Interpolation of rainfall for the Albuquerque area : a comparison to the Primary Local Climatological Data Site

A Professional Project submitted in partial fulfillment of the requirements for the degree of Master of Water Resources, Water Resources Program, University of New Mexico.The main objective of this project was to assess the ability of the Primary Local Climatological Data Site (LCD) to estimate precipitation around the City of Albuquerque. In other words, this study determined how representative LCD precipitation data is as a point source station for characterizing precipitation within the Albuquerque metro area and nearby sites. Because the LCD is located in a permanent position near the Albuquerque International Airport, precipitation associated with rainfall localized away from the LCD, but within the Albuquerque metro area, is not accounted for in the LCD precipitation data record. This study quantified regional variability in rainfall that cannot be captured by any single station by combining three primary source precipitation programs into a single comprehensive precipitation monitoring network (StudyNet) for the study period 2000 – 2010. Sub-regions within the StudyNet Region were identified and subjectively demarcated to include variations in topography, elevation, predominant wind and weather patterns, and areas with higher and lower densities of monitoring stations. Estimated Annual Precipitation (EAP) was determined by evaluating the results from objective seasonal spatial interpolations. Using estimated precipitation range for each interpolation, displayed by color scales, minimum and maximum precipitation levels were assigned for each sub-region. Maximum and minimum estimated precipitation values associated with each sub-region were then averaged to determine the EAP for each sub-region. For additional analysis, sub-regional EAP results were summed and averaged for each year to determine the StudyNet Estimated Annual Precipitation (StudyNetEAP). As results show, the LCD values were closest to spatially averaged precipitation in the Spring Season precipitation with a strong positive Pearson correlation of .95 and the lowest root mean squared difference (.40) in the Spring. Winter Season precipitation was underestimated by the LCD 64% of the time and had a moderately positive correlation coefficient (.72). During wet periods, especially the monsoon season when convective precipitation patterns display high spatial variability, the LCD frequently underestimates yearly precipitation, 90% of the time, with an average underestimation of 4.82 in/yr. Water Year precipitation was shown to have the highest root mean squared difference (2.09) and had a strong positive Pearson’s correlation of .87 between LCD and StudyNetEAP values. Quantifying seasonal precipitation shows that the LCD systematically underestimated StudyNetEAP in every season, and also in the annual mean. During dry periods the LCD was better at capturing both sub-regional and StudyNet wide precipitation. During times of sporadic and intense storm events the LCD often missed sub-regional precipitation events and consistently underestimated total precipitation for the StudyNet region. However, when LCD measurements are adjusted by sub-regional factor differences, the constant bias is removed. After removing the sub-regional bias the LCD measurement can be used to estimate seasonal sub-regional precipitation and thus regional precipitation for the greater Albuquerque area. Additionally, high seasonal correlations between LCD precipitation and StudyNetEAP show that the LCD provides a useful metric for determining dry and wet years for the StudyNet region as a whol