Statistical downscaling of daily maximum and minimum temperatures in the Central Pampas region

BETTOLLI, MARIA LAURA; PENALBA, OLGA C; RAGGIO, GABRIELA A

Workshop; WCRP VAMOS/CORDEX Workshop on Latin-America and Caribbean CORDEX LAC: Phase I - South America; 2013

Empirical downscaling is a widely used technique for exploring the regional and local-scale response to large-scale circulation. Different statistical techniques have been explored in many regions of the world. However, statistical downscaling (SD) has not been extensively developed for South America regions. The La Plata Basin in Southeastern South America, is the second most important basin of the continent after the Amazon basin. Its morphologic and climatic variety has generated a diversity of interests related to its hydrological resources. The central Pampas is extending itself towards the south of the basin and is the most important region of agricultural production in Argentina. Therefore, the exploration and development of statistical downscaling techniques are of special interest for the region. The purpose of this work is to downscale daily maximum and minimum temperatures at local scales in the Central Pampas using the analogue statistical downscaling method. To this end, daily mean fields of the NCEP-NCAR Reanalysis 1 in the area bounded by 60°S, 15°S, 90°W and 44.5°W were used as predictor variables. Daily maximum (Tmax) and minimum (Tmin) temperatures from 25 meteorological stations located in the Central Pampas region were also used for the period 1979-2000. This information was provided by the Argentina’s National Weather Service. The statistical downscaling is based on the analogue method which consists in finding, for each day in the record, the most similar days (called analogue fields). This similarity between the fields was assessed by the Euclidean distance. For each day, the memory effect was filtered (in a window of ± 5 days) and the first 10 analogue fields were retained. With the surface information (Tmax and Tmin) of each of these 10 analogues, five different types of estimations were performed: 1) the average, 2) the median, 3) the first analogue field, 4) the weighted average giving more importance to first analogues and 5) the temperature value of a random analogue. For the two variables and for each type of estimation, different validation measures were considered. The bias, the absolute error (AE) and the mean square error (RMSE) were analyzed. The correlation between the estimated and the observed time series was also performed. The analysis was focused on evaluating the ability of the method in representing spatial and temporal variability of the variables as well as the annual cycle. The results show that the best predictor of Tmax and Tmin was air temperature at 850 hPa. The method has a good performance at estimating spatial distributions, seasonal cycles and temporal variability. Both variables and all estimations showed annual cycle correlations greater than 0.98, though winter temperatures tend to be overestimated. The estimations of Tmin were better than for Tmax. However, for both temperatures, AE and RMSE errors showed values between 0 and 1°C, specially for the estimations (1), (2) and (4). Winter interannual correlations were mostly between 0.7 and 0.81, while summer interannual correlations showed values between 0.64 and 0.87. In all cases, the spatial distribution of estimated fields is in agreement with the observed fields. The statistical model reproduces the spatial structure of isoline distributions and also the gradients.