# Hapi: A Multivariable Land-Surface Transformer for Medium-Range Hydrological Forecasting at Continental Scale

arXiv: https://arxiv.org/abs/2609.22702 · HTML: https://mlsanity.com/paper/2609.22702/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 22 of its 52 references that are in the map.
- First posted on arXiv 19 Sep 2026.

## Builds on (most cited first, 22 of 22)

- [U-Net: Convolutional Networks for Biomedical Image Segmentation](https://mlsanity.com/paper/1505.04597/) (2015)
- [Mamba: Linear-Time Sequence Modeling with Selective State Spaces](https://mlsanity.com/paper/2312.00752/) (2023)
- [RoFormer: Enhanced Transformer with Rotary Position Embedding](https://mlsanity.com/paper/2104.09864/) (2021)
- [Multi-Task Learning Using Uncertainty to Weigh Losses for Scene Geometry and Semantics](https://mlsanity.com/paper/1705.07115/) (2017)
- [Swin Transformer V2: Scaling Up Capacity and Resolution](https://mlsanity.com/paper/2111.09883/) (2021)
- [Perceiver IO: A General Architecture for Structured Inputs & Outputs](https://mlsanity.com/paper/2107.14795/) (2021)
- [Message Passing Neural PDE Solvers](https://mlsanity.com/paper/2202.03376/) (2022)
- [EVA-02: A Visual Representation for Neon Genesis](https://mlsanity.com/paper/2303.11331/) (2023)
- [ClimaX: A foundation model for weather and climate](https://mlsanity.com/paper/2301.10343/) (2023)
- [Gated Attention for Large Language Models: Non-linearity, Sparsity, and Attention-Sink-Free](https://mlsanity.com/paper/2505.06708/) (2025)
- [FuXi: A cascade machine learning forecasting system for 15-day global weather forecast](https://mlsanity.com/paper/2306.12873/) (2023)
- [Rotary Position Embedding for Vision Transformer](https://mlsanity.com/paper/2403.13298/) (2024)
- [Scaling transformer neural networks for skillful and reliable medium-range weather forecasting](https://mlsanity.com/paper/2312.03876/) (2023)
- [Towards Learning Universal, Regional, and Local Hydrological Behaviors via Machine-Learning Applied to Large-Sample Datasets](https://mlsanity.com/paper/1907.08456/) (2019)
- [Prithvi WxC: Foundation Model for Weather and Climate](https://mlsanity.com/paper/2409.13598/) (2024)
- [Differentiable, learnable, regionalized process-based models with physical outputs can approach state-of-the-art hydrologic prediction accuracy](https://mlsanity.com/paper/2203.14827/) (2022)
- [MC-LSTM: Mass-Conserving LSTM](https://mlsanity.com/paper/2101.05186/) (2021)
- [Flood forecasting with machine learning models in an operational framework](https://mlsanity.com/paper/2111.02780/) (2021)
- [AERIS: Argonne Earth Systems Model for Reliable and Skillful Predictions](https://mlsanity.com/paper/2509.13523/) (2025)
- [Rainfall-Runoff Prediction at Multiple Timescales with a Single Long Short-Term Memory Network](https://mlsanity.com/paper/2010.07921/) (2020)
- [RiverMamba: A State Space Model for Global River Discharge and Flood Forecasting](https://mlsanity.com/paper/2505.22535/) (2025)
- [Towards Generalized Hydrological Forecasting using Transformer Models for 120-Hour Streamflow Prediction](https://mlsanity.com/paper/2406.07484/) (2024)

## Abstract (arXiv)

Accurate flood forecasts several days in advance are essential for flood control, water-resource management, and emergency response. A central challenge is to produce high-resolution forecasts across continental domains where hydrological behavior varies widely from place to place. We developed Hapi, a U-Net Swin Transformer that uses fine three-dimensional patches and hierarchical shifted-window attention to forecast river discharge, surface runoff, snow water equivalent, and soil wetness index across the contiguous United States. The model produces medium-range forecasts (24--72~h) at 0.05circ resolution with adaptive task weighting and required only 0.11 seconds for a four-variable 72-h CONUS forecast on one A100 GPU. In a held-out 2024 potential-skill evaluation with ERA5-Land inputs prescribed over the forecast horizon, Hapi achieved the highest F1-score for floods in 20 of 21 comparisons across seven GloFAS return periods and three forecast leads. Independent validation against observed daily discharge at 3,881 U.S. Geological Survey gauges showed that Hapi achieved the highest median Nash--Sutcliffe efficiency at every lead, supported by regional-cluster bootstrap intervals. In a matched 24-h comparison of loss formulations, adaptive task balancing produced the lowest discharge errors and the highest F1-score for floods.