Analysis of flow signatures and catchment similarity indices for catchment classification in yesilirmak basin

Date

2019

Author

Soyugür, Batuhan

Metadata

Show full item record

This work is licensed under a Creative Commons Attribution-NonCommercial-NoDerivatives 4.0 International License.

Item Usage Stats

3
views

2
downloads

Catchment classification schemes aim to identify groups of hydrologically similar catchments to enable a mapping between catchment physical characteristics and hydro-climatic conditions with the catchment functioning. This mapping, together with the quantified uncertainties, potentially facilitates improved process understanding, transfer of this understanding to ungauged catchments, model parameter regionalization and hence improve operational applications and watershed management. Although many studies focusing on the topic of catchment classification exist in the literature, there is yet no consensus on the number and type of similarity metrics that should be included in such analysis. The aim of this study is to first carefully derive hydrologically relevant similarity metrics from catchment physical (topography, geology, soil, landcover), climatic (precipitation, evapotranspiration, aridity index) and hydrologic response characteristics (flow signatures), provide a basic framework for catchment classification, utilize Affinity Propagation clustering algorithm to determine the optimal number of groupings based on individual as well as a combination of these similarity metrics then compare results with multiple external clustering validation indices (Cramer’s V, Adjusted Rand Index etc.). The study area is comprised of 31 sub-catchments (varying between 6.8 km2 and 10048.8 km2) located in the Yesilirmak Basin, Turkey, where daily streamflow and monthly meteorological variables are available. The metrics that are based on flow signatures summarize a number of behavioral functions of the catchment system including those derived from flow duration curve (overall water balance, vertical redistribution) as well as temporal redistribution of flow (seasonality etc.). Comparison of the clusters obtained from flow signatures, catchment physical properties and climatic properties indicated that the climate properties are the most important descriptors of the flow signatures in the catchments. Secondarily, the topographic descriptors have the second correspondence with the flow signature clusters. After climate and topographic group of descriptors, land cover, geology and soil have the correspondence with the flow signatures, respectively.

Subject Keywords

Hydrology., Catchment similarity, Flow signatures, Cluster analysis, Affinity propagation algorithm.

URI

http://etd.lib.metu.edu.tr/upload/12623289/index.pdf
https://hdl.handle.net/11511/43438

Collections

Graduate School of Natural and Applied Sciences, Thesis