Evaluating downscaled products with expected hydroclimatic co-variances DOI Creative Commons
Seung H. Baek, Paul Ullrich, Bo Dong

et al.

Geoscientific model development, Journal Year: 2024, Volume and Issue: 17(23), P. 8665 - 8681

Published: Dec. 9, 2024

Abstract. There has been widespread adoption of downscaled products amongst practitioners and stakeholders to ascertain risk from climate hazards at the local scale (e.g., ∼ 5 km resolution). Such must nevertheless be consistent with physical laws credible value users. Here we evaluate statistically dynamically by examining co-evolution temperature precipitation during convective frontal events (two mechanisms testable just precipitation). We find that two widely used statistical downscaling techniques (Localized Constructed Analogs version 2, LOCA2, Seasonal Trends Analysis Residuals Empirical Statistical Downscaling Model, STAR-ESDM) generally preserve expected co-variances over historical future projected intervals as compared European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) observation-based data (Livneh nClimGrid-Daily). However, both dampen intensification is otherwise robustly captured in global models (i.e., prior downscaling) process-based dynamical across five different regional models. In case this leads appreciable underestimation event intensity. This study one first quantify a likely ramification stationarity assumption underlying methods identify phenomenon where projections change diverge depending on production method employed. Finally, our work proposes useful evaluation diagnostics can universally applied wide range products.

Language: Английский

A Non‐Sigmoidal‐Curve‐Dependent Dynamic Threshold Method Improves Precipitation Phase Partitioning in the Northern Hemisphere DOI Creative Commons
Lina Liu, Liping Zhang, Qin Zhang

et al.

Water Resources Research, Journal Year: 2025, Volume and Issue: 61(4)

Published: April 1, 2025

Abstract Given the significant impact of precipitation phase transitions on water and energy balances, accurate partitioning is essential for hydrological modeling. Many commonly used methods (PPMs) rely sigmoidal curve assumptions to determine thresholds, leading biased results. Here we developed a non‐sigmoidal‐curve‐dependent dynamic threshold method (NSDT) establish time‐varying spatially varying thresholds classifying into rain, snow, sleet in Northern Hemisphere. The NSDT avoids curve‐fitting errors by directly calculating from snowfall rainfall frequency curves. In this method, relative humidity elevation are two most influential variables phase, single‐threshold dual‐threshold strategies employed separately across different ranges. results show that station derived have marked spatial variability. Furthermore, performs well robustly, with accuracy exceeding 80% over wet‐bulb temperature range [−10°C, 10°C] at each range, subinterval, sub‐time period. outperforms six PPMs, especially high elevations. Regarding [−4°C, 4°C], exhibits improvements ranging 1.0% 11.8% (0.4%–14.5%) all (relative humidity) subintervals compared other PPMs. Overall, herein improves partitioning, which expected enhance simulation land surface models provide theoretical basis more understanding processes.

Language: Английский

Citations

0
A machine learning-based water supply forecasting model to quantify the impact of snow water equivalent on seasonal streamflow variability over the western U.S DOI
Haowen Yue, Yihan Wang, Lujun Zhang

et al.

Journal of Hydrology, Journal Year: 2025, Volume and Issue: 660, P. 133465 - 133465

Published: May 5, 2025

Language: Английский

Citations

0
Crowdsourced Data Reveal Shortcomings in Precipitation Phase Products for Rain and Snow Partitioning DOI Creative Commons
Guo Yu, Keith S. Jennings, Benjamin J. Hatchett

et al.

Geophysical Research Letters, Journal Year: 2024, Volume and Issue: 51(24)

Published: Dec. 23, 2024

Abstract Reanalysis products support our understanding of how the precipitation phase influences hydrology across scales. However, a lack validation data hinders evaluation reanalysis‐estimated phase. In this study, we used novel dataset from Mountain Rain or Snow (MRoS) citizen science project to compare 39,680 MRoS observations January 2020 July 2023 conterminous United States (CONUS) assess three products. These included Global Precipitation Measurement (GPM) mission Integrated Multi‐satellitE Retrievals for GPM (IMERG), Modern‐Era Retrospective Analysis Research and Applications (MERRA‐2), North American Land Data Assimilation System (NLDAS‐2). The overall critical success indices detecting rainfall (snowfall) IMERG, MERRA‐2, NLDAS‐2 were 0.51 (0.79), 0.49 (0.77), 0.54 (0.53), respectively. show that IMERG MERRA‐2 reasonably classify snowfall, whereas overestimates rainfall. All performed poorly in subfreezing snowfall above 2°C. Therefore, crowdsourced provides unique source improve capabilities reanalysis

Language: Английский

Citations

1
iRainSnowHydro v1.0: A distributed integrated rainfall-runoff and snowmelt-runoff simulation model for alpine watersheds DOI
Yuning Luo, Ke Zhang,

Yuhao Wang

et al.

Journal of Hydrology, Journal Year: 2024, Volume and Issue: 645, P. 132220 - 132220

Published: Oct. 22, 2024

Language: Английский

Citations

0
Evaluating downscaled products with expected hydroclimatic co-variances DOI Creative Commons
Seung H. Baek, Paul Ullrich, Bo Dong

et al.

Geoscientific model development, Journal Year: 2024, Volume and Issue: 17(23), P. 8665 - 8681

Published: Dec. 9, 2024

Abstract. There has been widespread adoption of downscaled products amongst practitioners and stakeholders to ascertain risk from climate hazards at the local scale (e.g., ∼ 5 km resolution). Such must nevertheless be consistent with physical laws credible value users. Here we evaluate statistically dynamically by examining co-evolution temperature precipitation during convective frontal events (two mechanisms testable just precipitation). We find that two widely used statistical downscaling techniques (Localized Constructed Analogs version 2, LOCA2, Seasonal Trends Analysis Residuals Empirical Statistical Downscaling Model, STAR-ESDM) generally preserve expected co-variances over historical future projected intervals as compared European Centre for Medium-Range Weather Forecasts Reanalysis v5 (ERA5) observation-based data (Livneh nClimGrid-Daily). However, both dampen intensification is otherwise robustly captured in global models (i.e., prior downscaling) process-based dynamical across five different regional models. In case this leads appreciable underestimation event intensity. This study one first quantify a likely ramification stationarity assumption underlying methods identify phenomenon where projections change diverge depending on production method employed. Finally, our work proposes useful evaluation diagnostics can universally applied wide range products.

Language: Английский

Citations

0