Extension of a-CAMF to Select Groups of Contiguous Cells for Intervention: Computational Cost Vs. Solution Quality

EasyChair Preprint 11022

Abstract

The CAMF method selects in a raster geo-database of a hydrological catchment those cells for which a certain intervention, e.g. afforestation, leads to maximal reduction of sediment export through the outlet of the catchment. This optimization problem is solved by an iterative process in which independent cells with maximal capacity for sediment yield reduction in the catchment are selected. An accelerated version of CAMF (A-CAMF) was implemented with the aim to reduce the computational cost of the iterative process. The cells selected by A-CAMF are not always in the same region and may not be contiguous, which is a disadvantage for some applications, where performing the intervention at several distributed sites requires extra effort and can causes extra costs. In order to provide alternative solutions for these scenarios using A-CAMF, we extend the method to evaluate the selection of sets of contiguous cells in the minimization of sediment yield in a river catchment. We compare the performance of A-CAMF with and without region-growing (in solution quality and execution time) for minimizing sediment outflow using a raster geo-database of the Tabacay catchment in Ecuador.

Keyphrases: Afforestation, Minimizing flow, Sediment yield reduction, contiguity, region growing

@booklet{EasyChair:11022,
  author    = {Grethell Castillo Reyes and Dirk Roose and Gerdys Ernesto Jiménez Moya and René Estrella and Floris Abrams and Jos Van Orshoven},
  title     = {Extension of a-CAMF to Select Groups of Contiguous Cells for Intervention: Computational Cost Vs. Solution Quality},
  howpublished = {EasyChair Preprint 11022},
  year      = {EasyChair, 2023}}