How GIS Can Help With Tribal Safety Planning

Size: px

Start display at page:

Download "How GIS Can Help With Tribal Safety Planning"

Brook Bradley
6 years ago
Views:

1 How GIS Can Help With Tribal Safety Planning Thomas A. Horan, PhD Brian Hilton, PhD Arman Majidi, MAIS Center for Information Systems and Technology Claremont Graduate University

2 Goals & Objectives This study investigates how advances in GIS can enhance the availability, analysis, and use of transportation safety information on Tribal Lands. Conducted in partnership with Esri, the study focuses on traffic safety processes including awareness, detection, reporting, analysis, programming and improvements. The study develops GIS prototypes that will be evaluated through a series of collaborative assessments. The study includes spatial analysis of severe crashes and fatalities both in and around Tribal Areas.

3 Methodology Review and assess current issues, barriers, uses, data, and interests in using GIS for Tribal transportation safety. Devise a series of prototype applications that could be used by Tribes to assist in their transportation safety data gathering, planning, assessment and implementation efforts. Assess the Tribes interest in implementing GIS applications for planning, analysis, and programming in Tribal safety planning. Conduct spatial analysis of the value, barriers and implementation paths for promising GIS prototype applications.

4 Task 2: Prototypes To date, we have developed 6 easy to use prototype applications for Tribal safety 1. Hotspot Application Hot spot analysis is a spatial cluster detection method that identifies spatial concentrations of crashes and fatalities. It can provide preliminary indicators of historical trends in Tribal safety conditions crash and fatalities. 2. Roadway safety Application This application allows for visualization and analysis of road safety audits and related segment analysis on roads on or near Tribal lands. Road segment analysis identifies segments of roads and highways that are more prone to traffic incidents due to the characteristics of the roadway (such as potholes, curves, extreme sloop grading, etc.) 3. Pedestrian Safety Application This application aggregates and analyzes a variety of information related to Tribal pedestrian safety. This information can include crash and fatality information, community input data, bike lane availability, and intersection cross walk identification. This information is aggregated and analyzed for better visualization and representation of particular concerns, for possible improvements, and for suitable locations of intersection crosswalks and bike lanes.

5 Task 2: Prototypes (cont.) To date, we have developed 6 easy to use prototype applications for Tribal safety 4. Safety Emphasis Application This application integrates a range of Tribal Safety quantitative and qualitative analyses into an integrated view of identified emphasis areas for safety improvement. In addition to the safety hotspot, segment and community data, it allows for spatial identification of proposed emphasis areas into an integrated summary of proposed safety improvements. 5. Safety Story Application This application allows the Tribal Transportation safety plans to be analyzed and dis played as a story map. It provides a narrative of Tribal Safety conditions, analyses, and priority improvements. The story map can be used to convey the overall Tribal Safety approach to a variety of stakeholders. 6. Safety Application Portal The Tribal Safety portal provides an easy to access means to obtain the Tribal safety applications for use, store tribal maps and analyses, and share among the Tribal com munity. It allows for maps to quickly be used, updated, and integrated into a variety of reports.

6 Hotspot Safety Application Regional Mille Lacs

7 Regional Roadway Safety Application Mille Lacs

8 Pedestrian Safety Application Regional Mille Lacs

9 Safety Emphasis Application Regional Mille Lacs

Obvious Safety Concerns http://agis.maps.arcgis.

10 Safety Story Application Show specific areas of concern outlined in surveys or by community leaders Identify Non Obvious Safety Concerns

11 Safety Application Portal An online collaborative environment Group sites devoted to Tribe specific concerns Online web mapping application

12 Hotspot Application Feedback How useful is it to have hotspot identification of crash clusters? A) Very useful B) Partially Useful C) Somewhat Useful D) Not very useful E) Not useful at all # OF VOTES HOTSPOT RESPONSE A RESPONSE B RESPONSE C RESPONSE D RESPONSE E

13 Roadway Safety Application Feedback How useful is it to have mapping of high risk road segments? A) Very useful B) Partially useful C) Somewhat useful D) Not very useful E) Not useful at all

14 Pedestrian Safety Application Feedback How useful is it to have maps of pedestrian safety incidents? A) Very useful B) Partially useful C) Somewhat useful D) Not very useful E) Not useful at all AXIS TITLE 11 PEDESTRIAN SAFETY RESPONSE A RESPONSE B RESPONSE C RESPONSE D RESPONSE E RESPONSE

15 Safety Emphasis Application Feedback How useful is it to maps linking high risk safety areas with (proposed) priority safety improvements? A) Very useful B) Partially useful C) Somewhat useful D) Not very useful E) Not useful at all

16 Safety Story Application Feedback How useful is it to have maps and related data integrated into safety plans via Story Maps? A) Very useful B) Partially useful C) Somewhat useful D) Not very useful E) Not useful at all

17 Safety Application Portal Feedback How useful is it to have the safety application portal? A) Very interested B) Interested C) Indifferent D) Partially interested E) Not interested

18 Interactive Feedback How interested are you to use these applications? A) Very interested B) Interested C) Indifferent D) Partially interested E) Not interested

19 Interactive Feedback What is your favorite prototype? A) Hotspots & Road Segments FAVORITE PROTOTYPE B) Pedestrian Safety 5 C) Safety Risk Assessment D) Safety Story App # OF VOTES E) Safety Application Portal 1 RESPONSE A RESPONSE B RESPONSE C RESPONSE D RESPONSE E RESPONSE

20 Next Phase Four Tribal Areas (Leech Lake, Mille Lacs, Red Lake, and White Earth) were selected for additional GIS based statistical analysis. These areas were selected due to their high percentage of Fatal and Incapacitating Injuries, as well as containing a sufficiently large sample size. Route Segments data represents road centerlines for all public roads within the state of Minnesota. This data was obtained from the Minnesota Department of Transportation (MnDOT). The Route Segments dataset was developed to fill a need at MnDOT for a continuous, statewide GIS base map of the transportation system. The Route Segments dataset was created in the Summer First, a 10 mile buffer around the Tribal area was created to comparison. Then another more accurate method to identify crashes adjacent to these Tribal Areas using 15 minute drive time was used. A drive time polygon is an area that encompasses all accessible streets that lie within a specified drive time from a specific point.

21 Detailed Descriptives for Selected Tribes Tables 5 7 provide detailed descriptives for the 4 selected Tribes, the surrounding area, and the combination of both areas. Table 5 Crashes Occurring on Tribal Areas Non Incapcitating Fatal Incapcitating Injury Injury Possible Injury Total Leech Lake % % % % 500 Mille Lacs % % % % 303 Red Lake % % % % 11 White Earth % % % % 333 Table 6 Crashes Occurring Adjacent to Tribal Areas (15 minute Drive time) Fatal Incapcitating Injury Non Incapcitating Injury Possible Injury Total Leech Lake % % % 1, % 1,779 Mille Lacs % % % % 1,769 Red Lake % % % % 792 White Earth % % % % 1,103 Table 7 Crashes Occurring both on Tribal Areas and Adjacent Areas Non Incapcitating Fatal Incapcitating Injury Injury Possible Injury Total Leech Lake % % % 1, % 2,279 Mille Lacs % % % 1, % 2,072 Red Lake % % % % 803 White Earth % % % % 1,436

Hotspot Analysis Various Hot Spot Analyses were conducted on these fours data sets. Hot Spot Analysis calculates the Getis Ord Gi* statistic for each feature in a dataset.

22 Hotspot Analysis Various Hot Spot Analyses were conducted on these fours data sets. Hot Spot Analysis calculates the Getis Ord Gi* statistic for each feature in a dataset. The resultant z scores and p values indicate where features with either high or low values cluster spatially. This study is interested in identifying features that are statistically significant hot spots feature that have a high value and are surrounded by other features with high values. The local sum for a feature and its neighbors is compared proportionally to the sum of all features; when the local sum is very different from the expected local sum, and that difference is too large to be the result of random chance, a statistically significant z score results. These values are typically displayed as illustrated in Figure 1. In the analysis performed in this study, one question is where are there unexpectedly high spatial clusters of injuries, in particular, fatal injuries, given all injuries?. This question was examined in relation to the Tribal Area, the Adjacent Area, and both areas combined. Figure 1

Comparing Tribal Areas (road mile normalized z score) with their Adjacent Areas (road mile normalized z score) indicates the following: 1 All Tribal Areas have z scores greater than their Adjacent

23 Comparing Tribal Areas (road mile normalized z score) with their Adjacent Areas (road mile normalized z score) indicates the following: 1 All Tribal Areas have z scores greater than their Adjacent Areas, that is, they are worse off. 2 Ranking these in terms of severity (greatest difference between z scores) indicates that Red Lake is the worst Tribal Area followed by White Earth, Mille Lacs, and Leech Lake. Figure 7 Comparing Tribal Areas with their Adjacent Areas Tribal Area Avg Z Score Adjacent Area Avg Z Score Absolute Value Z Score Rank in Terms of Severity Leech Lake Mille Lacs Red Lake White Earth Very high (positive) or very low (negative) z scores, associated with very small p values, are found in the tails of the normal distribution. In the case of the output presented here, very high z scores are indicative of hot spots while very low z scores are indicative of cold spots.

24 Hotspot Analysis Red Lake Figure 3 Statistically Significant Hot Spots and Cold Spots for Red Lake Hot Spot Analysis Tribal Area Hot Spot Analysis Combined Areas

Comparing the Hotspot Analyses This step was followed by an analysis where each Tribal Area was compared to the normalized Hot Spot values for the Combined Tribal Areas and Adjacent Areas.

25 Comparing the Hotspot Analyses This step was followed by an analysis where each Tribal Area was compared to the normalized Hot Spot values for the Combined Tribal Areas and Adjacent Areas. In this case, Leech Lake has the highest proportion of Hot Spots per Road Mile per Combined Area followed by White Earth, Red Lake, and Mille Lacs. Figure 7 Hot Spots per Road Mile per Combined Area

In this case, Mille Lacs has the highest proportion of Hot Spots per Road Mile

26 Comparing the Hotspot Analyses Next the analysis normalized the Hot Spot values based on the total number of road miles within the Tribal Areas. In this case, Mille Lacs has the highest proportion of Hot Spots per Road Mile followed by Red Lake, White Earth, and Leech Lake. Figure 5 Tribal Areas Hot Spots per Road Mile

Comparing the Hotspot Analyses Following this, the next step in the analysis was to normalize the Hot Spot values based on the total number of road miles within the Combined Tribal Areas and Adjacent

27 Comparing the Hotspot Analyses Following this, the next step in the analysis was to normalize the Hot Spot values based on the total number of road miles within the Combined Tribal Areas and Adjacent Areas. In this case, Mille Lacs has the highest proportion of Hot Spots per Road Mile followed by White Earth, Leech Lake, and Red Lake. Figure 6 Combined Tribal Areas and Adjacent Areas Hot Spots per Road Mile

28 Summary of Findings There are a range of GIS applications that can be useful to Tribal stakeholders. The value of the applications is relative to how it assists in the workflow of safety planning and application. Each Tribal land studied had traffic crash hotspots, through often there were greater hotspots in adjacent areas. When the spatial crashes were normalized by road miles, each tribe has a greater spatial rate of occurrence, as compared to adjacent areas. Normalizing by exposure is the next step in analysis.

29 Contact Us For More Information Center for Information Systems and Technology Contact: Thomas Horan (909) Visit us on the web

Recipient Organization Thomas A. Horan Recipient Or ganization Recipient Organization Brian Hilton Recipient Identifying Number

Recipient Organization Thomas A. Horan Recipient Or ganization Recipient Organization Brian Hilton Recipient Identifying Number Human-centered solutions to advanced roadway safety Using GIS to Improve Tribal Safety: Applications, Trends, and Implementation Dimensions Thomas A. Horan School of Business University of Redlands Brian