0000000000424967
AUTHOR
Kalle Remm
Interpolative mapping of mean precipitation in the Baltic countries by using landscape characteristics
Maps of the long-term mean precipitation involving local landscape variables were generated for the Baltic countries, and the effectiveness of seven modelling methods was compared. The precipitation data were recorded in 245 meteorological stations in 1966–2005, and 51 location-related explanatory variables were used. The similarity-based reasoning in the Constud software system outperformed other methods according to the validation fit, except for spring. The multivariate adaptive regression splines (MARS) was another effective method on average. The inclusion of landscape variables, compared to reverse distance-weighted interpolation, highlights the effect of uplands, larger water bodies …
Precipitation pattern in the Baltic countries under the influence of large-scale atmospheric circulation and local landscape factors
The main objectives of the study are to analyse the mean precipitation pattern in the Baltic countries (Estonia, Latvia, Lithuania) and the influence of local landscape factors on it, to derive a regionalization, i.e. select regions having coherent fluctuations in precipitation, and to analyse relationships between precipitation and characteristics of the large-scale atmospheric circulation. Monthly precipitation data from 123 stations in the Baltic countries from 1966 to 2005 are analysed in relation to 52 landscape variables and 13 circulation variables. The landscape variables characterize landscape around the meteorological stations: land cover, land surface elevation, distance from the…