Environmental Variables Used in Maxent Machine Learning Approach

<p> Six environmental variables are represented as raster maps, and they are used in conjunction with locations of downed trees to inform the variable importance and relationships derived from results of the Maxent ML approach. These variables are processed as ESRI ASCII raster maps (*.asc), and input into the Maxent Model (Phillips et al. 2006). Four environmental variables (elevation, generalized slope, Euclidean Distance From River [EDFR], Inundation Probability) are raster maps that have a resolution (1.0m) and pixel locations coincident to the topography raster map that is input into HEC-RAS and featured in Castillo et al. (under review). Two classified variables (vegetation community, soil type) are informed by data form Ecological Mapping Systems of Texas (Elliott et.al., 2014), but transformed rasters with the same spatial information as those of the four environmental variables</p>
<p> Castillo, C. R., Güneralp, I, Hales, B. U., Güneralp, B. (in review). Scale-Free Structure of Surface-Water Connectivity within a Lowland River-Floodplain Landscape. Geophysical Research Letters.</p>
<p> Elliott, L.F., Treuer-Kuehn, A., Blodgett, C.F., True. C.D., German, D. Diamond, D.D. (2014). Ecological System of Texas: 391 Mapped Types, edited by Texas Parks and Wildlife Department and Texas Water Development Board, Austin, TX. url: https://tpwd.texas.gov/landwater/land/programs/landscape-ecology/ems/</p>
<p> Phillips, S.J., Anderson, R.P. Schapire, R.E. (2006). Maximum Entropy modelling of species geographic distribution. Ecological Modelling, 190, 231-259.</p>
<p> Zevenbergen, L.W. and C.R. Thorne. (1987), Quantitative analysis of land surface topography. Earth Surface Processes and Landforms, 12, 47-56.</p>