2.2.3. Mapping Inundation
For each survey, we mapped inundation by digitizing a polygon around the wetted edge visible in the aerial imagery. We inferred visible boundaries where vegetation or shadows obscured the water’s edge. We estimated inundation area uncertainty for each survey based on the resolution of the imagery used to digitize survey features. This procedure resulted in two separate buffered polygons representing the upper and lower bounds of maximum and minimum inundation within the valley bottom.
We delineated each inundation survey polygon into three flow type classes on a continuum from more lotic (free-flowing) to more lentic (ponded, but still flowing). We defined these classes as follows: 1) free-flowing – not obstructed by river channel or by a channel-spanning structural element, 2) overflow flow that is spilling over active channel boundaries and onto the floodplain or otherwise exposed in channel surfaces (e.g., bars, benches and/or ledges) and 3) ponded structurally forced backwater creating a pond or pool upstream of a channel-spanning structural element (e.g., a beaver dam; Wheaton et al., 2015). Once inundation types were classified, we used these data to derive the total area of each inundation type. We then divided inundated area by the valley bottom area which gave us the percent of both total inundation and each inundation type, allowing for comparison of inundation across reaches. We also estimated the integrated wetted width by dividing the total inundated area by the valley bottom length.
To characterize the diversity of inundation types (a proxy for habitat complexity), we used the Shannon’s Evenness Index (also referred to as Shannon Equitability or Shannon Evenness) to calculate a value for each site and survey, a metric frequently used to describe spatial heterogeneity (e.g., Laurel & Wohl, 2019; Wyrick & Pasternack, 2014). The Shannon’s Evenness Index value is calculated as follows:
\begin{equation} SHEI=\ \frac{-\sum_{i=1}^{m}{(P_{i}*lnP_{i})}}{\ln v}\nonumber \\ \end{equation}
where Pi is equal to the proportion of the valley bottom occupied by each inundation type i and v is equal to the number of inundation types present in the valley bottom. In our study, v was equal to four to include the three inundation types (free-flowing, ponded, overflow) and dry conditions.