Loading...
Please wait, while we are loading the content...
Similar Documents
Parameter Estimation in a Land-Surface Model Using Atmospheric Data Assimilation: Finding Distributions for Use in an Ensemble Prediction System
| Content Provider | Semantic Scholar |
|---|---|
| Author | Hacker, Joshua P. |
| Copyright Year | 2006 |
| Abstract | One of the difficulties with mesoscale prediction in the PBL is a fundamental lack of model fidelity. Errors in the mean behavior (biases) can be estimated and accounted for if a large number of cases and a good observing system are available. But probabilistic prediction using ensemble techniques relies on a model that effectively reproduces error growth. That is, a forecast from a perturbed initial state needs to diverge from an unperturbed forecast at approximately the same rate as either forecast diverges from the true PBL evolution. It is well known that mesoscale models, where the resolved-scale effect of PBL turbulence is usually parameterized, suffer from a lack of variability when compared to the real atmosphere. This results in a fundamental lack of error growth in the model, and difficulty producing a useful ensemble system for PBL forecasts. Lack of internal variability in models has lead to a proliferation of the so-called “multi-model” ensembles for mesoscale prediction (e.g. Hou et al. 2001; Grimit and Mass 2002; Stensrud and Yussouf 2003). Multimodel ensembles can be formed by varying PBL parameterization schemes within a single dynamical modelling framework such as the Weather Research and Forecast (WRF) model or the Penn State/NCAR Mesoscale Model (MM5). They have demonstrated a skillful ensemble mean, relative to the individual members, and |
| File Format | PDF HTM / HTML |
| Alternate Webpage(s) | https://ams.confex.com/ams/pdfpapers/111207.pdf |
| Language | English |
| Access Restriction | Open |
| Content Type | Text |
| Resource Type | Article |
