Canopy Basemap from Lidar

I'm always looking for a way to create a better basemap for my cartographic work. One of my favorite tools is the bare Earth model created by stripping canopy data from LiDAR. Sometimes having tree canopy data is useful for cartographic purposes, so here's a quick tutorial on how I've extracted and symbolized that data for use in ArcMap.

Our goal is to end up with a map that looks like this. I have used a hillshade generated from LiDAR and have it set to 50% transparent and the canopy layer is underneath with a color ramp of leaf green for high, apple green for medium, and yucca yellow for low (none). I've added elevation contours, and a water layer created using this tutorial.

To get to this map, you'll need to have two elevation layers rendered from LiDAR. I'm working with data that can be found here. Feel free to download White_2015_16_zLAS.zip and work alongside me. Once you've got all of your zLAS files in a LAS Dataset, add it to the table of contents and zoom in to view an area. You'll also want to turn on the LAS Dataset from the toolbars menu.

Once the LAS Dataset toolbar is open, make sure the drop down menu Filters is set to All, and the drop down menu with colored dots is set to Elevation, as shown below.

It should render out an image that looks like this hot mess. The "static" looking area is what tree canopy looks like when it's made into a TIN (Triangulated irregular network).

Using the LAS Dataset to Raster tool we'll convert what we see above into a raster image to work with. Let's call it elev_w_canopy since it's a combination of elevation and canopy.

Now let's create a bare Earth model (BEM) where we use the LAS Dataset tools to strip away the canopy.

The resulting terrain should look smoother than the terrain with canopy. Let's use the LAS Dataset to Raster tool to make this into a raster also. Let's call it elev.

Once you've created elev, you can create all kinds of wonderful derivative products from it like contours, and hillshade. Since we're making a map in this tutorial go ahead an render out some contours at 100' intervals, and generate a hillshade for use later.

Now for the magic. Since we have a raster of elev and elev_w_canopy, and what we're after is literally the difference between them, we can use the Raster Calculator tool to execute that.

You'll end up with a raster image that looks something like this. Note that if you zoom in, there will be lots of gaps and the image itself appears noisy and grainy.

Let's deal with that noise by using Focal Statistics. Set the neighborhood to circle, the radius to 8 pixels, and the statistics type to maximum as shown below. Let's call the output canopy01.

You should end up with a raster that looks something like this.

Apply a thoughful color ramp to it, and stretch the histogram appropriately (equalize gets rid of the outliers nicely).

Add your hillshade above the canopy layer now and set your transparency to 50%. You should be looking at something that looks like this.

Bartoo Hall Gallery, Tennessee Tech University

It's always cool when my photos get used somewhere. Bartoo Hall on campus at Tennessee Tech is the latest place to take advantage of my regional photos. Below are lots of photos of the images they used with some dork posing beside them. ;)



Topographic Position Index (TPI) as an Alternative to Hypsometric Shading

Hypsometric Shading

Hypsometric Shading is a method that I have been using to generate beautiful basemaps well before I knew what it was called. Sometime during my undergraduate days I had the idea to view a hillshade at 50% transparency and lay it atop a color-ramp-ed elevation layer. The resulting output was already well known to cartographers, and appears to have been first done in 1503 by one of my very favorite historic figures, Leonardo da Vinci.

While I am no Leonardo, I offer these examples of my previous work as a means to better understand what hypsometric shading looks like.

Sparta, White County, Tennessee

Pickett State Park, Tennessee

Topographic Position Index (TPI)

While hypsometric shading is generally pleasing one may need to draw the users attention to other types of features. By using a combination of a hillshade at 50% and a Topographic Position Index (TPI) raster, one could draw attention to ridges and valleys. For example, I offer this map of an area within Scotts Gulf in White County.

Topographic Position Index (TPI) as an Alternative to Hypsometric Shading

The TPI layer draws attention to areas, which are on average lower or higher than their surroundings. I first learned of TPI as an analysis tool through Jenness Enterprises's Land Facet Corridor Designer extension for ArcGIS. As an alternative to that tool I applied some of my Photoshop image processing skills when I realized the nature of TPI. It's unsharp mask. In other words, a gaussian blur is applied to the original image, and then that is subtracted from the original image and a histogram stretch applied. With ArcMap, one can designate the histogram stretch method very specifically without altering the values of the raster, and one can apply a color ramp.

Without invoking the tool, one can use focal statistics in place of the gaussian blur, and use raster calculator to subtract the focal statistics layer. In other words, it would look like this:

All this is just to say, "Hey, this may look cool." Let me know if and how you use it. :)


Tennessee Mississippian Strata - Fossils - St Louis, Warsaw, and Salem Limestones

St Louis, Warsaw, and Salem Limestones

collection_no identified_name identified_rank accepted_name accepted_rank early_interval max_ma min_ma
64639 Naticopsis (Naticopsis ?) n. sp. buttsi species Naticopsis (Naticopsis) buttsi species Meramecian 345 336
Bellerophon cf. scissile species Bellerophon scissile species Meramecian 345 336
133416 Stethacanthus altonensis species Stethacanthus altonensis species Visean 346.7 330.9
40773 Meekopora cf. clausa species Meekopora genus Meramecian 345 336
162328 Beyrichiella confluens species Beyrichiella genus Osagean 353.8 342.8
Cytherella glandella species Cytherella genus Osagean 353.8 342.8
Glyptopleura costata species Glyptopleura genus Osagean 353.8 342.8
Paraparchites carbonaria species Paraparchites genus Osagean 353.8 342.8
Paraparchites nicklesi species Paraparchites nicklesi species Osagean 353.8 342.8
Savagella lindahli species Savagella genus Osagean 353.8 342.8
Data Provider The Paleobiology Database
Data Source The Paleobiology Database
Data License Creative Commons CC-BY
License URL http://creativecommons.org/licenses/by/4.0/
Documentation URL http://paleobiodb.org/data1.2/occs/list_doc.html
Data URL http://paleobiodb.org/data1.2/occs/list.csv?datainfo&rowcount&strat=St.%20Louis%20limestone
Data URL http://paleobiodb.org/data1.2/occs/list.csv?datainfo&rowcount&strat=warsaw%20limestone
Access Time Mon 2018-02-12 17:04:28 GMT
Title PBDB Data Service

Longwave Light on Cave Formations

Last year I purchased a handheld longwave light in the hopes that it would assist in non-destructive mineral identification in caves. The model I purchased was Convoy S2+ Nichia 365nm UV LED 1Mode OP Reflector Flashlight since it ran on 18650 rechargeable batteries and seemed of rugged construction. After having been caving with it for more than a year I've found it to be rugged enough for my typical caving trip, though I generally carry it in a pelican case with my camera. It is lightweight, and doesn't seem to drain the battery quickly, though I use it only for a handful of minutes on any given caving trip.

To no one's surprise calcite is the dominant mineral in the caves of Middle Tennessee that I frequent. Trace impurities within calcite which can color it dramatically in visible light do not significantly change the phosphorescent color spectra. Other minerals present in caves of Middle Tennessee include chert (which does not phosphoress), dolomite (which reflects purple under longwave), and gypsum (which with phosphoress green the same as calcite).

Example 1: Flowstone in Blowhole Cave

Painted light from my headlamp (white balance adjusted)

Flowstone, A Passage, Blowhole Cave, Cannon County, Tennessee

Actively painting the formation with the longwave light
Longwave light, Flowstone, A Passage, Blowhole Cave, Cannon County, Tennessee

Phosphorescence from mineral after turning off longwave light
Phosphorescent light, Flowstone, A Passage, Blowhole Cave, Cannon County, Tennessee

The top image being what one sees in the "natural" colored light of one's headlamp, or if we were standing outside holding it under sunlight. The following image is where the mineral is being actively excited by the longwave light. The final image is the afterglow (captured small bit by small bit in numerous photos stacked in Photoshop), which is the most diagnostic of the three images in terms of mineral identification. Note the location of red color in natural light is slightly visible in the other images.

Example 2: Dolomite vugs in the Bangor Limestone in Gourdneck Cave

Illuminated by an external camera flash
Dolomite vugs, Gourdneck Cave, Gourdneck Cave Preserve, Marion County, Tennessee

Actively painting vugs with the longwave light
Dolomite vugs in longwave light, Gourdneck Cave, Gourdneck Cave Preserve, Marion County, Tennessee

The lack of phosphorescence in this example is what I believe to be so striking.

Example 3. Gypsum crust in Cumberland Caverns

Painted light from my headlamp (white balance adjusted)
Gypsum crust, Cumberland Caverns, Warren County, Tennessee

Actively painting the formation with the longwave light
Gypsum crust, longwave light, Cumberland Caverns, Warren County, Tennessee

Composite of the previous two photographs
Gypsum crust, longwave blend, Cumberland Caverns, Warren County, Tennessee

This sequence shows that only a portion of the gypsum is phosphorescent, but I have no explanation why other than perhaps there is a patina on the surrounding gypsum blocking the longwave from exciting the mineral. This would suggest that older formations don't phosphoress where the younger ones do.

The research leaves me with more questions than answers.