Space-time interpolation and uncertainty assessment of an extreme precipitation index using geostatistical cosimulation

Abstract

The rainfall regime in southern Portugal is clearly Mediterranean. In such regions, the research on space-time patterns of extreme precipitation is an important contribution to assess desertification dynamics, among other impact studies applications. For the space-time interpolation and uncertainty assessment of an extreme precipitation index, in the southern region of continental Portugal, we explore the application of direct sequential cosimulation (coDSS), which allows incorporating covariates such as altitude. The technique is illustrated using precipitation observations measured at 105 monitoring stations with data within the period 1970-2000. The results provide evidences of an increase of the spatial continuity of the extreme precipitation index in the last decades. The relationship between elevation and the index is decreasing along time. The promising results from this study indicate the coDSS technique as a valuable tool to deepen the knowledge on the space-time patterns of extreme precipitation indices.