Paper No. 21
Presentation Time: 9:00 AM-6:30 PM

MODELING EXTREME FLOOD EVENTS IN AFGHANISTAN


JOHNSON, William C.1, KASTENS, Jude2, LEE, Eunmok2, DOBBS, Kevin E.3 and MACPHERSON, G.L.4, (1)Dept. of Geography, University of Kansas, 1475 Jayhawk Blvd, Rm. 213, Lawrence, KS 66045, (2)Kansas Biological Survey, Lawrence, KS 66045, (3)Kansas Applied Remote Sensing Program, University of Kansas, 124 Higuchi Hall, Lawrence, KS 66045, (4)Dept. of Geology, Univ of Kansas, 1475 Jayhawk Blvd, 120 Lindley Hall, Lawrence, KS 66045, wcj@ku.edu

Researchers at the Kansas Biological Survey (KBS) developed a topographic floodplain model (FLDPLN) that has been used to identify and map river valley floodplains for many rivers and stream networks, using elevation data ranging from 90-m SRTM to 1-m LiDAR. FLDPLN uses simple hydrologic principles to estimate floodplain extent as a function of floodwater depth: backfill flooding to approximate backwater effects, and spillover flooding to approximate floodwater rerouting. Resulting floodplain “depth to flood” (DTF) databases are completely seamless and modular. Consequently, FLDPLN can be used for river valley identification, flood extent estimation, hydrologic connectivity indexing, and also for scenario modeling (e.g., a particular flood event, or connectivity impacts of adding or removing a levee). Model outputs have applications for river valley morphology assessment, ecological modeling, wetland identification and extent mapping, new reservoir site selection, and flood disaster response mitigation and damage assessment, and can also facilitate traditional hydraulic studies. Water-stressed regions do have extreme flood events, as recent history has shown, which necessitates planning for not only extreme drought but also for catastrophic flood events. Accordingly, FLDPLN is being applied in Afghanistan to channel ways representing three levels within the drainage hierarchy: the Helmand River, its major tributary the Arghandab River (upon which Kandahar is situated), and an alluvial fan-dominated, high-energy drainage. U.S. military drone-acquired, high-resolution (~1 m) LiDAR x, y and z data are being used to develop the 5- and 10-m depth flood-event models. Because of the level of computing involved using these data, the model has also been run for lower-resolution databases, such as the re-sampled LiDAR database (5 m) and IFSAR data (~11 m; INTERMAP). Results provide a quantitative approach to evaluating the potential risk to urban/village infrastructure as well as to irrigation systems, agricultural fields and archaeological sites.