Monday, February 20, 2017

Cartographic Fundamentals

Introduction

  The first objective of this lab is to incorporate the fundamental cartographic elements of maps using UAS data. The UAS data was collected in Eau Claire county Wisconsin. The sport field image was collected in March of 2016, and the Hadely cemetery image was collected in September of 2016. The second objective for this lab is use these maps to identify spatial patterns and explain about them is some detail. The fundamental cartographic elements include a north arrow, a scale bar, a locator map, a watermark (place to put the author's name), and the data sources. These important cartographic fundamentals are essential to use when working with UAS data as it helps the reader to make spatial relationships. Without out the cartographic elements the reader can not understand the whole story, and he or she can assume things which is bad. For example, people often have an aerial image and claim it to be a map. This is incorrect. If there are plotted points on an aerial image but nothing else, the reader doesn't know what the scale is, the orientation, or where the aerial image was taken. To make the aerial image into a map, a north arrow, scale, locator map, watermark, and the data sources must be on it. The data sources can consist of the UAS platform used to collect the data, the height at which it was flown, and the coordinate system which the map is projected in.
  Once these elements have been added, then the reader can identify spatial patterns within the data. Example spatial patterns that one could find using UAS data include wind direction based on the the slope faces of sand dunes, glacial extent based on terminal moraine and elevation profile, tree types grouped based on the amount of sunlight taken in, point bars along a meandering river and where they form or could be expected to form, and how a landscape changes after a flooding event.

Methods

What is the Difference Between a DSM and DEM, and a Georeferenced Mosaic and Orthorectified Mosaic?
 A Digital Surface Model (DSM) represents the first detected surface by the sensor. This surface can be many different things. It can be trees, houses, streetlights, cars, the ground, or any other feature. A Digital Elevation Model (DEM) is a representation of the earths ground topography. It only takes into account the earths ground surface.
  A georeferenced mosaic is a collection of raster images of the earth which is georeferenced, or assigned to, its correct location on the earth. It assigns it a coordinate system which is necessary to perform spatial analysis. An orthorectified mosaic is also a collection of raster images of the earth, but it assigns each pixel to a real location on the earth. It can assign values in Z dimension for trees, buildings, and other structures.

What are the DSM statistics? Why Use Them?
  For the Hadley cemetery, the minimum elevation was 283.35 m, the maximum was 310.46 m, and the mean was 289.35 m. The standard deviation was 4.45 m. The pixel depth was 32 bit, and the linear unit was one meter. Its spatial reference was WGS_1984_UTM_Zone_15N.
  For the Eau Claire sportfield, the minimum elevation was 17.95 m, the maximum was 22.61 m, and the mean was 20.21 m. The standard deviation was 1.05 m. Its spatial reference was also WGS_1984_UTM_Zone_15N. Also, its linear unit was one meter. The pixel depth was 32 bit.
  These are important descriptive statistics because without them the data doesn't have much meaning. Knowing the linear unit, and basic statistics of elevation help so that some analysis can be done. For example, one can assume that about 68 percent of all the elevation values will fall between 284.90 m and 293.78 m.

 How is a Hillshade Done? What are the Different Regions of the DSMs?
  The Hillshade is performed in ArcScene using the Hillshade (spatial analyst) tool. To run the Hillshade on the DSM, The DSM was put in the Input Raster box, and then a name was given with its file path in the Output Raster box. This then can be opened in ArcMap where cartographic elements can be added.
  In the Hadley cemetery DSM there are three main regions: the southeast, the southwest, and the north. There are many trees located in the southeast region, while corn stalks populate the southwest region, and prairie grass is in the north region. In the Eau Claire sport field DSM there are five regions. The northwest, the southwest, the center third, and the two areas which lay just outside the sport track. The two areas just outside the sport track are lined with trees, and the other regions don't have any significant vegetation.

Results

What Types of Patters Can Be Seen in the Orthomosaics, and the DSMs? Describe the Regions.
How do These Patterns Align with the Descriptive Statistics?
  Below, in figures 2.0 and 2.1, are maps which show the orthomosaics overlaid with the hillshaded DSMs. Also, an oblique view is given for each map. The Hadley cemetery map, displayed in figure 2.1, has the DSM overlaid with the orthomosaic. The oblique view also shows this, but with a 3D angle. The Eau Claire sport track map, shown in figure 2.1, has the hillshade overlaid with the orthomosaic, while the DSM is displayed in the oblique view.
Hadley Cemetery Map
Fig 2.0: Oblique View, and DSM overlaid with an Orthomosaic of the Hadley Cemetery
  In the Hadley cemetery map, the highest elevation occurs in the southeast corner, where there are many trees. The lowest elevation is in the north region where there is prairie grass. In the southwest corner, where the corn stalks are, is where the in between elevation occurs. This can really be seen in the oblique map as the trees stand out as really tall, and the grass north of the road is really low. The orthomosaic was set to 40 percent transparency so the DSM can be seen. The tops of the trees look very white, but other than that its difficult to pick out any other of the DSM hues. Overlaying the DSM and the orthomosaic helps for the eye to see where the highest points are, but not the lower points. The patterns on the DSM and orthomosaic align perfectly, which is why they were overlaid with each other. Comparing the image to the descriptive statistics, it is clear where the maximum is, but it is unclear where the minimum elevation is exactly. Knowing that linear units is meters is nice, but it doesn't really mean anything in this map because the images are overlaid with each other there for not allowing for all the gray hues to show through. However, in the oblique map, the elevation can clearly be differentiated across all three of the main regions.

Eau Claire Sport Track Map
Fig 2.1: Oblique View, and Hillshade overlaid with an Orthomosaic of the Eau Claire Sport Track
  In the Eau Claire sport track map, the highest elevation occurs in the northeast corner while the lowest elevation occurs in the southwest corner. This can be clearly seen in the DSM. The color scheme makes it look like a lot of elevation is lost from one end of the map to the other, but in reality the range of elevation is a partly 4.7 m. The trees can be seen both in the hillshade/orthomosaic and the DSM maps. These trees are located in the two regions located just east and west of the sport track. By looking at the maps and comparing them to the statistics, it looks like the mean (20.21 m) falls nearly in the middle of the sport track, the highest point (22.61 m) is on the northeast side of the sport track, and the lowest point (17.95 m) is located to the southwest of the sport track. The spatial reference of WGS_1984_UTM_Zone_15N makes for low distortion, so measuring distances along the map should prove to be accurate. The patterns found in the orthomosaic and the DSM are identical once again, and could be overlaid, but are separated to show what the DSM looks like on its own.

What Anomilies or Errors are Noted in the Data Sets?
  A big error which is confusing is the elevation values in the Eau Claire Sport Track map. For some reason the values are only between 17.9 m and 22.6 m. That low of an elevation is not accurate for the Eau Claire area. Another error is the height in the oblique maps. The trees in the sport track map don't seem tall enough. The vertical exaggeration was calculated by extent, but knowing how this relates to the actual height remains unknown.

Where is the Data Quality the Best? Where do you Note Poor Data Quality?
How Might This Relate to the Application?
  The data quality is the best in both of the overlaid maps. The resolution is very high in both of these maps. This is the case because the images were taken directly from the Phantom 30 UAS platform and no manipulation of these maps were needed. 
  The data quality is the poorest in the oblique maps. Although these maps make for a nice visual, they wouldn't be that useful for conducting detailed analysis with. Both of the obliques had to be created in ArcScene where the color schemes where changed, along with the vertical exaggeration. These maps have poor quality because they had to be converted to raster when exported into ArcMap. This makes the resolution worse as the pixels get distorted easily. Unfortunately, there is no way to avoid this because cartographic elements cannot be added in ArcScene.

Conclusion

What Makes UAS Data Useful as a Tool to the Cartographer and GIS User?
  To summarize, UAS data can be a usefull to both cartographers and GIS users. The elevation data such as the kind used in this lab, can be used to generate maps. For the GIS user, there are different tools such as the Hillshade which can be used to help aid with the direct UAS imagery. The imagery can be used in ArcScene to create a nice 3D visual of the study area, but it is isn't that useful for much more than that. UAS data can be used for very accurate precision of the earths surface, and can be use things which need elevation such as the direction water flows, or how much sunlight a certain spot receives.

What Limitation Does the Data Have? What Should the User Know About the Data When Working With it?
  Some limitations of UAS data is that it isn't very pervasive. It only focuses in on a small area. Another downfall of the UAS data is that there aren't attributes associated with the data, because it's raster data. Users should know that the imagery is raster data, and should be careful when mapping it so distortion of the resolution is minimal. Users of UAS data should also know that scale is a very important cartographic part of the map. Knowing the scale will allow the user to create spatial connections between areas much easier. Lastly, the user should know that just taking an image from the aerial imagery and claiming it to be a map is incorrect. All maps need the fundamental cartographic elements described in the introduction.

What Other Forms of Data Can UAS Data be Combined With to Make it Even More Useful?
  Other forms of data which can be combined with UAS data included precipitation data, land cover data, soil type data, and geologic data. Precipitation data can be coupled with UAS data to see what the earths surface looks with specific precipitation value. Wetter places should have a greener surface, while dryer places should have a browner surface. Land-cover data can also be coupled with UAS data. This helps to see what type of vegetation is on the surface while also seeing the actual vegetation in the UAS data. Soil type data, and geologic data can be used in the same way to see what the land looks like above the surface where a specific soil type or geologic unit resides. In Conclusion, UAS data can be used in many useful ways. It's important to make sure that the user actually produces a map of the data and not just an image when presenting the data.

Sources

Posted by at

No comments:

Post a Comment

Subscribe to: Post Comments (Atom)