1. 1 Seattle JSM Session G. P. Patil August 7, 2006

2. 2 Statistical GeoInformatics of Hotspot Detection and Prioritization for Early Warning and Disaster Management G.P. Patil 1, Luiz Duczmal 2, Reza Modarres 3, and Stephen Rathbun 4 1 Penn State University, University Park, PA, U.S.A. 2 Universidade Federal de Minas Gerais, Belo Horizonte, Brazil 3 George Washington University, Washington, D.C., U.S.A. 4 University of Georgia, Athens, GA, U.S.A.

3. 3 Statistical Issues in Disaster Response Disaster means occurrence of Widespread Destruction and Distress. Widespread means spread in space and time. Destruction is of resource. Distress is to life. Thus disaster impacts resource and life in space and time over a geographic region or over a network. Response is before, during, or after, leading to early warning, prediction, prevention, or to intervention, monitoring or to management, mitigation, restoration. And the response issue involves where, when, whether, what, how, and why. Thus statistical issues in disaster response amount to statistical geoinformatics of hotspot detection and prioritization for monitoring, etiology, early warning, and management.

4. 4 Disaster Data Geospatial Temporal Warehousing Various Sources and Data Products Remote Sensing and Image Analysis Data Compression and Segmentation Data Mining, Pattern Analysis, … Resource Identification and Measurement Environmental Resources, Ecological Resources, Human Resources, Health Resources,… Geospatial Temporal Cells, Network Nodes, and Indicators Construction and Validation Indicator Based Cellular Surface, Network Surface Pixels or Individuals as Units in the Cells/Nodes

5. 5 Some Examples Human Disease: West Nile Virus, Lyme Disease, Dengue Fever, Avian Flu. Crop Disease: Spray or Not Spray. Poverty Alleviation: Poverty patch trajectory. Youth Unemployment: Pockets of unemployment and their trajectory. Watershed Surveillance; Water Harvesting; Invasive Species; Biodiversity Hazardous Waste Site Management Regional Vulnerability Assessment Networked Infrastructure Security: Drinking water networks, subway networks, hospital network, sensor networks. Robotics Command and Control Oil Spill Carbon Sources and Sinks Anthrax: Home to office mobility matrix for outbreak study. Tsunami: Tsunami inundation mapping and evacuation routing.

6. 6 Abstract Geoinformatic surveillance for spatial and spatiotemporal hotspot detection and prioritization, early warning, and sustainable management is a critical need for the 21 st century. A hotspot can mean an unusual phenomenon, anomaly, aberration, outbreak, or critical area. Hotspot delineation and prioritization may be required for etiology, management, or early warning. Case studies may relate to different kinds of disasters, such as tsunami inundation, wildfires, tornadoes, disease outbreaks, persistent poverty trajectories, etc. A declared need is around for geoinformatic surveillance statistical science and appropriate software infrastructure with capability for detection and prioritization of arbitrarily shaped hotpsots including emerging hotspots. Our detection innovation employs the notion of an upper level set, and is accordingly called the upper level set scan statistic system (Patil and Taillie 2004, Environmental and Ecological Statistics, 11, 183-197). The prioritization innovation employs the notion of a partially ordered set, and is accordingly called the poset prioritization system (Patil and Taillie, 2004, Environmental and Ecological Statistics, 11, 198-228). In this presentation, we will discuss these innovations and the investigations in progress relating to a variety of case studies involving natural disaster management and early warning.

7. 7 Agency Databases Thematic Databases Other Databases Homeland Security Disaster Management Public Health Ecosystem Health Other Case Studies Statistical Processing: Hotspot Detection, Prioritization, etc. Data Sharing, Interoperable Middleware Standard or De Facto Data Model, Data Format, Data Access Arbitrary Data Model, Data Format, Data Access Application Specific De Facto Data/Information Standard Agency Databases Thematic Databases Other Databases Homeland Security Disaster Management Public Health Ecosystem Health Other Case Studies Statistical Processing: Hotspot Detection, Prioritization, etc. Data Sharing, Interoperable Middleware Standard or De Facto Data Model, Data Format, Data Access Arbitrary Data Model, Data Format, Data Access Application Specific De Facto Data/Information Standard National Applications Biosurveillance Carbon Management Coastal Management Community Infrastructure Crop Surveillance Disaster Management Disease Surveillance Ecosystem Health Environmental Justice Environmental Management Environmental Policy Homeland Security Invasive Species Poverty Policy Public Health Public Health and Environment Robotic Networks Sensor Networks Social Networks Syndromic Surveillance Tsunami Inundation Urban Crime Water Management Statistical Geoinformatics of Hotspot Detection, Prioritization, Early Warning, and Management NSF Digital Government Project #0307010 PI: G. P. Patil Federal Agency Partnership CDCDOD EPANASA NIHNOAA USFSUSGS gpp@stat.psu.edu Websites: http://www.stat.psu.edu/~gpp/ http://www.stat.psu.edu/hotspots/ http://www.digitalgovernment.org/news/stories/2004/1104/1104_hotspots_heyman.jsp http://www.digitalgovernment.org/news/stories/2006/0306/0306_patil_heyman.jsp NSF Digital Government hotspot geoinformatics project, federal agency partnership, and national applications for digital governance.

8. 8 Geographic and Network Surveillance for Arbitrarily Shaped Hotspots Overview Geospatial Surveillance Upper Level Set Scan Statistic System Spatial-Temporal Surveillance Typology of Space-Time Hotspots Hotspot Prioritization Ranking Without Having to Integrate Multiple Indicators Statistical Geoinformatics for Hotspot Detection, Prioritization, Early Warning and Sustainable Management Upper Level Set Scan System Definition: A hotspot is that portion of the study region with an elevated risk of an adverse outcome Federal Agency Partnerships CDC DOD EPA NASA NIH NOAA USFS USGS Features of ULS Scan Statistic: Identifies arbitrarily shaped hotspots Applicable to data on a network Confidence sets and hotspot ratings Computationally efficient Generalizes to space-time scan Poset Prioritization System Objective: Prioritize or rank hotspots based on multiple indicator and stakeholder criteria without having to integrate indicators into an index, using Haase diagrams and partially ordered sets. Example: Prioritization of disease clusters with Multiple Indicators National Applications and Case Studies Biosurveillance Carbon Management Costal Management Community Infrastructure Crop Surveillance Disaster Management Disease Surveillance Ecosystem Health Environmental Justice Sensor Networks Robotic Networks Environmental Management Environmental Policy Homeland Security Invasive Species Poverty Policy Public Health Public Health and Environment Syndromic Surveillance Social Networks Stream Networks g Changing Connectivity of ULS as Level Drops G.P. Patil, R. Acharya, M. Haran, W.L. Myers, and P. Patankar The Pennsylvania State University R. Modarres George Washington University Example: West Nile Virus First isolated in 1937, this mosquito born disease, indigenous to north Africa, the Middle East and west Asia was first introduced into the United States in 1999. Disease Count Quintiles Population Quintiles Disease Rate QuintilesLikelihood Quintiles Comparison of ULS Scan with Circular Scan ULS ScanCircular Scan Confidence set for ULS Hotspot Hotspot Membership Rating Example: Lyme Disease Infections from the bacterium Borelia burgdorfei vectored by ticks from the genus Ixodes. ULS Scan Cylindrical Scan Example: Human-environment indicator values for 16 European countries. There are a total of 3,764,448 admissible linear extensions. The cumulative rank function for Sweden exceeds that of all remaining countries. The crf’s of all countries dominate that of Ireland. The remaining countries cannot be uniquely ordered based on their crf’s. Belgium, Netherlands and United Kingdom have identical crf’s. Admissible linear extensions are comprised of rankings compatible with the rankings of all indicators. Treating each linear extension as a voter, the cumulative rank function is obtained from the frequencies at which each object receives each rank. Disease Rates Comparison of ULS Scan with Cylindrical Scan 1997 1998 1999 Year 2000 2001 2002 2003 Haase Diagram The crf’s also form a partially ordered set. There are only 182 admissible linear extensions for this poset, yielding the cumulative rank function: One more iteration yields the rankings in the data table. Center for Statistical Ecology and Environmental Statistics S.L. Rathbun University of Georgia

9. 9 NSF Digital Government hotspot geoinformatics project, federal agency partnership, and national applications for digital governance. - Publications - G. P. Patil and C. Taillie (2003). Geographic and network surveillance via scan statistics for critical area detection. Statistical Science, 18 (4), 457-465. G.P.Patil (2003). BOF Report: Geoinformatic surveillance: Hotspot detection and prioritization across geographic regions and networks for digital government in the 21st century. In: Proceedings of the 4th Annual International Conference on Digital Government Research, Boston, MA, May 18-22, 2003, pp. 421-422. G. P. Patil and C. Taillie. (2004a). Upper level set scan statistic for detecting arbitrarily shaped hotspots. Environmental and Ecological Statistics, 11 (2), 183-198. G. P. Patil and C. Taillie (2004b). Multiple indicators, partially ordered sets, and linear extensions: Multi-criterion ranking and prioritization. Environmental and Ecological Statistics, 11 (2), 199-228. G.P. Patil and C. Taillie (2004c). Geoinformatic surveillance hotspot prioritization using linear extensions of partially ordered sets for multi-criterion ranking with multiple indicators. In: Proceedings of the 5th Annual International Conference on Digital Government Research, Seattle, WA, May 24-26, 2004, pp. 57-58. G.P.Patil, S. Rathbun, R. Acharya, P. Patankar, R. Modarres (2005). Upper Level Set Scan System for Detecting Arbitrarily Shaped Hotspots for Digital Governance, In: Proceedings of the 6th Annual International Conference on Digital Government Research, Atlanta, GA, May 18-25, 2005, pp. 281- 282. G. P. Patil, R. Modarres, W.L. Myers, P. Patankar (2006a). Spatially Constrained Clustering and Upper Level Set Scan Hotspot Detection in Surveillance Geometrics. Environmental and Ecological Statistics, Special Institutional Thematic Issue: Penn State Cross-Disciplinary Classroom in Statistical Ecology and Environmental Statistics, Volume 13 (4). In press. Reza Modarres and G.P. Patil (2006b). Hotspot Detection with Bivariate Data. Journal of Statistical Planning and Inference (S.N. Roy Centennial Volume) In press. G.P. Patil, R. Acharya, R. Modarres, W.L. Myers, and S.L. Rathbun (2006c). Hotspot Geoinformatics for Digital Governance. In: Encyclopedia of Digital Government, Volume II, Idea Group Publishing, Hershey, PA, 2006, A.V. Anttiroiko, M. Malkia (Editors).

10. 10 G.P. Patil, Raj Acharya, Amy Glasmeier, Wayne Myers, Shashi Phoha, and Stephen Rathbun (2006d). Hotspot Detection and Prioritization – Geoinformatics for Digital Governance, In: Digital Government: Advanced Research and Case Studies. Springer Publ., H. Chen. L. Brandt, V. Gregg, R. Traunmüller, S. Dawes, E. Hovy, A. Macintosh, C. Larson (Editors). G.P. Patil (2006e). Digital Governance and Hotspot Geoinformatics for Monitoring, Etiology, Early Warning, and Sustainable Management, In: Proceedings of the 7th Annual International Conference on Digital Government Research, San Diego, CA, May 21-24, 2006, p. 75-76. S.L. Rathbun and G.P. Patil (2006f). Spatiotemporal Geoinformatic Disease Surveillance, Joint Statistical Meetings, August 6-10, 2006, Seattle, WA. G.P. Patil, Raj Acharya, Wayne Myers, Shashi Phoha, and Rajan Zambre, Hotspot Geoinformatics for Detection, Prioritization, and Security (2006g). In: Encyclopedia of Geographical Information Science, Shashi Shekhar and Hui Xiong (Editors). G.P. Patil, K. Sham Bhat, and Michael Kase (2006h). Evaluation of Multiple Indicators for Conditions of Watersheds in the Atlantic Slope Consortium: Multicriteria Prioritization and Ranking with Differential Weights, Stepwise Aggregations, Hasse Diagrams, Poset Cumulative Rank Frequency Operators, and Markov Chain Monte Carlo Methods. Center for Statistical Ecology and Environmental Statistics Technical Report Number 2006-0503, The Pennsylvania State University, University Park, PA. G.P. Patil, Jessica Newlin, K. Sham Bhat, and Michael Kase (2006i). Evaluation of Multiple Indicators for Stream Channel Stability Near Bridges in the United States: Multi-Criteria Prioritization and Ranking with Differential Weights, Stepwise Aggregation, Hasse Diagrams, Poset Cumulative Rank Frequency Operators, and Markov Chain Monte Carolo Methods. Center for Statistical Ecology and Environmental Statistics Technical Report Number 2006-0504, The Pennsylvania State University, University Park, PA. G.P. Patil and K. Sham Bhat (2006j). Unmasking Weight Camouflage of a Composite Index Based on Multiple Indicators. Center for Statistical Ecology and Environmental Statistics Technical Report Number 2006-0712. The Pennsylvania State University, University Park, PA. Wayne L. Myers and Ganapati P. Patil (2006k). Partial Order and Rank Range Runs for Compositional Complexes. Center for Statistical Ecology and Environmental Statistics Technical Report Number 2006-0912, The Pennsylvania State University, University Park, PA. G.P. Patil, R. Bruggemann, and A. Warke (2006l). On Multicriteria Prioritization and Ranking, Using Posets, Hasse Diagrams, and Monte Carlo Algorithms with Applications—A Comparative Study (in preparation).

11. 11 Publications in Press Springer Environmental and Ecological Statistics Monographs Series Editors: G.P. Patil, T. Gregoire, Andrew Lawson Volume 1: Landscape Pattern Analysis for Assessing Ecosystem Condition By G. D. Johnson and G. P. Patil Volume 2: Pattern-Based Compression of Multi-Band Image Data for Landscape Analysis By W. L. Myers and G. P. Patil Springer Journal on Environmental and Ecological Statistics Editor-in-Chief: G.P. Patil Deputy Editor: T. Gregoire Special Issue 1: Hotspot Geoinformatics: December 2004 Special Issue 2: Hotspot Geoinformatics: December 2006

12. 12

13. 13

14. 14 The Spatial Scan Statistic Move a circular window across the map. Move a circular window across the map. Use a variable circle radius, from zero up Use a variable circle radius, from zero up to a maximum where 50 percent of the population is included.

15. 15 A small sample of the circles used

16. 16 Detecting Emerging Clusters Instead of a circular window in two dimensions, we use a cylindrical window in three dimensions. Instead of a circular window in two dimensions, we use a cylindrical window in three dimensions. The base of the cylinder represents space, while the height represents time. The base of the cylinder represents space, while the height represents time. The cylinder is flexible in its circular base and starting date, but we only consider those cylinders that reach all the way to the end of the study period. Hence, we are only considering ‘alive’ clusters. The cylinder is flexible in its circular base and starting date, but we only consider those cylinders that reach all the way to the end of the study period. Hence, we are only considering ‘alive’ clusters.

17. 17 Major epicenter on Staten Island Dead bird surveillance system: June 14 Dead bird surveillance system: June 14 Positive bird report: July 16 (coll. July 5) Positive bird report: July 16 (coll. July 5) Positive mosquito trap: July 24 (coll. July 7) Positive mosquito trap: July 24 (coll. July 7) Human case report: July 28 (onset July 20) Human case report: July 28 (onset July 20)

18. 18 Hospital Emergency Admissions in New York City Hospital emergency admissions data from a majority of New York City hospitals. Hospital emergency admissions data from a majority of New York City hospitals. At midnight, hospitals report last 24 hour of At midnight, hospitals report last 24 hour of data to New York City Department of Health A spatial scan statistic analysis is performed every morning A spatial scan statistic analysis is performed every morning If an alarm, a local investigation is conducted If an alarm, a local investigation is conducted

19. 19 Issues

20. 20 Geospatial Surveillance

21. 21 Spatial Temporal Surveillance

22. 22 Attractive Features Identifies arbitrarily shaped clusters Identifies arbitrarily shaped clusters Data-adaptive zonation of candidate hotspots Data-adaptive zonation of candidate hotspots Applicable to data on a network Applicable to data on a network Provides both a point estimate as well as a confidence set for the hotspot Provides both a point estimate as well as a confidence set for the hotspot Uses hotspot-membership rating to map hotspot boundary uncertainty Uses hotspot-membership rating to map hotspot boundary uncertainty Computationally efficient Computationally efficient Applicable to both discrete and continuous syndromic responses Applicable to both discrete and continuous syndromic responses Identifies arbitrarily shaped clusters in the spatial-temporal domain Identifies arbitrarily shaped clusters in the spatial-temporal domain Provides a typology of space-time hotspots with discriminatory surveillance potential Provides a typology of space-time hotspots with discriminatory surveillance potential Hotspot Detection Innovation Upper Level Set Scan Statistic

23. 23 Data-adaptive approach to reduced parameter space  0 Data-adaptive approach to reduced parameter space  0 Zones in  0 are connected components of upper level sets of the empirical intensity function G a = Y a / A a Zones in  0 are connected components of upper level sets of the empirical intensity function G a = Y a / A a Upper level set (ULS) at level g consists of all cells a where G a  g Upper level set (ULS) at level g consists of all cells a where G a  g Upper level sets may be disconnected. Connected components are Upper level sets may be disconnected. Connected components are the candidate zones in  0 These connected components form a rooted tree under set inclusion. These connected components form a rooted tree under set inclusion. –Root node = entire region R –Leaf nodes = local maxima of empirical intensity surface –Junction nodes occur when connectivity of ULS changes with falling intensity level ULS Scan Statistic

24. 24 Upper Level Set (ULS) of Intensity Surface Hotspot zones at level g (Connected Components of upper level set)

25. 25 Changing Connectivity of ULS as Level Drops g

26. 26 ULS Connectivity Tree Schematic intensity “surface” N.B. Intensity surface is cellular (piece-wise constant), with only finitely many levels A, B, C are junction nodes where multiple zones coalesce into a single zone A B C

27. 27 A confidence set of hotspots on the ULS tree. The different connected components correspond to different hotspot loci while the nodes within a connected component correspond to different delineations of that hotspot

28. 28 Ecological Indicators and Early Warning

29. 29 Mapping Priority Hotspots of Vegetative Disturbance for Carbon Budgets

30. 30 Early Detection of Biological Invasions

31. 31 Key Crop Areas Crops NOAA Weather Threat Locations Plants Infected Non-infected Sentinel Ground Cameras Air/Space Platforms Hyperspectral Imagery Signature Library Data Processing Anomaly Report Crop Attack Decision Support System Ground Truthing Site Identification Module Signature Development Module

32. 32 Crop Biosurveillance/Biosecurity

33. 33 Hyperspectral Imagery Signature Library Image Segmentation (hyperclustering) Proxy Signal (per segment) Disease Signature Similarity Index (per segment) Tessellation (segmentation) of raster grid Signature Similarity Map Hotspot/ Anomaly Detection Crop Biosurveillance/Biosecurity Data Processing Module

34. 34 Space-Time Poverty Hotspot Typology Federal Anti-Poverty Programs have had little success in eradicating pockets of persistent poverty Federal Anti-Poverty Programs have had little success in eradicating pockets of persistent poverty Can spatial-temporal patterns of poverty hotspots provide clues to the causes of poverty and lead to improved location- specific anti-poverty policy ? Can spatial-temporal patterns of poverty hotspots provide clues to the causes of poverty and lead to improved location- specific anti-poverty policy ?

35. 35 Dimensions of Tract Poverty in Four Metropolitan Areas 1970-1990 Concentrated Persistent Camden, NJ Detroit, MI Oakland, CA Memphis, TN Growing Shifting

36. 36 Growing poverty

37. 37 Persistent poverty

38. 38 Oakland 1970 Poverty dataOakland 1980 Poverty data Oakland 1990 Poverty data Shifting poverty

39. 39 Camden 1990 Poverty data Concentrated poverty Camden 1970 Poverty dataCamden 1980 Poverty data

40. 40 Hotspot Persistence Hotspot Persistence Space (census tract) 1970 1980 1990 2000 Time (census year) Persistent Hotspot Long Duration Space (census tract) 1970 1980 1990 2000 Time (census year) One-shot Hotspot Short Duration Persistence is a property of space-time hotspots Persistence can be assessed by the projection of the space-time hotspot onto the time axis

41. 41 Typology of Persistent Space-Time Hotspots-1 Space (census tract) 1970 1980 1990 2000 Time (census year) Stationary Hotspot Space (census tract) 1970 1980 1990 2000 Time (census year) Shifting Hotspot Space (census tract) 1970 1980 1990 2000 Time (census year) Expanding Hotspot Space (census tract) 1970 1980 1990 2000 Time (census year) Contracting Hotspot

42. 42 Typology of Persistent Space-Time Hotspots-2 Space (census tract) 1970 1980 1990 2000 Time (census year) Bifurcating Hotspot These hotspots are connected in space-time However, certain time slices of the hotspot are disconnected in space Space (census tract) 1970 1980 1990 2000 Time (census year) Merging Hotspot Spatially disconnected time slice

43. 43 Trajectory of a Persistent Space-Time Hotspot A space-time hotspot is a three-dimensional object Visualization can be done by displaying the sequence of time slices---called the trajectory of the hotspot Time slices of space-time hotspot Space (census tract) 1970 1980 1990 2000 Time (census year) Merging Hotspot

44. 44 Trajectory of a Merging Hotspot 19701980 1990 2000

45. 45 Trajectory of a Shifting Hotspot 19701980 1990 2000

46. 46 Network-Based Surveillance Subway system surveillance Subway system surveillance Drinking water distribution system surveillance Drinking water distribution system surveillance Stream and river system surveillance Stream and river system surveillance Postal System Surveillance Postal System Surveillance Road transport surveillance Road transport surveillance Syndromic Surveillance Syndromic Surveillance

47. 47 NYC Drinking Water Quality Within-City Sampling Stations 892 sampling stations Each station about 4.5 feet high and draws water from a nearby water main Sampling frequency increased after 9-11 Currently, about 47,000 water samples analyzed annually Parameters analyzed:  Bacteria  Chlorine levels  pH  Inorganic and organic pollutants  Color, turbidity, odor  Many others

48. 48 Sampling Station Locations City-WideManhattan

49. 49 Crisis-Index Surveillance

50. 50 Scalable Wireless Geo-Telemetry with Miniature Smart Sensors Geo-telemetry enabled sensor nodes deployed by a UAV into a wireless ad hoc mesh network: Transmitting data and coordinates to TASS and GIS support systems

51. 51 Wireless Sensor Networks for Habitat Monitoring

52. 52 (left) The overall procedure, leading from admissions records to the crisis index for a hospital. The hotspot detection algorithm is then applied to the crisis index values defined over the hospital network. (right) The -machine procedure for converting an event stream into a parse tree and finally into a probabilistic finite state automaton (PFSA). Syndromic Surveillance

53. 53 Ocean SAmpling MObile Network OSAMON

54. 54 Tsunami Inundation Mapping

55. 55 On PULSE: The Progressive Upper Level Set Scan Statistic System for Geospatial and Spatiotemporal Hotspot Detection Ganapati Patil Department of Statistics Penn State University University Park, PA 16802 Email: gpp@stat.psu.edu Luiz Duczmal Department of Statistics Universidade Federal de Minas Gerais, Brazil Email: duczmal@est.ufmg.br Murali Haran Department of Statistics Penn State University University Park, PA 16802 Email: muh10@psu.edu Pushkar Patankar Computer Sci. & Engr. Penn State University University Park, PA 16802 Email: pushkar@psu.edu

56. 56 PULSE: Abstract In this paper, we attempt to improve on the upper level set (ULS) scan system by improving the nodal quality and quantity of the upper level set tree, using PULSE, the progressive upper level set scan system. Here we generalize the ULS notion of the rank of a cell, while the original ULS algorithm ranked cells only by a single response variable, typically disease rate, we now propose ranking cells by other criteria as well, such as their likelihood values. The comparative performance of PULSE is investigated with circular scan and with genetic algorithmic scan.

57. 57 Hotspot Detection for Continuous Responses Human Health Context:  Blood pressure levels for spatial variation in hypertension  Cancer survival (censoring issues) Environmental Context:  Landscape metrics such as forest cover, fragmentation, etc.  Pollutant loadings  Animal abundance

58. 58 Hotspot Model for Continuous Responses Simplest distributional model: Additivity with respect to the index parameter k suggests that we model k as proportional to size: Scale parameter  takes one value inside Z and another outside Z Other distribution models (e.g., lognormal) are possible but are computationally more complex and applicable to only a single spatial scale

59. 59 The upper level scan statistic (ULS), its theory, design and extension to bivariate observations are discussed. The general theory of bivariate hotspot detection is explained, including the bivariate binomial model, the multivariate exceedance approach, and the bivariate Poisson distribution. We propose the intersection method that is simple to implement, using univariate hotspot detection methods. We study the sensitivity of the joint hotspots to the degree of association between the variables. An application for the mapping of crime hotspots in the counties of the state of Ohio is presented. Hotspot Detection with Bivariate Data Reza Modarres and G.P. Patil George Washington U. and Penn State U.

60. 60 Incorporating Spatial Autocorrelation Covariate Adjustments with known/unknown effects SAR Models STAR Models CAR Models Bayesian Hierarchical Analysis

61. 61

62. 62

63. 63 Spatiotemporal Geoinformatic Disease Surveillance S.L. Rathbun and G.P. Patil University of Georgia and Penn State University

64. 64 Disease surveillance for early warning requires quick and efficient methods for delineation of disease hotspots in space and time. The upper level set (ULS) scan statistic is a computationally efficient approach for delineating hotspots of arbitrary shape. However, the current version of the ULS scan statistic assumes that the data are independently distributed, an assumption that may be untenable for georeferenced data. We shall investigate the statistical properties of the ULS scan statistic under a variety of models for spatial dependence including spatial probit models for the spatial distribution of hotspots and generalized linear mixed models with conditional autoregressive random effects. Simple methods are sought for testing the significance of hotspots, adjusting for the effects of spatial dependence. Our approach is illustrated using zoonotic disease data.

65. 65 We also present a prioritization innovation. It lies in the ability for prioritization and ranking of hotspots based on multiple indicator and stakeholder criteria without having to integrate indicators into an index, using Hasse diagrams and partial order sets. This leads us to early warning systems, and also to the selection of investigational areas. Prioritization Innovation Partial Order Set Ranking

66. 66 HUMAN ENVIRONMENT INTERFACE LAND, AIR, WATER INDICATORS RANK COUNTRYLANDAIRWATER 1Sweden 2Finland 3Norway 5 Iceland 13 Austria 22 Switzerland 39 Spain 45 France 47 Germany 51 Portugal 52 Italy 59 Greece 61 Belgium 64 Netherlands 77 Denmark 78 United Kingdom 81 Ireland 69.01 76.46 27.38 1.79 40.57 30.17 32.63 28.34 32.56 34.62 23.35 21.59 21.84 19.43 9.83 12.64 9.25 35.24 19.05 63.98 80.25 29.85 28.10 7.74 6.50 2.10 14.29 6.89 3.20 0.00 1.07 5.04 1.13 1.99 100 98 100 82 100 98 100 for land - % of undomesticated land, i.e., total land area-domesticated (permanent crops and pastures, built up areas, roads, etc.) for air - % of renewable energy resources, i.e., hydro, solar, wind, geothermal for water - % of population with access to safe drinking water

67. 67 Hasse Diagram (Western Europe)

68. 68 Ranking Partially Ordered Sets – 5 Linear extension decision tree a b dc e f a c e b b d ff d ed f e e f c f d ed f e e f d f e e f c f e e f c c f d ed f e e f d f e e f c b a b a d Jump Size: 1 3 3 2 3 5 4 3 3 2 4 3 4 4 2 2 Poset (Hasse Diagram)

69. 69 Cumulative Rank Frequency Operator – 5 An Example of the Procedure In the example from the preceding slide, there are a total of 16 linear extensions, giving the following cumulative frequency table. Rank Element123456 a91416161616 b71215161616 c0410161616 d026121616 e00141016 f0000616 Each entry gives the number of linear extensions in which the element (row label) receives a rank equal to or better that the column heading

70. 70 Cumulative Rank Frequency Operator – 6 An Example of the Procedure 16 The curves are stacked one above the other and the result is a linear ordering of the elements: a > b > c > d > e > f

71. 71 Rank range run sequence for the 4-index data set. The bottom of each vertical line represents the minimum rank and the top of the line is the maximum rank for the indicators.

72. 72 Maximality level vs. consistency level for the 4-index data set

73. 73 End-member Elimination results for the 4-index data set

74. 74 First stage screening First stage screening –Significant clusters by SaTScan and/or upper level sets upper level sets Second stage screening Second stage screening –Multicriteria noteworthy clusters by partially ordered sets and Hass diagrams Final stage screening Final stage screening –Follow up clusters for etiology, intervention based on multiple criteria using Hass diagrams based on multiple criteria using Hass diagrams Multiple Criteria Analysis, Multiple Indicators and Choices, Health Statistics, Disease Etiology, Health Policy, Resource Allocation

75. 75 Logo for Statistics, Ecology, Environment, and Society