💾 Archived View for radia.bortzmeyer.org › fosdem › event-11441.gmi captured on 2023-12-28 at 20:15:53. Gemini links have been rewritten to link to archived content
⬅️ Previous capture (2023-06-14)
-=-=-=-=-=-=-
Huidae Cho
Type devroom
Improving the performance of geospatial computation for web-based hydrologic modeling
Starts on day 2 (2021-02-07) at 14:40 (Brussels time, UTC+1) in room Geospatial (duration 00:40)
Matrix room #geospatial:fosdem.org
The longest flow path is one of the most important geospatial parameters that is used for hydrologic analysis and modeling. However, there are not many available GIS tools that can compute this watershed parameter. At the same time, there have been almost little to no efforts in improving its computational efficiency since its first, to the presenter's best knowledge, introduction by Smith (1995) when the geospatial data resolution was relatively coarser. In this talk, the presenter introduces a new algorithm that applies Hack's law to the discovery of the longest flow path and its efficient implementation as a GRASS module called r.accumulate. He compares its performance to that of commercial ArcHydro's Longest Flow Path tool. Lastly, he introduces a proof-of-concept version of the Web-based Hydrologic Modeling System (WHydroMod) built using GRASS, PyWPS, MapServer, and OpenLayers, and discusses how r.accumulate can be used to improve the efficiency of geospatial computation for WHydroMod.