforms both operational and deep learning baselines, generating accurate high-resolution rainfall maps with well-calibrated uncertainty, even when only few stations are available and in cross-regional scenarios.
format Preprint
id arxiv_https___arxiv_org_abs_2605_05912
institution arXiv
publishDate 2026
record_format arxiv
spellingShingle From Drops to Grid: Noise-Aware Spatio-Temporal Neural Process for Rainfall Estimation
Sarabia, Rafael Pablos
Nyborg, Joachim
Birk, Morten
Assent, Ira
Machine Learning
Computer Vision and Pattern Recognition
High-resolution rainfall observations are crucial for weather forecasting, water management, and hazard mitigation. Traditional operational measurements are often biased and low-resolution, limiting their ability to capture local rainfall. Accurate high-resolution rainfall maps require integrating sparse surface observations, yet existing deep learning densification methods are hindered by rainfall's skewed, localized nature, noise, and limited spatio-temporal fusion. We present DropsToGrid, a Neural Process-based method that generates dense rainfall fields by fusing temporal sequences from noisy, irregularly distributed private weather stations with spatial context from radar. Leveraging multi-scale feature extraction, temporal attention, and multi-modal fusion, the model produces stochastic, continuous rainfall estimates and explicitly quantifies uncertainty. Evaluations on real-world datasets demonstrate that DropsToGrid outperforms both operational and deep learning baselines, generating accurate high-resolution rainfall maps with well-calibrated uncertainty, even when only few stations are available and in cross-regional scenarios.
title From Drops to Grid: Noise-Aware Spatio-Temporal Neural Process for Rainfall Estimation
topic Machine Learning
Computer Vision and Pattern Recognition
url https://arxiv.org/abs/2605.05912