Delineating riparian zones for entire river networks using geomorphological criteria

Abstract

Riparian zone delineation is a central issue for riparian and river ecosystem management , however, criteria used to delineate them are still under debate. The area inundated by a 50-yr flood has been indicated as an optimal hydrological descriptor for ripar-ian areas. This detailed hydrological information is, however, not usually available for 5 entire river corridors, and is only available for populated areas at risk of flooding. One of the requirements for catchment planning is to establish the most appropriate location of zones to conserve or restore riparian buffer strips for whole river networks. This issue could be solved by using geomorphological criteria extracted from Digital Elevation Models. In this work we have explored the adjustment of surfaces developed under 10 two different geomorphological criteria with respect to the flooded area covered by the 50-yr flood, in an attempt to rapidly delineate hydrologically-meaningful riparian zones for entire river networks. The first geomorphological criterion is based on the surface that intersects valley walls at a given number of bankfull depths above the channel (BF-DAC), while the second is based on the surface defined by a threshold value indicating 15 the relative cost of moving from the stream up to the valley, accounting for slope and elevation change (path distance). As the relationship between local geomorphology and 50-yr flood has been suggested to be river-type dependant, we have performed our analyses distinguishing between three river types corresponding with three valley morphologies: open, shallow vee and deep vee valleys (in increasing degree of valley 20 constrainment). Adjustment between the surfaces derived from geomorphological and hydrological criteria has been evaluated using two different methods: one based on exceeding areas (minimum exceeding score) and the other on the similarity among total area values. Both methods have pointed out the same surfaces when looking for those that best match with the 50-yr flood. Results have shown that the BFDAC approach 25 obtains an adjustment slightly better than that of path distance. However, BFDAC requires bankfull depth regional regressions along the considered river network. Results have also confirmed that unconstrained valleys require lower threshold values than 4046 HESSD 9, 4045-4071, 2012 constrained valleys when deriving surfaces using geomorphological criteria. Moreover, this study provides: (i) guidance on the selection of the proper geomorphological criterion and associated threshold values, and (ii) an easy calibration framework to evaluate the adjustment with respect to hydrologically-meaningful surfaces.