## Poster Session #1: UC Ballroom

#### Title

Quantifying Error in Regional Climate Models Using Data Assimilation

#### Presentation Type

Poster

#### Faculty Mentor’s Full Name

Marco Maneta

#### Faculty Mentor’s Department

Geosciences

#### Abstract / Artist's Statement

Predictions of the amount of hydrologic quantities such as precipitation are important because they provide the foundation for design of water resource management strategies. Regional climate models are used to make predictions of these quantities, but how well can they estimate the actual amount for a given area? Data Assimilation fuses ground observations with model data, yielding a more accurate estimate of the true state of a process. Using a computer algebra system to perform statistical calculations on model data and conveying the results in the context of a Geographical Information System, this correction can be applied to any type of model prediction, given there exists ground measurements of the quantity of interest. This study finds the amount of error, in the form of a correction, on model data by incorporating precipitation measurements collected at six Snotel stations within the Bitterroot Range in Southwest Montana. The correction calculated is a quantification of the amount of error in the prior probability distribution of predicted values. The corrected model is a reflection of the ability of Snotel stations to capture the amount of precipitation in the surrounding area. The corrected model in this study shows the model underestimated average daily precipitation in the high-relief topography of the Bitterroot Range, in the western section of the study area, and a negligible correction in the Bitterroot Valley and a portion of the Sapphire Range in the Northeast section of the study area. The results of data assimilation enhance the information from regional climate models, and provide insight on the influence of ground observations so a more accurate prediction can be made.

Quantifying Error in Regional Climate Models Using Data Assimilation

UC Ballroom

Predictions of the amount of hydrologic quantities such as precipitation are important because they provide the foundation for design of water resource management strategies. Regional climate models are used to make predictions of these quantities, but how well can they estimate the actual amount for a given area? Data Assimilation fuses ground observations with model data, yielding a more accurate estimate of the true state of a process. Using a computer algebra system to perform statistical calculations on model data and conveying the results in the context of a Geographical Information System, this correction can be applied to any type of model prediction, given there exists ground measurements of the quantity of interest. This study finds the amount of error, in the form of a correction, on model data by incorporating precipitation measurements collected at six Snotel stations within the Bitterroot Range in Southwest Montana. The correction calculated is a quantification of the amount of error in the prior probability distribution of predicted values. The corrected model is a reflection of the ability of Snotel stations to capture the amount of precipitation in the surrounding area. The corrected model in this study shows the model underestimated average daily precipitation in the high-relief topography of the Bitterroot Range, in the western section of the study area, and a negligible correction in the Bitterroot Valley and a portion of the Sapphire Range in the Northeast section of the study area. The results of data assimilation enhance the information from regional climate models, and provide insight on the influence of ground observations so a more accurate prediction can be made.