Greve, Peter (2012) Evaluating soil water content in a WRF-NOAH downscaling experiment. (Diploma thesis), Universität Hohenheim, Stuttgart und Kiel, 90 pp.

Text Dipl. 2012 Greve, P.pdf - Reprinted Version Restricted to Registered users only Available under License German copyright act UrhG. Download (10MB) | Contact

Abstract

The main objective of this thesis is to evaluate the performance of the NOAH land surface model (LSM) using soil water content (SWC). SWC is one of the key variables in land surface physics, controlling the coupled energy and land water balance. The evaluation is conducted by comparing SWC generated by NOAH with in-situ station observations and with SWC generated by two other LSM. These are 1) TESS EL implemented in the ERA-Interim global reanalysis data set and 2) the Climate Prediction Center global reanalysis of SWC. The NOAH data was generated by an hindcast evaluation run covering Europe within the period from 1990 to 2008, using the Weather Research and Forecasting (WRF) model. Statistical analysis indicates an overall well performance of WRF-NOAH. The spatial patterns of seasonal mean SWC are quite similar, but absolute values differ much between the models, especially at a regional scale. It is shown that this disagreement most likely appears due to differences in precipitation, model physics and adjustments of SWC by the data assimilation system in ERA-Interim. Patterns of variability do correspond well in space and time. The most dominant mode of variability is the annual cycle, being largest in southern and western Europe. At interannual time scales, all models show patterns similar to those of the North A tlantic Oscillation (NAO), despite correlations with the NAO time series are rather small. Evaluation against station observations in southern France and southern Germany is performed on a 6-hourly basis within a two year period covering 2007 and 2008. It is shown that annual and daily cycles are the most important modes of variability. Despite WRF-NOAH generates similar structures, amplitudes are significantly larger. In winter the areal mean SWC of WRF-NOAH is too large in comparison with the observations, which is most probably due to an overestimation of precipitation. Too low SWC during summer in southern France is possibly related to an overestimation of evapotranspiration in a cropland environment, whereas at the few stations located in forest and at sandy soils, WRF-NOAH's SWC is significantly larger. It is pointed out which aspects of the model physics may lead to those differences.

Actions (login required)

View Item