TY - JOUR

T1 - Spatial interpolation of climate variables in Northern Germany

T2 - Influence of temporal resolution and network density

AU - Berndt, C.

AU - Haberlandt, U.

N1 - Funding Information: The authors would like to thank the German Weather Service for providing the meteorological observations and Ross Pidoto for correcting an early version of the manuscript. This research was partly funded by the German Environment Agency (RESTNI, project number: 3710422182) and the Lower Saxony Water Management, Coastal Defence and Nature Conservation Agency (KliBiW, no project number available). The article publication was funded by the Open Access fund of Leibniz Universität Hannover. Finally, the reviewer and the editor are gratefully acknowledged for their contributions.

PY - 2018/2

Y1 - 2018/2

N2 - Study region: Region in Lower Saxony (North Germany) covered by the measuring range of the weather radar device located near Hanover (approx. 50.000 m2). Study focus: This study investigates the performance of various spatial interpolation techniques for climate variables. Meteorological observations are usually recorded as site-specific point information by weather stations and estimation accuracy for unobserved locations depends generally on station density, temporal resolution, spatial variation of the variable and choice of interpolation method. This work aims to evaluate the influence of these factors on interpolation performance of different climate variables. A cross validation analysis was performed for precipitation, temperature, humidity, cloud coverage, sunshine duration, and wind speed observations. Hourly to yearly temporal resolutions and different additional information were considered. New hydrological insights: Geostatistical techniques provide a better performance for all climate variables compared to simple methods Radar data improves the estimation of rainfall with hourly temporal resolution, while topography is useful for weekly to yearly values and temperature in general. No helpful information was found for cloudiness, sunshine duration, and wind speed, while interpolation of humidity benefitted from additional temperature data. The influences of temporal resolution, spatial variability, and additional information appear to be stronger than station density effects. High spatial variability of hourly precipitation causes the highest error, followed by wind speed, cloud coverage and sunshine duration. Lowest errors occur for temperature and humidity.

AB - Study region: Region in Lower Saxony (North Germany) covered by the measuring range of the weather radar device located near Hanover (approx. 50.000 m2). Study focus: This study investigates the performance of various spatial interpolation techniques for climate variables. Meteorological observations are usually recorded as site-specific point information by weather stations and estimation accuracy for unobserved locations depends generally on station density, temporal resolution, spatial variation of the variable and choice of interpolation method. This work aims to evaluate the influence of these factors on interpolation performance of different climate variables. A cross validation analysis was performed for precipitation, temperature, humidity, cloud coverage, sunshine duration, and wind speed observations. Hourly to yearly temporal resolutions and different additional information were considered. New hydrological insights: Geostatistical techniques provide a better performance for all climate variables compared to simple methods Radar data improves the estimation of rainfall with hourly temporal resolution, while topography is useful for weekly to yearly values and temperature in general. No helpful information was found for cloudiness, sunshine duration, and wind speed, while interpolation of humidity benefitted from additional temperature data. The influences of temporal resolution, spatial variability, and additional information appear to be stronger than station density effects. High spatial variability of hourly precipitation causes the highest error, followed by wind speed, cloud coverage and sunshine duration. Lowest errors occur for temperature and humidity.

KW - Climate data

KW - Geostatistics

KW - Interpolation

KW - Kriging

UR - http://www.scopus.com/inward/record.url?scp=85044443301&partnerID=8YFLogxK

U2 - 10.1016/j.ejrh.2018.02.002

DO - 10.1016/j.ejrh.2018.02.002

M3 - Article

AN - SCOPUS:85044443301

VL - 15

SP - 184

EP - 202

JO - Journal of Hydrology: Regional Studies

JF - Journal of Hydrology: Regional Studies

ER -