Emily McRae's GIS Blog

This module focuses on methods of classifying raster images, including supervised and unsupervised classifications. The above map shows an example of supervised image classification completed in ERDAS IMAGE and mapped in ArcGIS Pro. By creating features in an image to feed an algorithm, an image was classified based on several polygons created manually. These images were matched across the entire image to create classes. Spectral bands were then compared through histograms to identify the bands which need to be separated due to overlap in like features. The image was then displayed based on these classes, as well as acres belonging to that class.

This lab follows methods of spatial enhancement using various techniques in both ERDAS IMAGE and ArcGIS Pro. The process summary for the different sections can be seen below, as well as some finished map products that were produced using these methods.  Exercise 1: In this exercise, using the glovis.usgs.gov site, I explored the different freely available satellite data from various satellites and how to view the metadata. I did make an account with the website to be able to download data, however I did not receive the confirmation email yet so I am using the data that was provided already in the UWF drive and did not download any data from the website. The image file in the folder was in a .tar.gz format. To convert it into a .tif file, the process involved unzipping the file into a .tar folder, and then once again unzipping that folder into a folder which contains 7 tif files and then supporting documents. It seems this process progressively sheds layers of the original file into

This lab provided an introduction to using the ERDAS software for processing and deriving information about image and digital satellite data. Additionally, it provided an introduction to calculating and understanding the electromagnetic radiation as it related to satellite imagery.  There were two parts to this lab. See process summary below for more information about the activities performed in the lab, and struggles encountered along the way. Part A This lab largely focuses on practicing simple movements in the ERDAS program. To begin, I opened the ERDAS IMAGE 2020 32-bit program and added the files using the folder icon. To open files, the process involves selecting the file type, in this case raster, and then setting the file properties with opening them, including selections for the display, for multiple file types, and for framing such as fit to frame, and clear display. I explored the pan and zoom options, including zooming in and out, panning using the mouse, and using the

In this lab, a study area in Pascagoula, MS is classified up to the 1 st  and 2 nd  classifications under the USGS Standard Land Use / Land Cover Classification System, at a minimum mapping unit of 1:2000 inches, using an aerial photograph. The process involves standardizing best judgement in a way that can be replicable and reliable. Once all land parcels are drawn and classified, 30 random points are placed across the area and ground truthing is performed using Google Maps Street View. The percent accuracy is calculated based on the number of correctly identified points divided by the total number of sample points, multiplies by 100. The ground truth results yielded a 73% accuracy result.  The map below shows the results of the classification and ground truthing.  The following classifications were identified based on their characteristics: 11- Residential: Areas with regular patterns of small homes, green or grassy yards, and following small streets. 12- Commercial and Services: Are

This map demonstrates, through a grey scale aerial image, how one can classify features on a landscape visually using tone and texture. Tone being the scale of lightness to darkness and texture being a scale of very coarse to very fine. These classifications were done by eye, and may be subjective to the analysts interpretation, as well as the images clarity, scale, and resolution. This map identifies features in a greyscale aerial photo by 4 means: pattern (Objects/places were chosen that are clearly visible and interpretable as they are part of a larger repeating pattern, but that had they been presented in isolation or at a very large scale, would not be clear as to what they are), shadows (Objects were chosen which could not have been identified had they not been casting a shadow which revealed information about what the object is), association (Objects were chosen that could not be identified without looking at the surrounding environment and making logical connections about what

This lab explores the complexities in representing real geographic phenomena accurately, both in terms of physical features and socioeconomic or other patterns, based on decision made by the cartographer. It looks at the distortions that can occur when data are displayed at varying scales, resolutions, and the very interesting and perplexing issue of the modified area unit problem (MAUP). The MAUP is an of yet, unsolved issue in spatial analysis, by which aggregation at different boundaries can yield very different results for analysis. This can even be exploited with gerrymandering, a practice of drawing political boundaries to intentionally to group voters and representation in favor of a political party.  This lab looked at the affects of scale on vector data, resolution on raster data, and measuring the degree of gerrymandering in congressional districts within the contiguous US. I will briefly detail the analysis process for each, and provide examples and explanations of the findi

This weeks module focused on surface interpolation using the Spline, Thiessen, and Inverse Distance Weighted (IDW) methods.  In the example below, water quality sample points taken across Tampa Bay were used as input data and run through the 3 surface interpolation methods to get an understand of water quality across the entire Tampa Bay study area. The methods are compared based on their ability to achieve a good impression of what water quality may actually look like across the surface.  I will detail how each method works,  their strengths and limitations as it applies to this example, and the result of this method when applied to Tampa Bay water quality assessment.  I will note that none of these methods take into account other highly influential factors of the water landscape including flow direction, elevation, watershed, aquifer boundaries, etc.  Spline Conceptually, how it works: The spline method interpolated the value of the surface where data points are not recorded, by pas

Triangular irregular networks (TIN) and digital elevation models (DEM) are explored and compared as models for representing terrain. TINs, as indicated in the name, are models composed of nodes and connecting vertices (Bolstad, 2016, Chpts 2 and 11). DEMs are generally composed of pixels. They both have data on terrain and ground elevation and can be used to explore similar topics.  In this module, I explored TINs and DEMs in five different examples. I will focus my discussion here on one example, where I believe the difference between the two data types is best illustrated. In this example, a TIN is compared to a DEM derived from input shapefiles, and both are used to create contour lines across a terrain.  A point and study area feature layer are added to a local scene, and used as inputs to create a TIN of the elevation. A DEM is then created using the Spline tool, which uses input data to create a smooth surface passing through the X,Y, and Z values of the input points. This tool c

In this module, a completeness assessment is conducted on a Street Centerline data set provided by the Jackson County, OR.  The completeness of a data set refers to the extend that the data set contains the features it represents. Unlike positional accuracy standards, which have established methodology and accuracy thresholds from national and international standardization organizations, measuring the completeness of linear features like roads is not standardized (Bolstad, 2016, Chpt 14). This means that it can be left up to the party assessing the data to create their own methodology.  In assessing this data, a greater understanding of the data quality can be achieved and can also be included in data quality reporting. Users can determine if this data set is adequate for their purposes and choose to use it or seek another source.  Methodology: This assessment looks at the completeness of road networks based on length of road segments within Jackson County, OR as compared to a road net

In this module, formally established methods of determining accuracy standards are reviewed, notably the National Standard for Spatial Data Accuracy (NSSDA). This is a national standard for determining the confidence of point positional accuracy. This method uses a sample of real and test points selected at random to generalize the accuracy of the entire data set. As an example of this practice, I tested the accuracy of two road maps of Albuquerque streets, one provided by the city of Albuquerque, and one a USA street map. Those these data layers are lines and not points, points are placed at intersections and used to test the accuracy of the road segments. Other methods for determining positional accuracy of line segments, such as the epsilon band method, are not formally standardized ( Bolstad, 2016, Chpt 14) . Using NSSDA guidelines and procedures, confidence of both data sets is determined and reported to the nearest foot. Methods:  1) Divide study area: Study are is divided into

This weeks module uses an example of GPS points and a reference point to demonstrate methods of determining accuracy and precision in GIS.  The above map uses a set of given GPS point locations to determine the average location of all points, and then the difference from actual points to this average. The 68th percentile is determined, and the distance away from the average location at the 68th percentile is used to create a buffer.  Calculating Precision Precision refers to the closeness in value a set of values, in this case geographic coordinates, are to each other. It does not indicate anything about the closeness to the actual value, or the actual location of the point in this case.  In this example, a GPS is used to take 50 sample measurements of a single location. The precision of these values can be measured using percentiles. It is generally accepted that those values in the 68th percentile, or those values which make up 68% of all values when ordered numerically from smallest

 This module was on conducting damage assessment on coastal flooding using ArcGIS Pro analysis tools. A case study of Hurricane Sandy is used. Using imagery and various analytic tools, a general assessment of structure damage caused by flooding and wind is conducted for an area in New Jersey.  Structures by structural damage level after Hurricane Sandy. Process summary: 1) 2 mosaic datasets are created, one for the study are before the storm and one for after, using aerial rasters. The mosaics are created within the gdb being used. The result are two projected aerial images with grids highlighting the study area. When turning on and off these two layers, damage such as flooding and missing buildings is apparent. 2) Within the gdb, new domains are created that will be used to categorize damage by type and severity. This is done by adding Domains under the gdb, and the categories for Inundation, Structure Damage, Wind Damage, and Structure Type are added alon with options for each domain

In this module, various methods of assessing coastal flooding are used to determine flood damage. In this first example, the affects of Hurricane Sandy on a part of the New Jersey coastline is assessed using DEMs. A DEM of the area before the storm and a DEM of the area after the storm are used, along with satellite data from before the storm, the assess which areas experienced erosion and deposition, likely from debris. The study are DEM of after the storm is subtracted from the pre-storm DEM to generate a change layer, which can be used to assess areas based on their surface elevation change. This method, combined with visual analysis and comparison of previous buildings and surface features, can be used to determine where buildings have been damaged or washed away. In this second example, storm surge of 1 meter is analysed as it is likely to affect building locations in Collier County, Fl. Two data sources are used, a Lidar DEM and a photogrammetry DEM from the USGS. The L

Overview:  This module's lab involved using three different hotspot techniques in ArcGIS Pro to determine crime densities within areas in Washington DC and Chicago.  See figures below for a brief overview of the process of creating each hotspot type. Local Moran's I Hotspot map is shown above. Moran's I process summary:  1)    Join the boundaries (census tracts) to the points (number of homicides in 2017): Right click on the polygon layer, in this case census tracts, select Spatial Join. The target feature will automatically be set to the polygon layer, set join features to the points, in this case number of homicides in Chicago in 2017. Set operation to ‘one to one’, which avoids duplicating point features, ‘keep all target features’ box is left checked, and set the match option is set to ‘completely contains’, which will produce a column within the census tracts layer that holds a count of all point features that fell within that polygon. 2)       Calculate th