2025-08-21 パシフィック・ノースウェスト国立研究所 (PNNL)
Models overproduce precipitating storm cells like the one shown here. This leads to overestimates of heavy rain rates and underestimates of light rain rates relative to observation.
(Image by Jared Rackley | NOAA National Weather Service)
＜関連情報＞
- https://www.pnnl.gov/publications/increasing-model-spatial-resolution-fails-reduce-simulated-storm-biases
- https://agupubs.onlinelibrary.wiley.com/doi/10.1029/2024JD041090
シミュレートされた対流セルおよびシステム成長のバイアスが大気不安定性とモデル解像度に依存する要因 Dependencies of Simulated Convective Cell and System Growth Biases on Atmospheric Instability and Model Resolution
Zhixiao Zhang, Adam C. Varble, Zhe Feng, James N. Marquis, Joseph C. Hardin, Edward J. Zipser
Journal of Geophysical Research: Atmospheres Published: 22 November 2024
DOI:https://doi.org/10.1029/2024JD041090
Abstract
This study evaluates convective cell properties and their relationships with convective and stratiform rainfall within a season-long convection-permitting weather research and forecasting simulation over central Argentina using radar, satellite, and radiosonde measurements from the RELAMPAGO-CACTI field campaign. The simulation slightly underestimates radar-estimated rainfall over the ∼3.5-month evaluation period but underestimates stratiform rainfall by 46% and overestimates convective rainfall by 43%. As convective available potential energy (CAPE) increases, the convective rainfall overestimation decreases, but the stratiform rainfall underestimation increases such that the contribution of convective to total rainfall remains constantly high biased by ∼26%. Overestimated convective rainfall arises from the simulation generating 2.6 times more precipitating convective cells (14,299) than observed by radar (5,662) despite similar observed and simulated cell growth processes, with relatively wide cells contributing mostly to excessive convective rainfall. Relatively shallow cells, typically reaching heights of 4–7 km, contribute most to the cell number bias. This cell number bias increases as CAPE decreases, potentially because cells and their updrafts become narrower and more under-resolved as CAPE decreases. The gross overproduction of precipitating shallow cells leads to overly efficient precipitation and inadequate detrainment of ice aloft, thereby diminishing the formation of robust stratiform rainfall regions. Decreasing model horizontal grid spacing from 3 to 1 or 0.333 km for low (<300 J kg−1) and high CAPE (>1,000 J kg−1) cases results in minimal change to cell number, depth, and convective-to-stratiform partitioning biases. This suggests that improving prediction of these convective properties depends on factors beyond solely increasing model resolution.
Key Points
- A convection-permitting simulation overestimates the convective contribution to total rainfall, while underestimating stratiform rainfall
- A large excess of simulated shallow convective cells increases as instability decreases, contributing to the stratiform rainfall bias
- Increasing model resolution does not improve convective cell and convective-stratiform rainfall partitioning biases
Plain Language Summary
The ability of a storm-resolving weather model to predict rainfall over central Argentina was evaluated with data from a field campaign. Although the model accurately predicted the total amount of rain, it produced far too much relatively heavy rainfall and not enough light rainfall. The overestimation of heavy rainfall increased as the atmosphere became less favorable for intense storms, which correlated with far too many predicted storm cells, especially ones that were relatively shallow. The excessive frequency of storm cells prevented the formation of widespread lighter rainfall that was much more frequent in observations. Increasing the spatial resolution of the model to better resolve storm circulations did not improve predictions, suggesting model representation of storm precipitation formation, and growth processes requires improvement beyond model resolution to better predict storm rainfall intensities.
