An approach to computing topographic wetness index based on maximum downslope gradient

摘要

As an important topographic attribute widely-used in precision agriculture, topographic wetness index (TWI) is designed to quantify the effect of local topography on hydrological processes and for modeling the spatial distribution of soil moisture and surface saturation. This index is formulated as TWI = ln(a/tan beta), where a is the upslope contributing area per unit contour length (or Specific Catchment Area, SCA) and tan beta is the local slope gradient for estimating a hydraulic gradient. The computation of both a and tan beta need to reflect impacts of local terrain on local drainage. Many of the existing flow direction algorithms for computing a use global parameters, which lead to unrealistic partitioning of flow. beta is often approximated by slope gradient around the pixel. In fact, the downslope gradient of the pixel is a better approximation of beta. This paper examines how TWI is impacted by a multiple flow routing algorithm adaptive to local terrain and the employment of maximum downslope gradient as beta. The adaptive multiple flow routing algorithm partitions flow by altering the flow partition parameter based on local maximum downslope gradient. The proposed approach for computing TWI is quantitatively evaluated using four types of artificial terrains constructed as DEMs with a series of resolutions (1, 5, 10, 20, and 30 m), respectively. The result shows that the error of TWI computed using the proposed approach is generally lower than that of TWI by the widely used approach. The new approach was applied to a low-relief agricultural catchment (about 60 km(2)) in the Nenjiang watershed, Northeastern China. The results of this application show that the distribution of TWI by the proposed approach reflects local terrain conditions better.