Quantifying measurement uncertainty and spatial variability in the context of model evaluation
Abstract
In an effort to improve wind forecasts for the wind energy sector, the Department of Energy and the NOAA funded the second Wind Forecast Improvement Project (WFIP2). As part of the WFIP2 field campaign, a large suite of in-situ and remote sensing instrumentation was deployed to the Columbia River Gorge in Oregon and Washington from October 2015 - March 2017. The array of instrumentation deployed included 915-MHz wind profiling radars, sodars, wind- profiling lidars, and scanning lidars. The role of these instruments was to provide wind measurements at high spatial and temporal resolution for model evaluation and improvement of model physics. To properly determine model errors, the uncertainties in instrument-model comparisons need to be quantified accurately. These uncertainties arise from several factors such as measurement uncertainty, spatial variability, and interpolation of model output to instrument locations, to name a few. In this presentation, we will introduce a formalism to quantify measurement uncertainty and spatial variability. The accuracy of this formalism will be tested using existing datasets such as the eXperimental Planetary boundary layer Instrumentation Assessment (XPIA) campaign. Finally, the uncertainties in wind measurement and the spatial variability estimates from the WFIP2 field campaign will be discussed to understand the challenges involved in model evaluation.
- Publication:
-
AGU Fall Meeting Abstracts
- Pub Date:
- December 2017
- Bibcode:
- 2017AGUFMGC24F..04C
- Keywords:
-
- 3305 Climate change and variability;
- ATMOSPHERIC PROCESSES;
- 3360 Remote sensing;
- ATMOSPHERIC PROCESSES;
- 1620 Climate dynamics;
- GLOBAL CHANGE;
- 1878 Water/energy interactions;
- HYDROLOGY