Temporal Resolution of Wind Forcing Required for River Plume Simulations. Issue 3 (5th March 2019)

• Record Type: Journal Article
• Authors: Qu, Lixin; Hetland, Robert D.
• Is Part Of: Journal of geophysical research. Volume 124:Issue 3(2019)
• Page Start: 1459
• Page End: 1473
• Publication Date: 2019-03-05
• Subjects: Oceanography -- Periodicals; 551.4605
• DOI: 10.1029/2018JC014593
• Languages: English
• ISSNs: 2169-9275

Abstract

Abstract: Wind is a primary forcing agent for river plume variability. Consequently, the temporal resolution of wind forcing is an important factor to consider for river plume simulations. This study evaluates river plume simulation errors caused by temporal subsampling of wind forcing data. We use an idealized model of a river plume over a continental shelf and force the model with temporally filtered winds to quantify the effect of temporal subsampling on simulation accuracy. The simulation error is proportional to the fraction of energy missing in the high‐frequency wind absent from the forcing. These results set requirements for temporal wind resolution in realistic simulations of river plumes. Spectral analysis of observed wind records at the Mississippi River, Columbia River, and Merrimack River regions indicates that, for simulation errors due to insufficient temporal wind resolution to be smaller than 5% of the variance, 3‐hourly or 4‐hourly wind data are reasonable. Though horizontal variations in wind forcing are lost, analyzing fast Fourier transformation spectrum of a single‐point wind measurement in the simulation region is helpful for estimating simulation errors due to temporal resolution and hence aid in properly selecting temporal resolutions. Plain Language Summary: Temporal subsampling of winds causes errors in river plume simulations. The primary finding of this study is that the simulation error is proportional to the fraction of the energy missing in the high‐frequency wind caused by subsampling. Analyzing the fast Fourier transformation spectrum of a single‐point wind measurement in the simulation region is helpful for estimating simulation errors due to temporal resolution. This study helps us better understand the importance of high‐frequency information in the wind and better select an appropriate temporal resolution of wind forcing for a river plume simulation. Key Points: River plume simulation error due to temporal resolution is proportional to the missing fraction of wind energy caused by subsampling Analyzing FFT spectrum of a single‐point wind measurement is helpful for estimating simulation errors