Development of a Global Representative Hillslope Dataset for Earth System Models

Earth System models (ESMs) provide global projections of future climate states, and an important component of this task is the simulation of the exchange of water, energy, and carbon at the land surface. Topographic relief has a strong influence on on these exchanges through its effects on gravity-driven water redistribution and aspect-modulated surface radiation patterns. To date, however, the role of topography has rarely been accounted for explicitly in ESMs, in part due to the mismatch between the spatial scales at which ESM simulations are typically performed (10s to 100s of km) and those of topographic structures, which may have significant power down to the meter scale. One method of addressing this problem is through the use of "representative hillslopes", in which the hydrological and radiative effects of topography are explicitly modeled on a conceptual hillslope, and the results are subsequently upscaled to the ESM grid scale. However, due to the multiscale nature of topographic relief, the spatial scale of hillslopes that is "most representative" can be ambiguous. In this study, we describe a method for objectively identifying the spatial scale of the most influential topographic features. Stream networks and the associated catchments are then determined at the identified spatial scale, and the resulting geomorphic parameters (e.g. height above nearest drainage, slope, aspect) are used to construct a representative hillslope. We will present the results of applying this methodology globally using a 90m digital elevation model to create a gridded dataset of representative hillslopes at approximately 1 degree spatial resolution that can be used in an ESM.