Density Estimation Converts points to a raster

Slides:



Advertisements
Similar presentations
Original Figures for "Molecular Classification of Cancer: Class Discovery and Class Prediction by Gene Expression Monitoring"
Advertisements

Spatial Autocorrelation using GIS
Spatial statistics Lecture 3.
Basic geostatistics Austin Troy.
Spatial Statistics II RESM 575 Spring 2010 Lecture 8.
Introduction to ArcGIS Geostatistical Analyst & Fragstats Represent the Data Explore the Data Fit the Model Perform Diagnostics Compare Models Classify.
Chap 3-1 EF 507 QUANTITATIVE METHODS FOR ECONOMICS AND FINANCE FALL 2008 Chapter 3 Describing Data: Numerical.
Geographic Information Systems
Spatial Analysis Longley et al., Ch 14,15. Transformations Buffering (Point, Line, Area) Point-in-polygon Polygon Overlay Spatial Interpolation –Theissen.
Mahanalobis Distance Dr. A.K.M. Saiful Islam Source:
Ordinary Kriging Process in ArcGIS
Density Estimation Converts points to a raster The density of points in the neighborhood of a pixel No “Z” value is used ArcMap has a simple “Point Density”
Advanced GIS Using ESRI ArcGIS 9.3 Arc ToolBox 5 (Spatial Statistics)
Chap 3-1 Statistics for Business and Economics, 6e © 2007 Pearson Education, Inc. Chapter 3 Describing Data: Numerical Statistics for Business and Economics.
Fundamentals of GIS Lecture Materials by Austin Troy except where noted © 2008 Lecture 14: More Raster and Surface Analysis in Spatial Analyst Using.
Density vs Hot Spot Analysis. Density Density analysis takes known quantities of some phenomenon and spreads them across the landscape based on the quantity.
Habitat Analysis in ArcGIS Use of Spatial Analysis to characterize used resources Thomas Bonnot
University of Wisconsin-Milwaukee Geographic Information Science Geography 625 Intermediate Geographic Information Science Instructor: Changshan Wu Department.
1 Doing Statistics for Business Doing Statistics for Business Data, Inference, and Decision Making Marilyn K. Pelosi Theresa M. Sandifer Chapter 11 Regression.
ArcGIS Overview Lecture 1: Software Layer characteristics Thematic maps.
ESRM 250 & CFR 520: Introduction to GIS © Phil Hurvitz, KEEP THIS TEXT BOX this slide includes some ESRI fonts. when you save this presentation,
Esri International User Conference | San Diego, CA Technical Workshops | Spatial Statistics: Best Practices Lauren Rosenshein, MS Lauren M. Scott, PhD.
Using ESRI ArcGIS 9.3 Spatial Analyst
Basic Geographic Concepts GEOG 370 Instructor: Christine Erlien.
Why Is It There? Getting Started with Geographic Information Systems Chapter 6.
15. Descriptive Summary, Design, and Inference. Outline Data mining Descriptive summaries Optimization Hypothesis testing.
NR505 GIS Applications in Wildlife Sciences
Intro to Raster GIS GTECH361 Lecture 11. CELL ROW COLUMN.
Data Types Entities and fields can be transformed to the other type Vectors compared to rasters.
Statistical analysis Outline that error bars are a graphical representation of the variability of data. The knowledge that any individual measurement.
Ripley K – Fisher et al.. Ripley K - Issues Assumes the process is homogeneous (stationary random field). Ripley K was is very sensitive to study area.
Spatial Interpolation Chapter 13. Introduction Land surface in Chapter 13 Land surface in Chapter 13 Also a non-existing surface, but visualized as a.
Local Indicators of Spatial Autocorrelation (LISA) Autocorrelation Distance.
Concepts and Applications of Kriging
NR 143 Study Overview: part 1 By Austin Troy University of Vermont Using GIS-- Introduction to GIS.
What’s the Point? Working with 0-D Spatial Data in ArcGIS
1 Overview Importing data from generic raster files Creating surfaces from point samples Mapping contours Calculating summary attributes for polygon features.
L15 – Spatial Interpolation – Part 1 Chapter 12. INTERPOLATION Procedure to predict values of attributes at unsampled points Why? Can’t measure all locations:
Chapter1: Introduction Chapter2: Overview of Supervised Learning
Phil Hurvitz Avian Conservation Lab Meeting 8. March. 2002
Spatial Analysis with Raster Datasets-1 Francisco Olivera, Ph.D., P.E. Srikanth Koka Department of Civil Engineering Texas A&M University.
INTERPOLATION Procedure to predict values of attributes at unsampled points within the region sampled Why?Examples: -Can not measure all locations: - temperature.
Spatial statistics Lecture 3 2/4/2008. What are spatial statistics Not like traditional, a-spatial or non-spatial statistics But specific methods that.
Descriptive Statistics ( )
Chapter 13 Simple Linear Regression
Statistical analysis.
Chapter 8 Raster Analysis.
Business and Economics 6th Edition
15. Descriptive Summary, Design, and Inference
Introduction to Spatial Statistical Analysis
Spatial analysis Measurements - Points: centroid, clustering, density
Statistical analysis.
Raster Analysis Ming-Chun Lee.
Population rrmtt.wikispaces.com
Summary of Prev. Lecture
Nonparametric Density Estimation
Spatial statistics Topic 4 2/2/2007.
Lecture 6 Implementing Spatial Analysis
Chapter 10 Problems Even problems are at end of text. 19. What is a kernel in a moving window operation? Does the kernel size or shape change for different.
Review- vector analyses
Tabulations and Statistics
Preparing raster files for Condatis
URBDP 422 Urban and Regional Geo-Spatial Analysis
Concepts and Applications of Kriging
Nicholas A. Procopio, Ph.D, GISP
Concepts and Applications of Kriging
Nicholas A. Procopio, Ph.D, GISP
Business and Economics 7th Edition
4. Focal Analysis 4.1 Definition of focal analysis
Spatial Analysis with Raster Datasets-1
Presentation transcript:

Density Estimation Converts points to a raster The density of points in the neighborhood of a pixel No “Z” value is used ArcMap has a simple “Point Density” tool Each pixel=number of points within radius Kernel Density is related to Kernel Smoothing but different

Point Density in ArcMap Distance=0.3 Distance=3

Point Density in ArcMap Distance=10

Kernel Smoothing Kernel Smoothing is interpolation

Density Estimation Using Kernels Creates a raster from points Weight (attribute) optional Not really interpolation “Kernel function” applied to points near target pixel Different functions are available High parameters make a “wide” pile, small values make a “narrow” pile Density analysis takes known quantities of some phenomena and spreads it across the landscape based on the quantity that is measured at each location and the spatial relationship of the locations of the measured quantities.Why map density? Density surfaces are good for showing where point or line features are concentrated. For example, you might have a point value for each town representing the total number of people in the town, but you want to learn more about the spread of population over the region. Since all the people in each town do not live at the population point, by calculating density you can create a surface showing the predicted distribution of the population throughout the landscape. F(x) = 1/nh2 S K x-X/n A density function in GIS calculates a magnitude per unit area from point features that fall within a neighborhood around each cell. Conceptually, a neighborhood is defined around each raster cell center, and the number of points that fall within the neighborhood is totaled and divided by the area of the neighborhood.If a population field setting other than NONE is used, the item's value determines the number of times to count the point. For example, an item with a value of 3 would cause the point to be counted as three points. The values can be integer or floating point. If an area unit is selected, the calculated density for the cell is multiplied by the appropriate factor before it is written to the output raster. For example, if the input ground units are meters, comparing a unit scale factor of meters to kilometers will result in the values being different by a multiplier of 1,000,000 (1000 x 1000).Possible uses include finding density of houses, wildlife observations, or crime reports. The population field can be used to weigh some points more heavily than others, depending on their meaning, or to allow one point to represent several observations. For example, one address might represent a condominium with 6 units, or some crimes might be weighed more severely than others in determining overall crime levels.Increasing the radius will not greatly change the calculated density values. Although more points will fall inside the larger neighborhood, this number will be divided by a larger area when calculating density. The main effect of a larger radius is that density is calculated considering a larger number of points, which can be farther from the raster cell. This results in a more generalized output raster. How to calculate density in ArcGIS 9.1with Spatial Analyst: 1 Click the Spatial Analyst dropdown arrow and click Density. 2 Click the Input data dropdown arrow and click the input layer. 3 Click the Population field dropdown arrow and click the field you want to use. 4 Click either Kernel or Simple Density type. 5 In the Search radius text box, type a value to determine the distance to search for points or lines from each cell in the output raster. 6 Click the Area units dropdown arrow and choose the units in which the density values should be presented. 7 Specify an output cell size. 8 Type a name for the result, or leave the default to create a temporary result. 9 Click OK. More info at http://webhelp.esri.com/arcgisdesktop/9.2

Width of Kernel Determines smoothness of surface narrow kernels produce bumpy surfaces wide kernels produce smooth surfaces

Kernel Density in ArcGIS 10 Under Spatial Analyst -> Kernel Density The kernel function is based on the quadratic kernel function described in Silverman (1986, p. 76, equation 4.5). Kernel estimators are based on probability “kernels”, which are regions around each point location containing some likelihood of animal presence. The width of the kernel is based on the smoothing parameter (h), which can be determined in a number of different ways. The main automated methods for smoothing parameter selection are reference (which is based on assumptions of bivariate normality) and least-squares cross validation (which is based on properties of the data). Kernel methods are either adaptive (where the kernel width increases as the distance between kernels increases) or fixed (always the same kernel width). Adaptive kernels tend to perform poorly, often over-estimating home range areas (Powell 2000; Kernohan et al. 2001). Kernel estimators have a number of features that make them useful for home ranges: they work well with small amounts of data (approximately 50 locations), they are robust to autocorrelation, they are nonparametric, they allow multiple centers of activity, and they result in a utilization distribution (UD) rather than a simple home range outline (Kernohan et al. 2001). A UD is a grid where the value for each cell represents the probability of the animal occurring in that cell. Among other uses, a UD allows for a more precise estimate of home range overlap than a simple outline.

Overview This analysis show where point features are concentrated. Estimations are based on probability “kernels” regions around each point location containing some likelihood of point presence. The width of the kernel is based on the smoothing parameter (h) The output is often called a Utilization Distribution (UD) Grid. Methods include: minimum convex polygons, bivariate ellipses, adaptive and fixed kernels Kernel estimators are based on probability “kernels”, which are regions around each point location containing some likelihood of animal presence. The width of the kernel is based on the smoothing parameter (h), which can be determined in a number of different ways. The main automated methods for smoothing parameter selection are reference (which is based on assumptions of bivariate normality) and least-squares cross validation (which is based on properties of the data). Kernel methods are either adaptive (where the kernel width increases as the distance between kernels increases) or fixed (always the same kernel width). Adaptive kernels tend to perform poorly, often over-estimating home range areas (Powell 2000; Kernohan et al. 2001). Kernel estimators have a number of features that make them useful for home ranges: they work well with small amounts of data (approximately 50 locations), they are robust to autocorrelation, they are nonparametric, they allow multiple centers of activity, and they result in a utilization distribution (UD) rather than a simple home range outline (Kernohan et al. 2001). A UD is a grid where the value for each cell represents the probability of the animal occurring in that cell. Among other uses, a UD allows for a more precise estimate of home range overlap than a simple outline.

Kernel Density in ArcGIS

Kernel Density Cell Size = 0.05 Search Radius = 0.4?

Kernel Density Cell Size = 0.05 Search Radius = 10

How to select parameters? What should the cell size be? What should the search radius be?

Origins of Computer Viruses

Origins of Email Spam

Kernel Density Analysis Amelia O’Connor

Kernel Density Output

Other tool extensions for kernel density: Home Range Tools Animal Movement Biotas Home Ranger 1.5 KernelHR

Spatial Stats Toolbox New in ArcGIS 10 Additional tools in ArcGIS 10.2 By Lauren Rosenshein

Hot-Spot Analysis Layer may show “hot-spot” but is it really? Z-score and P-value are required Z-score = high or low values together? P-value = random?

Hot-Spot Analysis High z-values indicate a significantly high or low value 2.5=cluster of high or low values P-value is the chance a pattern is random 0.01=probably not random

Hot-Spot Analysis Tool

Citations Bugoni, L., D'Alba, L., and Furness, R. W. (2009) Marine habitat use of wintering spectacled petrels Procellaria conspicillata, and overlap with longline fishery. Marine Ecology Progress Series 374:273-285. Mitchell, Brian R. (2007) Comparison of Programs for Fixed Kernel Home Range Analysis http://www.wildlife.org/wg/gis/newsletter/jun06/hrcompar.htm Silverman, B. W. Density Estimation for Statistics and Data Analysis. New York: Chapman and Hall, 1986. ArcGIS 10 resource center; Kernel Density (Spatial Analyst) http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//009z0000000s000000.htm http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#/Understanding_density_analysis/009z0000000w000000/ http://help.arcgis.com/en/arcgisdesktop/10.0/help/index.html#//009z00000011000000.htm

Extra Slides

Density Estimation Simple point density: Golf courses Fail Rockware