Special Topics and Meth and Analyzation

        Welcome to the continuation of our look at statistical analysis with ArcMAP. Recall that the theme being explored with statistics is methamphetamine lab busts around Charleston West Virginia. These past two weeks of analysis have been the bulk of the work for this project. The overall objectives of the analysis portion of this project are to review and understand regression analysis basics, and a couple key techniques. Define what the dependent and independent variables for the study are as they apply to the regression analysis. Perform (multiple renditions) of an Ordinary Least Squares (OLS) regression model. Finally, complete 6 statistical sanity checks based on the OLS model outcomes.
       In the previous post we looked at a big overview of the area that is being analyzed. There are 54 lab busts from the 2004-2008 time frame from the DEA's National Clandestine Laboratory data. Decennial census data from 2000 and 2010 at the census tract level was spatially joined to these 54 lab busts. The data was then normalized into a percentage by census tract into 31 categories for analysis in the OLS model. These 31 categories of data were then fed into the model and systematically removed while analyzing their affect on the model. Ultimately as good a model as possible was arrived at with some results below.

        This is an extract of the table that I put together from the ArcMAP generated output to depict the OLS results. This was a cleaner format than the straight screen shot because it incorporates the descriptions of the individually labeled data elements. Key things to note for the table are that there are now only 11 variables being incorporated into the OLS model of the original 31. How were variables removed you might ask? There are six checks or questions to answer to determine the validity of a variables use in the OLS model: does an independent variable help or hurt the model; is its relationship to the dependent variable as expected; are there redundant explanatory variables; is the model biased; are there variables missing or unexplained residuals; how well does the model predict the dependent variable? The first three of these were generally grouped into 1 solid check for determining if a variable should stay or go. The remaining checks were applied to the model results as a whole. The key attributes to look at for a variable fall in line with [a], [b], [c],  as depicted on the table. As long as you had a coefficient not near zero, probability lower than .4, and VIF less than 7.5 a variable could stay. Not all of these match this criteria now, however you have to look at model functionality as a whole. The R-Squared [d] value is right at .7 (rounded up) which means that the model as it is accounts for 70% of the meth labs location based on the variables in use. This is pretty good when working with sociological data of this type. After looking at this data table its time to transition to the visual interpretation seen below.

     This map depicts the standard residual for the OLS model depicted in the table. It symbolizes areas using a standard deviation style outlook. However rather than wanting a more Gaussian curve style of data showing some of every color you ideally want values to be in the -0.5 to +0.5 range because that is said to be highly accurate. Darker browns indicate areas that the model predicted less meth labs than there actually were, and darker blues indicate high value areas where the model expected more meth labs than were actually present. Remember though that from our table above we are only doing a good job of predicting 70% of the total meth labs and the majority of the study area is still within 1 standard deviation.
    This weeks focus was not to describe the data results, but to accomplish the analysis leading up to it. Please follow up next week for a look at the finalized product. Thank you.

Remote Sensing: Unsupervised Classification

Welcome to this weeks discussion on unsupervised classification with remotely sensed imagery. This is a multifaceted lab looking at a number of different processes culminating in the unsupervised classification and manual reclassifying of the resulting raster dataset for a permeability analysis. The main objective was to understand and perform an unsupervised classification in both ArcMAP and ERDAS Imagine. Imagery for two different areas was provided, ultimately the UWF area as seen in the map below was the final subject matter for exploration of these topics.
Unsupervised classification is a classification method such that a software suite utilizes an algorithm to determine which pixels in the raster image are most like other pixels throughout the image and groups them accordingly. After the software has grouped the various pixels together it is up to the user to define what the grouped classes represent. For this type of classification the software is given certain user defined parameters such as number of iterations to run, confidence or threshold percentage to reach, and sample sizes. These essentially tell the software how long to run, what the minimum "correctly grouped" pixel percentage is, and how many pixels to look at adjusting at a time.  Lets look at how this was applied this week.

A high definition true color image of the UWF campus was used for the above analysis. This entailed performing a clustering algorithm on the true color image to group like pixels together and then export them as a slightly less defined image for storage space and processing speed concerns. The clustering algorithm created 50 classes, or shades of pixels which approximated the true color image. That was essentially the unsupervised classification. The software was told to produce 50 classes with 95% accuracy overall. Then I manually reclassified each of those 50 original classes into one of 5 labeled classes. I accomplished this by highlighting the pixel shade and reviewing it against the true color image and assigning it to the classes described. Four of the five classes are straight forward and represent what they say, with some possible error. The mixed class however is there because certain pixel shades applied to different items that represented both permeable or impermeable surfaces. For example some dead grass showing could show a tan pixel while a tan rooftop could also be showing the same value. So recoding this pixel to be grass or buildings would be wrong for at least some of that cluster of pixels. To account for this the mixed class was created, which is why you can see some rooftops as blue, grassy areas as blue or green and some blue sprinkled throughout.
Overall this is a fairly course analysis, but it does do a great job of exercising the process and creating likely results. Next week we will look at supervised classification in case that question came to mind. Thank you.

Remote Sensing and Thermal Analysis

Hello and welcome to this week in Remote Sensing where ill be discussing this weeks look at multi-spectral image analysis specifically looking for trends in thermal or infrared energy. This week was designed to focus on being able to compose a series of different raster bands into a composite image in both ERDAS Image and ArcMAP. Then having a multi-band composite image being able to manipulate that bands being displayed by color, and interpreting the resulting image. A couple different images were provided by UWF to exercise these skills and to ultimately come up with a user derived analysis of some particular feature. Lets look at my composite map and then we will discuss how I met the objectives above.

This is a thermal overview of Florida's emerald coast. The image is as of February 2011 provided by UWF. I have made a composite of the original 7 bands of information available. The main map and two insets are all the same image with different band combinations or visualizations of the available data. The central feature to all three images is a large oblong clearing. What is this clearing? It is one of the many available military firing ranges located along the panhandle. What am I trying to do with it. Overall im trying to differentiate this area from its surroundings using thermal imagery. The purple image comes from a unique combination of infrared both short wave and thermal bands to provide brightness to the "hottest" areas. These are those areas that heat up and or emit the best. You can see that there is a very similar spectral signature all along the island to the south. Santa Rosa island is made up of white sand beaches and dunes and appears as the only feature that might be spectrally similar to the artillery ranges. The color inset is another infrared look at the area, but rather than a grayscale color has been used to help give characteristic spectral pattern to the other images. Additionally for reference you can see that when viewed in true color that ranges do stand out fairly easily, and it was this resolution and clarity of image that I wanted to have replicated using an all infrared color image. I think it worked well. What about you? Ultimately this banding combination is great at identifying land clearings specifically, but you can also see a good range of vegetation density and land water contrast even if that wasnt my original focus. Thank you.

Special Topics and Statistics and Meth Labs

                Hello and welcome back to my blog. We are beginning a fresh topic for the net few weeks worth of assignments and posts. This project will have us delving deep into the clandestine, the dangerous, and ultimately bad world of drugs. Specifically we will be examining the role of GIS statistical analysis as it applies to aiding law enforcement with determining ideal locations to find methamphetamine labs. Methamphetamine's have been around since the early 1920's, and illegal since the 60's which drove the illicit trade underground. Meth labs have been found in every state, but surprisingly only in about half the countries counties. This leaves a huge disparity in the national / state level problem and the county level. Over the next few weeks I will be analyzing tow particular counties of West Virginia, Putnam and Kanawha. These counties are credited with 187 meth lab busts from 2004-2008. The majority of which come after 2005 which saw the introduction of the Combat Methamphetamine Epidemic Act of 2005 aimed at eliminating the over the counter acquisition of pills which could be distilled into meth generating substances. Once again, the overall goal is to explore the uses of GIS in this incredibly relevant topic to the nation. Chances are we all know someone who has been impacted through drugs or drug use, or at minimum you can see it all too prevalent in the news. The idea behind this lab is to examine the socioeconomic  trend information that can aid in determining where meth labs are most likely to be to give the information to law enforcement. The end deliverable for this will be a scientific paper discussing the issue and analysis being done on the below study area. 

As stated earlier the study area in the Charleston area of West Virginia was home to 187 meth lab busts. This information has already been summarily broken down into a meth lab density by census tract in the above main map. This essentially means the total number of busts per census tract was divided by the area of the tract to give us the density values seen in the legend. This map also provides a basic overview of the subject counties and provides state context as well. Additionally an extra tidbit was added in the cities that are displayed, you can see that throughout this study area there are only 4 named cities that house over 10,000 people apiece. This may or may not factor into the study. Time will tell. 



I will leave you with this:

Remote Sensing and Multispectral Analysis

      Welcome to this weeks remote sensing topic revolving around multi-spectral analysis through spectral enhancement. Essentially this means to take existing spectral data and present it in a manner that might bring out certain relationships or patterns not readily present in other presentations. This weeks objective was to study an image set and identify certain spectral relationships that aren't readily seen looking at a standard true color image. This is done by manipulating that pixel values to show other relationships through gray scale panchromatic views of single spectral bands or different combinations of multiple bands such as that seen from a standard false color infrared image. Both ERDAS Imagine and ArcGIS were used to explore the given image. Several tools within ERDAS were used, such as the Inquire cursor to look at particular groups of pixels for there relevant brightness information. Histograms and contrast information were used to identify patterns within multi-spectral and panchromatic views of one or more spectral bands. The image manipulated below was provided by UWF, and the assignment was to identify three different sets of unique spectral characteristics present within the image and to build maps around those.

The first criteria involved locating the feature in spectral band 4 that correlates to a histogram spike in value between 12 and 18.

As stated above, the first criteria involves investigation specifically into band 4 which is generally associated with near infrared (NIR) energy and is good for looking at vegetation and soil and crop land and water contrasting. With this task I needed to look at the histogram and find the resulting spike seen in the lower right inset of the histogram in question. From there I specifcally made this the only "visible" feature in the map to the left of it. You can see that in both the standalone feature map and the gray scale band 4 that the water does stand out quite significantly.

The second criteria involved locating the feature that represents both a spike in the visual and NIR bands with a value around 200, and a large spike in the infrared layers of bands 5 and 6 around pixel values 9 to 11.

The main features of this image are displayed in a false natural color employing bands 5,4,3 in that order for red, green, and blue. This color combo does particularly well at letting the areas that are being inquired about be displayed You can see that the insets display different extents of the same data in different spectral scenes. The two separate breakdowns of pixels of value 200 in Visual bands and values 9-11 in the infrared bands are compared in the lower right. In the visual inset in the lower left this appears to be snow within some mountainous valleys.

The last criteria being looked for revolves around water features that when looking at bands 1-3 become brighter than usual, but remain relatively constant in bands 5 and 6.

A true or natural color image is opposite a custom arrangement of bands 6 displaying band 7 short wave IR, and green and blue bands showing their own respective colors for enhancement. If you look at the natural color image you can see a river in the upper right portion that is much darker than the water way that is featured in the other images. This inlet is then highlighted spectrally in the other images. Focusing on the green and blue and removing the red, allowing for pinks and reds to be limited to open ground you get a much better highlight of the likely shallower inlet area. You can see also that false color IR does highlight the bay only by contrasting it against the vegetation which stands out so well.

Wrapping it up this assignment was about taking one image which is a combination of bands of spectral information and organizing that information in different ways to make certain features more readily apparent over others. Thank you.

Remote Sensing: Preprocessing and Enhancement

Welcome to this weeks remote sensing topic of image pre-processing and spatial enhancement. This week focuses on acquiring remotely sensed information, utilizing programs like ERDAS Imagine or ArcGIS to modify the image to make its information more readily apparent then what might be available at first glance. This is known as spatial enhancement. To facilitate this weeks objectives of utilizing various filters and image transformations I was supplied with a 2003 Landsat 7 image which had some serious stripping effects due to a Scan Line Corrector error. Exploration using low pass, high pass, and Fourier Transformation led up to the actual work on the below map. Low pass enhancements essentially apply a mean kernel filter to smooth the detailed areas letting larger similarities show through. The opposite of this is a High pass filter which enhances the small details like road patterns. The Fourier Transformation was used as the basis of the processing of the below image which was used to reduce the stripping present in the lower inset.

An inset of the initial image is shown in the lower left with the culminated result of ERDAS Imagine and ArcMAP display in the large screen. Stripping is still prevalent but the result is much cleaner than the original. The information on the map details that a Wedge and Low pass transformations were key to the initial processing, followed by stretched symbology adjustments focusing on histogram display by standard deviation.
All other courses throughout this program so far have given already processed or corrected imagery to this point. This is an excellent intro to what to do to imagery that is not already at usable quality. Stay tuned as this concept is broadened throughout the coming weeks.

Special Topics and Mountaintop Removal Report

The past two posts for Special Topics have focused on the preparation and analysis of Mountaintop Removal (MTR) areas located around the Appalachian Mountain chain in the mid East United States. The analysis culminated in the report here for the Group 3 area. The overall purpose of the report is too compare 2005 MTR measurements provided by Skytruth.org to an internal study by UWF of 2010 available imagery.

The last post outlined the reclassification of the Landsat rasters for a particular area and here we are looking at the fruits of that labor. The group 3 analysis team, lead and coordinated by myself combined data for two Landsat areas and then eliminated areas of noise and interference. The noise and interference refer to areas that show similar spectral signatures to MTR sites. River erosion, urbanization, agricultural clearing, and roadways can all show similar reflectance to MTR sights. As such these areas were attempted to be removed by discounting areas within 400m of rivers, 400m of highways, and 50m of small streams and roads. One key area of interference remained afterward which was clouds on the imagery. When reclassified some of these areas could not be cleared away and left some false positive MTR sites. These have been highlighted on the map below, and likely account for the inaccuracy described on the map.

The map above is based around the two combined Landsat Raster images which were used to derive the MTR sites through unsupervised reclassification during the analysis phase. These are overlayed on the 2005 identified sites. With the exception of the areas in the north west explained on the image the data supports that there is an overall decline in MTR activity. This is likely as a result of being on the western edge of the study area as whole with MTR progressively moving eastward into more rich regions of the US coal fields. These fields arent depicted here. However this blog entry, and associated map are only one part of the project as a whole. There is other analysis and conclusions drawn as a part of this project at the following link: http://arcg.is/1Yxokr7 . This is an ArcGIS Online Story Map Journal dedicated to this project. Overlayed there with this data on the analysis tab is a depiction of the coal fields. I invite you to look at the more indepth discussion on this analysis and MTR activity through the jounal. Thank you.