GIS 5935 - Data Quality Standards

Continuing on from the first week of GIS5935, this week we were introduced to the more common data quality standards used for cartographic and digital geospatial data. These standards were put in place by organizations, such as the United States Geological Survey (USGS) and the Federal Geographic Data Committee (FGDC), to "provide a common language for reporting accuracy to facilitate the identification of spatial data for geographic applications" (FGDC, 1998). In particular, the lab allowed us to perform our own positional accuracy tests using the National Standard for Spatial Data Accuracy (NSSDA) standards. 

The NSSDA standard follows a seven step process which includes the following:

    1. Determining if the test should be for horizontal accuracy, vertical accuracy or both.

    2. Selecting test points from the data set being evaluated

    3. Selecting an independent data set of higher accuracy that corresponds to the data set being tested.

    4. Collecting measurements from identical points from each of those two sources.

    5. Calculating a positional accuracy statistic using either the horizontal or vertical accuracy statistic worksheet.

    6. Preparing an accuracy statement in a standardized report form.

    7. Including the report in a comprehensive description of the data set called metadata.

For this assignment we were testing for horizontal accuracy. The points I chose for step 2 can be seen below. 


Once the test points were created, along with the corresponding reference points, the coordinate information was exported and placed into an Excel spreadsheet to calculate the horizontal accuracy. This is done by determining the difference in error between the test and reference points, followed by determining the sum of the squared errors, average, Root Mean Squared Error (RMSE) and NSSDA, which is accurate to the 95th percentile. This process was then duplicated for a second dataset.

The following are my accuracy statements for both tested datasets:

ABQ_Streets_Sample.shp 
Positional Accuracy: Tested 15.81 feet horizontal accuracy at 95% confidence level. 

StreetMapUSA_Sample.shp
Positional Accuracy: Tested 383.58 feet horizontal accuracy at 95% confidence level.

Based on the accuracy statements it is clear that the ABQ_Streets_Sample shapefile has the better accuracy. The report in the metadata was not required for this lab. 

Comments

Post a Comment

Popular posts from this blog

GIS 5935 - Data Quality Assessment

Image
Like the second assignment, this module continues to focus on data quality. More specifically, this week focused on data quality of road networks. As such, we were tasked with performing a comparative assessment to determine the completeness of a street centerline shapefile and a Topologically Integrated Geographic Encoding and Referencing (TIGER) road shapefile for Jackson County, Oregon.  To perform this assessment we utilized a non standard method, based on that used by M. Haklay in the "Comparative Study of OpenStreetMap and Ordinance Survey datasets" in 2010. This included using a grid, which encompasses the study area, to clip the road network into smaller sections. These smaller sections give more detailed insight into areas which may be considered complete or incomplete. Once the roads were clipped to each grid cell/section the lengths were calculated and then compared.  My analysis determined that 134 of the grid cells/sections contained parts of the street centerlin
Read more

GIS 5935 - Surfaces - TINs and DEMs

Image
Having moved on from the modules regarding data quality, this week we were introduced to surfaces and surface models, more specifically Triangulated Irregular Networks (TINs) and Demographic Elevation Models (DEMs). TINs are based on vector data and are made up of a series of linked triangles that vary in shape and size, but display a 3D representation of a surface. DEMs are based on raster data, more specifically the spot heights associated with each grid part of a raster.  For this lab we were tasked with creating a TIN and DEM, as well as displaying contours based on both of them. We were then tasked with comparing the contours produced by both. Below I have included images of my results.  (a) Contour lines based on the TIN (b) Contour lines based on the DEM The contours produced as a result of the TIN appear to be more jagged, while the contours produced from the DEM have a smoother appearance. There is also a noticeable difference between the derived contours in areas that hav
Read more

GIS 5100 - Who is Gabrielle Hudson?

Image
Hi! My name is Gabrielle. I am currently a part time student in the MSc. Geographic Information Science Administration program at the University of West Florida (UWF), but I'm also a full time employee at the Bahamas National Geographic Information Systems (BNGIS) Centre, in The Bahamas. My official position is GIS Analyst and GI Education Coordinator, however due to the small staff compliment we all wear many hats. Some of my responsibilities include maritime delimitation, disaster preparedness and response, training, mentoring, data collection and data management. While I enjoy most aspects of GIS, my favorite part is producing a meaningful and esthetically pleasing map. I decided to pursue a masters level degree with the hope of increasing my knowledge in GIS, which will better prepare me for future opportunities at my current place of employment, or where ever the opportunity may present itself. The degree option at UWF seemed to perfectly match my goal of future advancement as
Read more