Michaela's GIS Journey

Spatial Enhancement and Multispectral Analysis This week's lab assignment included several exercises to further familiarize ourselves with ERDAS Imagine and to switch between programs since both ERDAS and ArcGIS Pro are best used in conjunction. Our primary focus in the lab centered around the Olympic Mountains in Washington State. The Process: A Detailed Look at Exercise 7 The culmination of our project was Exercise 7, where we synthesized the skills learned throughout our remote sensing course. We followed a meticulous four-step process to identify and map distinct environmental features: Histogram Analysis: We began by examining the histograms for each satellite data layer, looking for spikes that denote predominant pixel values. These spikes helped us pinpoint areas of interest by showing where features occur frequently and have consistent reflectance properties. Grayscale and Multispectral Viewing: Next, we analyzed the image in grayscale to identify contrast differences mor

Introduction to ERDAS Imagine  This map illustrates a classified image of forested lands in Washington State, beginning with data preparation in ERDAS Imagine. Although ERDAS Imagine has map-making capabilities, the subset data was transferred to ArcGIS Pro to create the final map product. In ERDAS Imagine, a new field was added to the attribute table, and a specific area of the image was selected for export. This preprocessing step enables smooth integration into ArcGIS Pro, where the final map, showcasing detailed land cover types, is produced. This workflow demonstrates the combined power of ERDAS Imagine for data classification and ArcGIS Pro for map creation, especially when handling complex spatial data.

 LULC Classification and Accuracy Assessment for Pascagoula, MS Map Summary: This project involved creating a Land Use and Land Cover (LULC) map of Pascagoula, MS, using aerial imagery to identify and classify various land features based on size, shape, color, and other visual elements. Using the USGS Standard Classification System at Level II, I achieved an overall accuracy of 83.33%. Commercial areas had the highest accuracy, easily recognized by their distinct structures and parking lots, while Residential areas presented challenges, primarily due to generalization. This map highlights the nuances of classifying urban and natural landscapes within a diverse region. 

  Understanding the Scale Effects on Vector Data, Basic Resolution Effects on Raster Data, and Gerrymandering Scale Effects on Vector Data : Scale significantly influences how vector data is portrayed. Smaller-scale maps (such as 1:100,000) offer less detail, resulting in the generalization or exclusion of more minor features. Conversely, larger-scale maps (like 1:1,200) provide greater precision, capturing intricate details such as complex boundaries. As the scale reduces, polygons become simplified, and more complex boundaries—like those of irregular coastlines—are smoothed out. This has a direct effect on the accuracy of calculated areas and perimeters. Regarding hydrographic features, moving from a 1:1,200 scale to a 1:100,000 scale demonstrated a notable reduction in detail. Basic Resolution Effects on Raster Data : Raster data divides the world into a grid where each cell holds a single value. The resolution of raster data refers to the dimensions of these cells. A higher resolu

 Interpolating Water Quality Data for Tampa Bay In this exercise, we applied various interpolation techniques to explore water quality conditions in Tampa Bay, Florida, using a dataset of samples collected over a short period. The primary focus was on Biochemical Oxygen Demand (BOD) concentrations, measured in milligrams per liter, a key indicator of water quality. By using methods like Thiessen , IDW , and Spline interpolation, I was able to generate surfaces that estimate BOD levels across Tampa Bay, providing insight into the spatial distribution of pollution. Each method offered a different approach—some emphasizing sharp transitions between points and others providing smoother, more continuous surfaces that better represent gradual changes in water quality conditions. Performing Interpolation Analysis  Thiessen interpolation   assigns each location the value of the nearest point, resulting in abrupt boundaries that  may not be suitable for continuous phenomena like pollution. Inv

In my exploration of elevation data models, I examined both Triangulated Irregular Networks (TINs) and Digital Elevation Models (DEMs) to understand their unique properties and applications. TINs are formed by connecting data points into triangles, providing highly detailed but angular representations of terrain. On the other hand, DEMs are grid-based and create smoother surface representations through interpolation techniques. I compared these models by analyzing contour lines generated from both TIN and DEM data. One key difference I found is that DEM contours are much smoother, especially in flatter areas with less elevation variation, while TIN contours are more angular, especially in areas of steeper terrain due to the triangular structure. The smoother DEM is more suitable for continuous surface representation, while the TIN excels in areas requiring more detailed elevation information. A comparison between DEM and TIN contours can be seen below: DEM TIN This exploration highligh