Benchmarking land surface models in high latitudes

Several groups of land surface modellers have set out a standard set of tests that will remain in place and be delivered to the modelling community as a benchmark. The idea is to draw a line in the sand at a particular moment in time and quantify the performance of the model, which can then be referred to later as changes are made. An example of such a set of benchmark tests was made for the UK community model JULES (Joint UK Land Environment Simulator). We set out to answer the following questions: 1. How well does the original model work with respect to some well established datasets? 2. How well does the model reproduce the seasonal river flows in the major rivers, the seasonal greening up and dying of the vegetation at the continental scale, and the seasonal fluxes of carbon dioxide (CO2) and evaporation