Archive for September, 2013
Friday, September 27th, 2013
In the world of remote sensing temporal changes are much more relevant in some data than others. Imagery and infrared can change in real time while terrain and bathymetric data tends to change more slowly. So, as a terrain provider, it is an advantage to not have requirements for continuous updates.
Typically a terrain data set collected in the last 10 years is considered new data, while an air photo from that vintage would be considered historic archive with very little value. Of course humans can change terrain quickly but with the exception of a few natural catastrophes, geomorphological changes are so languid that we come to expect terrain to stay the same. So when this new island off the cost of Pakistan popped up (yes pun intended) it had me rethinking the pace that terrain can change.
Tuesday, September 24th, 2013
As a map geek that has lived through the Google Earth revolution, it amazes me how far we have come in such a short time. The pre Keyhole (2001) world is hard for me to even imagine these days, and not just because I am a map guy. Think about what it took back then to look at an air photo of a place you were planning to visit or look at the terrain and profile of a ride you were thinking about doing; unless you had some great connections it was impossible. Back then, If a hotel advertised a beach front location near downtown you had to take their word, now all we do is enter the address and voila, you can not only see where the hotel is but you can see what it looks like from street view, simply amazing.
Every time I show my children some new place on Google Earth I reflect on how they will never know the pre-Google Earth world. Back then, map data providers could be called cartographers, seeing an air photo of your house was a big deal, and the world just seemed much bigger.
Monday, September 23rd, 2013
The bread and butter of terrain data and DEMs have always been hydrologic modeling. The reason is simple, water follows the rules, it goes downhill and terrain will determine where the water will go, or so I thought…..
As I write this, I am in the middle of an actual flood. Boulder, Colorado is experiencing a 100 year flood event and it has been an eye opening experience to see the difference between a model and the real world.
Our house sits on top of a ridge about 80 feet above flood level and 1200 feet away from the nearest drainage channels. If you evaluated a typical fluvial flood model, our house would be shown as well above flood levels of even a 500 year event. The problem is that the typical fluvial model only works when you focus on the drainage channel as the source of flood. In a pluvial event, where rain comes down faster than the drainage infrastructures can handle, the use of terrain data alone as an input for predicting the water movement is not enough. Not only must you know drainage infrastructure capacity but you need understand geology and the permeability of the soil, as well as the actual soil saturation.
Friday, September 20th, 2013
Map geeks are a unique breed that tend to be equally reliant on both the left and right sides of our brain. Call it indecision to form a self identity or call it a balanced intellect, but we fall somewhere between a pocket protecting engi-nerd and a black turtleneck, beret wearing artist.
My own college background is a perfect illustration. I entered college as a fine arts major, realized I would never move out of my parents house if my income was reliant on my artwork and stumbled across cartography as an art form that had actual job titles associated with it. This mix of opposing characteristics makes for a group of people that are usually smart and funny. Yeah I know the dot coms of the world already exploited the nerds just want to have fun vibe and you will be hard pressed to find many geospatial companies that go as far to push creativity as the Googles of the world, but hey, as a ratio of fun to dollars made, I would say mappers are the top of the heap. I mean come on we have a foosball table and our marketing departments encourage videos like these:
Friday, September 13th, 2013
As a medium resolution data set the NEXTMap® 5m terrain data plays an important role between high precision ground survey or LiDAR and lower resolution ‘free’ data. This sweet spot is the realm of preliminary planning or site location analysis when a project needs to be defined but not yet calculated. The use of the NEXTMap data during this initial phase may have higher upfront costs, but in the long run, the more precise a preliminary plan is the less re-work you will experience down the line. Site planning is just one of the many applications that the 5 meter data is suited for, and I will jump into others in following posts.
Tuesday, September 10th, 2013
If you have worked in GIS then you have probably fused raster elevation data. So, chances are that you have also torn your hair out if these data came from different sources and had accuracy biases or projection difference. Trying to get a seamless merging of two separate elevation sets is never as easy as it initially sounds. The problems can be as complex as the misalignment of the origin point in reprojection or a reference to pixel center vs. pixel corner. Or, they can be as simple as a vertical offset or a planar tilt.
Intermap® has worked very hard to simplify the data fusion process. Over the course of a couple years we have been developing a data fusion tool and a universal projector tool. With these tools now complete, we are able to reproject to hundreds of vertical and horizontal datums with validated precision. And, with our automated fusion tool the user loads the data and species the reference data then the tool does the rest. Overlapping areas area removed, vertical bias applied, data is tilted to minimize all edges around the fusion boundary and the remaining seams are blended to complete the process.
Friday, September 6th, 2013
Terrain is tricky to understand just by looking at a map. I just spent the weekend riding the Monarch Crest Trail near Salida, CO. The ride is a ‘shuttle’ style ride, meaning you get driven to the top of the trail and end down at the bottom. In theory, shuttle rides are much easier because the majority of climbing is eliminated, but the crest trail is a bit of an anomaly. My friends and I gave a quick scan of the map before we started and noted that we would be dropping 4000’ from start to end. The profile looked promising and we set off with visions of a downhill run with just a bit of climbing mixed in.
Wednesday, September 4th, 2013
Ski racers demand super stiff boots in order to push through fast turns. Unfortunately, misinformed ski boot shoppers believe that because fast racers use stiff boot those same boots will help them achieve higher speed. In fact the opposite is true, the stiff boots put the amateur skier in an inappropriate position for their abilities and they are unable to exhibit the force into the boots required to make them work. Instead of going faster, our poor skier ends up off balance and out of control with a significantly lighter wallet.
Consumer desire for increased DEM resolution is akin to buying ski boots. I often hear “I need the highest resolution data available for this area.” Sometimes this statement proves to be a true requirement but often when asked what the DEM data will be used for the actual DEM requirements for the application don’t warrant ‘high’ res terrain data. So the consumer ends up with a high cost for storage and processing of a heavy dataset that is overkill for their project. Just like the weekend warrior on race boots, using a data resolution higher than you need will slow you down and cost you money.