Areas at Risk for Volcanic Activity in Washington State

Introduction

The question that I created and attempted to answer was, "What areas in Washington state are at risk for volcanic activity?" More specifically, I wanted to show possible associated earthquakes due to volcanic activity; and which counties and major roads were at risk. This research question was designed for the population of Washington state, as well as those officials who provide risk assessments for volcanic and earthquake activity. 

Data Sources

I used different data from the USA geodatabase to help answer my question. This data was provided through the GIS I course offered through the University of Wisconsin- Eau Claire. There were a few data concerns, specifically, the topographic base map used did not match up perfectly to the data used from the USA geodatabase. Also, earthquakes within a 40 mile radius of a volcano may not have been generated due to volcanic activity. Also, population density affected by a volcano may extend further than 20 miles. 

Methods

My process began by creating a personal geodatabase to work from, labeled Final Project. From here, I worked through ArcMap and ArcCatalog to add the appropriate data that I intended to use. The different data layers I used from the USA geodatabase were: States, Counties, Volcanoes, Quakehis, and Maj_roads. By using different queries (Query and Spatial queries) and analysis tools (Clip, Buffer, Intersect), I was able to generate a map showing different areas that are at risk for volcanic activity. The diagram below is the data flow model that I used to answer my research question.

Results

The results of my project shows population densities (one dot = 20 people) within a 20 mile radius of a volcano, earthquakes within a 40 mile radius of a volcano, major roads, and major roads within a 40 mile radius of a volcano. Counties that may be affected by a volcano include, Yakima, Whatcom, Snohomish, Skamania, Skagit, Pierce, Lewis, Klickitat, King, Cowlitz, Clark, and Chelan. West Crater, Rainier, and St. Helens appear to be the three volcanoes that pose the greatest risk to population density, and major roads.

Evaluation

My overall impression of this project was a positive one. I felt very satisfied with my overall map and was impressed that I was able to answer the question that I posed. If I were to do this project again I would have searched for a geodatabase specifically for the state of Washington, so as to have more accessible information, such as damage, pollution, eruption extent, etc. Some challenges that I faced while doing this project included finding data that would help answer my question. Again, my results could have been much more accurate or helpful if such data as damages or eruption extent were included. 

Vector Analysis With ArcGIS

Goals and Objectives

The goals and objectives of this lab were to learn and apply various geoprocessing tools for vector analysis in ArcGIS to determine suitable habitat for bears in the study area of Marquette County, Michigan. I used a GPS MS Excel file to map the locations of the bears, and used other data to determine forest types in which the bears were found, as well as bears found near streams. I combined this information to find an ideal habitat for the bears. I then used DNR management location data to create a new bear habitat area that fit the above criteria and was also in the DNR management area. I finally excluded all areas that were within 5 kilometers from Urban or Built up lands to produce a final ideal bear habitat that is within the DNR management area.

Method

The first objective was to add the bear locations from the MS Excel file as an XY event theme. An"event theme" is a temporary display of X,Y data in ArcMap. Once they were mapped, I exported them to bring them into my geodatabase as a feature class.

The second objective was to determine the forest types where bears are found in central Marquette County, Michigan based on GPS locations of bears. I began by adding several feature classes to my map from the Marquette_bear_study geodatabase. These feature classes were landcover, streams, and study area. I used the intersect tool between the bear locations and landcover classes and created a new feature class named Bear_cover. I summarized the Minor Type field in the new table that was created to determine the sum of bears within each forest type. The top 3 habitat types were Mixed Forest Land, Forested Wetlands, and Evergreen Forest Land.

The third objective was to determine if bears were found near streams. I began by creating a 500 meter buffer around all the streams (remembering to use the dissolve tool) and then intersected the new feature class (Bears_within_500m_of_stream) with the Bear_cover feature class. 72% of the bears were found near streams.

The fourth objective was to find suitable bear habitat based on the two above criteria: within 500 meters of a stream and within the suitable forest cover. I began by querying the landcover feature class to select and create a feature class (Suitable_landcover) with only the minor type covers that the bears were found in. There were 6 total types. I then performed an intersect between the Bears_within_500m_of_stream and suitable_landcover features classes to obtain a near feature class (Ideal_habitat). I then used the dissolve tool to remove any internal boundaries, and make the polygons look much cleaner.

The fifth objective was to make recommendations to the Michigan DNR for a bear management plan. The DNR must find suitable bear habitat that is located on their management lands. I began by adding the dnr_mgmt feature class from the marquette_bear_study geodatabase, and then used the clip tool to include only DNR management areas within the study area. I then intersected the DNR_management and Ideal_habitat feature classes to obtain DNR management areas within ideal bear habitats (DNR_management_area_within_ideal_habitat). 

The sixth objective was to eliminate areas near urban or built up lands. I began by querying landcover to select the major type of cover "Urban or built up lands" and created a feature class (Urban_or_builtup_land). I then created a 5 kilometer buffer around this land (5km_buffer), and used the erase tool to remove the buffer and leave only the DNR management areas within ideal bear habitats at least 5 kilometers away from any urban or built up land.

The seventh objective was to create a cartographically pleasing map to display the different habitats and bear locations.

The eighth objective was to create a data flow model of the procedures used to determine suitable bear habitat in Marquette County, Mi.

Results

The area in light green is the ideal bear habitats based off of the two criteria; within 500 meters of a stream and within the suitable land cover types. The areas in pink are the DNR management areas within the ideal bear habitats. The brown circles are all the bear locations that were found through GPS. The upper right hand corner shows the state of Michigan as well as Marquette county.

Above is the data flow model I created on CorelDRAW X6 showing the procedure I used to determine ideal bear habitats in Marquette County.

Sources

Landcover data is from USGS NLCD
-http://www.mcgi.state.mi.us/mgdl/nlcd/metadata/nlcdshp.html
DNR management units
-http://www.dnr.state.mi.us/spatialdatalibrary/metadata/wildlife_mgmt_units.htm
Streams from
-http://www.mcgi.state.mi.us/mgdl/framework/metadata/Marquette.html

S