PHOTOGRAMMETRIC ENGINEERING & REMOTE SENSING
November 2016
833
GIS
&
Tips Tricks
By Al Karlin, Ph.D.,CMS, GISP
O
ne of the four major missions of the Southwest Flori-
da Water Management District (SWFWMD) involves
Flood Prevention. Toward that mission, the District
uses Hydrologic and Hydraulic (H&H) models to identify flood-
prone areas, and to implement best management practices to
assist in preventing flooding. As the starting point for most
H&Hmodels is a representation of terrain, the District collects
LiDAR (Light Detection and Ranging) point clouds and break-
lines to construct hydro-flattened and hydro-enforced Digital
Terrain Models (DTMs).
Through the years, as the LiDAR point clouds have become
less costly to obtain, LiDAR-handling software has become
more available, and budgets have become tighter, the District
has been using internal resources to construct and/or update
DTMs. Typically the District uses GeoCue LP360™, as either
the Esri™ Add-In or in the stand-alone 64-bit version for con-
structing the raster DTMs for a watershed.
In a scenario that we have seen often, the LiDAR vendor
“bulldozes” buildings, i.e., removes the LiDAR points falling
on buildings from the terrain, and a “basement sink” results
from errant triangulation of ground points along the sides of
the removed building footprint that remain on the ground
class. These “basement sinks” need to be slightly elevated,
flattened and the DTM updated, otherwise water will accu-
mulate in the sink. The following workflow demonstrates
how we repair a “basement” using LP360 and the Con(dition-
al) “IsNull” function from Spatial Analyst, but can be applied
to any portion of a DTM that requires updating/replacement.
In this example, the building pictured in the top right is rep-
resented in the bottom right in the DTM. Clearly the “bull-
dozed” building is not properly represented in the DTM as
water will flow along the “channels” and into the “basement”.
Fixing “Basement Sinks” and updating Digital Terrain Models
Step 1
Capture the Building footprint as a 2-D
shapefile
Use ArcGIS and heads-up digitize the building footprint (or
roofline) as a polygon shapefile. Make sure that the shapefile
has a proper projection defined. We will use this polygon to
reclassify the LiDAR points around the building. In this ex-
ample, I called the polygon “Building_Rooftop2D.shp”.
