Mapping technique of climate fields between GCM's and ice models

Abstract

Here, we present a mapping method OBLIMAP, which projects and interpolates fields like surface temperature, surface mass balance, and surface height between a geographical based coordinate system of a General Circulation Model (GCM) and a rectangular based Ice Model (IM). We derive an oblique stereographic projection and its inverse, which holds for any area at the Earth's surface, and which can be combined with two different interpolation methods. The first one is suited to interpolate the projected fields of a coarse GCM grid on a fine meshed IM grid. The second one is appropriate for the opposite case. Both grids are allowed to be arbitrary and irregularly spaced. Therefore the OBLIMAP technique is suitable for any GCM-IM combination. After a first scan of the GCM grid coordinates and the specification of the IM grid, fast mapping of various fields is possible. To and fro (GCM-IM-GCM) mapping tests with the Climate Community System Model (CCSM) at T42 resolution (~313 km) and the Regional Atmospheric Climate Model (RACMO) at ~11 km and ~55 km, show average temperature differences of less than 0.1 K with small standard deviations. OBLIMAP, available at GMD, is an accurate, robust and well-documented mapping method for coupling an IM with a GCM or to map state of the art initial and forcing fields available at geographical coordinates to any local IM grid with an optimal centered oblique projection. Currently, the oblique stereographic and the oblique Lambert azimuthal equal-area projections for both the sphere and the ellipsoid are implemented in OBLIMAP.