Comparison of the NLDAS Weather Forcing Model with Ground-Based Measurements over Agricultural Areas Throughout the Western United States
Location
Room 307/309
Event Website
http://water.usu.edu/
Start Date
4-10-2013 1:10 PM
End Date
4-10-2013 1:30 PM
Description
Clayton S. Lewis, Hatim M. E. Geli, Christopher M. U. Neale, James P. Verdin, and Gabriel Senay This analysis is being conducted in the context of standardizing and providing guidelines for evapotranspiration (ET) maps of crop water use estimates for the Western United States, a project funded by the USGS. ET estimate maps can be obtained using remote sensing-based methods at field/local scales and can be achieved with a reasonable accuracy. There are presently different remote sensing methods in the literature that provide reasonable levels of accuracy at these scales. Most of these methods use weather forcing data from ground-based weather stations with the assumption of being reasonably representative of the local conditions. Ground-based stations are generally sparse in nature for various reasons. However, at regional/continental scales this assumption will not be applicable and requires the use of other approaches to account for the variability of the near surface weather conditions. The North American Land Data Assimilation System (NLDAS), a gridded weather forcing dataset, can potentially provide this information. These NLDAS data are available at 1/8th of a degree (~ 6.8 miles × 8.6 miles), a relatively coarse resolution considering the requirement of estimating ET at field scales. In order to use the NLDAS weather forcing data in remote sensing of ET at field to regional scales with increased confidence it is important to compare it with ground-based observations. Such comparison will help to identify the associated uncertainties and biases. It can also help to quantify the uncertainties in the remote sensing-based model estimates of ET. This study presents a preliminary result of the comparison between ground-based weather data to the NLDAS gridded products. Ground-based, hourly data drawn from 16 networks throughout the West are used in the analysis. The comparison considered hourly temperature, humidity, precipitation, wind, and solar radiation data. Stations that were operated over non irrigated surfaces were thrown out as the study concentrates on those stations located in irrigated agricultural areas.
Comparison of the NLDAS Weather Forcing Model with Ground-Based Measurements over Agricultural Areas Throughout the Western United States
Room 307/309
Clayton S. Lewis, Hatim M. E. Geli, Christopher M. U. Neale, James P. Verdin, and Gabriel Senay This analysis is being conducted in the context of standardizing and providing guidelines for evapotranspiration (ET) maps of crop water use estimates for the Western United States, a project funded by the USGS. ET estimate maps can be obtained using remote sensing-based methods at field/local scales and can be achieved with a reasonable accuracy. There are presently different remote sensing methods in the literature that provide reasonable levels of accuracy at these scales. Most of these methods use weather forcing data from ground-based weather stations with the assumption of being reasonably representative of the local conditions. Ground-based stations are generally sparse in nature for various reasons. However, at regional/continental scales this assumption will not be applicable and requires the use of other approaches to account for the variability of the near surface weather conditions. The North American Land Data Assimilation System (NLDAS), a gridded weather forcing dataset, can potentially provide this information. These NLDAS data are available at 1/8th of a degree (~ 6.8 miles × 8.6 miles), a relatively coarse resolution considering the requirement of estimating ET at field scales. In order to use the NLDAS weather forcing data in remote sensing of ET at field to regional scales with increased confidence it is important to compare it with ground-based observations. Such comparison will help to identify the associated uncertainties and biases. It can also help to quantify the uncertainties in the remote sensing-based model estimates of ET. This study presents a preliminary result of the comparison between ground-based weather data to the NLDAS gridded products. Ground-based, hourly data drawn from 16 networks throughout the West are used in the analysis. The comparison considered hourly temperature, humidity, precipitation, wind, and solar radiation data. Stations that were operated over non irrigated surfaces were thrown out as the study concentrates on those stations located in irrigated agricultural areas.
https://digitalcommons.usu.edu/runoff/2013/AllAbstracts/36