RT journal article
T1 Mapping the spatial variability of rainfall from a physiographic-based multilinear regression: model development and application to the Southwestern Iberian Peninsula
A1 Ruiz Ortiz, Verónica
A1 Isidoro, Jorge M.G.P.
A1 Fernandez, Helena Maria
A1 Granja Martins, Fernando M.
A1 García López, Santiago
A2 Ciencias de la Tierra
A2 Ingeniería Industrial e Ingeniería Civil
K1 Rainfall
K1 Physiography
K1 Multilinear regression
K1 Interpolation
K1 Map algebra
K1 Iberian Peninsula
AB A physiographic-based multilinear regression model supported by GIS was developed to estimate spatial rainfall variability in the Southwest Iberian Peninsula. The area study includes a wide diversity of landscape features and comprises four Portuguese regions and one Spanish province (totalizing 28,860 km2). The region suffers a very strong Mediterranean influence, with a major cleavage between winter and summer seasons. Thus, the analysis was carried out separately for the wet (October to March) and dry (April to September) semesters. From an initial set of 10 explanatory physiographic variables, five were selected to be used in the multilinear regression, as they allowed generating models by map algebra that fitted well with the last 40 years of monthly rainfall data records. These records were obtained from 163 weather stations, filtered from an initial set of 230 (142 stations in Portugal and 88 in Spain). The correlation between the physiographic-based multilinear regression model and a model obtained by interpolation from rainfall historical data showed to be good or very good in approximately 75% of the area under study. Results show that physiographic-based models can be effectively used to estimate rainfall where there is a lack of rain gauges, or to densify spatial resolution of rainfall between rain gauges.
PB Springer
SN 0167-6369
YR 2022
FD 2022
LK http://hdl.handle.net/10498/34861
UL http://hdl.handle.net/10498/34861
LA eng
DS Repositorio Institucional de la Universidad de Cádiz
RD 10-may-2026