Representativity of the Iowa Environmental Mesonet Daryl Herzmann and Jeff Wolt, Department of Agronomy, Iowa State University The Iowa Environmental Mesonet.

Slides:



Advertisements
Similar presentations
Sampling Design, Spatial Allocation, and Proposed Analyses Don Stevens Department of Statistics Oregon State University.
Advertisements

Very simple to create with each dot representing a data value. Best for non continuous data but can be made for and quantitative data 2004 US Womens Soccer.
I OWA S TATE U NIVERSITY Department of Animal Science Using Basic Graphical and Statistical Procedures (Chapter in the 8 Little SAS Book) Animal Science.
Copyright © Allyn & Bacon (2007) Statistical Analysis of Data Graziano and Raulin Research Methods: Chapter 5 This multimedia product and its contents.
Spatial statistics Lecture 3.
Spatial Statistics II RESM 575 Spring 2010 Lecture 8.
Calculating & Reporting Healthcare Statistics
Statistical Analysis SC504/HS927 Spring Term 2008 Week 17 (25th January 2008): Analysing data.
Biostatistics Unit 2 Descriptive Biostatistics 1.
Sampling Distributions
Point Pattern Analysis
University of Wisconsin-Milwaukee Geographic Information Science Geography 625 Intermediate Geographic Information Science Instructor: Changshan Wu Department.
Describing Data: Numerical
6 - 1 Basic Univariate Statistics Chapter Basic Statistics A statistic is a number, computed from sample data, such as a mean or variance. The.
Plug & Play Middle School Common Core Statistics and Probability using TinkerPlots.
1 Statistical Analysis - Graphical Techniques Dr. Jerrell T. Stracener, SAE Fellow Leadership in Engineering EMIS 7370/5370 STAT 5340 : PROBABILITY AND.
May 06th, Chapter - 7 INFORMATION PRESENTATION 7.1 Statistical analysis 7.2 Presentation of data 7.3 Averages 7.4 Index numbers 7.5 Dispersion from.
Chapter 3 Statistical Concepts.
Chapter 3 – Descriptive Statistics
Chapter 3: Central Tendency. Central Tendency In general terms, central tendency is a statistical measure that determines a single value that accurately.
Spatial Statistics Applied to point data.
McGraw-Hill/IrwinCopyright © 2009 by The McGraw-Hill Companies, Inc. All Rights Reserved. Chapter 7 Sampling Distributions.
McGraw-Hill/Irwin Copyright © 2007 by The McGraw-Hill Companies, Inc. All rights reserved. Chapter 6 Sampling Distributions.
Chapter 15 Data Analysis: Testing for Significant Differences.
Variability The goal for variability is to obtain a measure of how spread out the scores are in a distribution. A measure of variability usually accompanies.
Measures of Central Tendency and Dispersion Preferred measures of central location & dispersion DispersionCentral locationType of Distribution SDMeanNormal.
Basic Statistics for Engineers. Collection, presentation, interpretation and decision making. Prof. Dudley S. Finch.
Describing Data: Displaying and Exploring Data. 4-2 GOALS 1. Develop and interpret a dot plot. 2. Construct and interpret box plots. 3. Compute and understand.
What is Central Tendency? Statistics analyzes and interprets large sets of numbers. To make the lists of data more comprehensible, central tendencies are.
Spatial Association Defining the relationship between two variables.
Using the Mesonet Daryl 30 March 2015.
1.1 EXPLORING STATISTICAL QUESTIONS Unit 1 Data Displays and Number Systems.
Biostatistics Class 1 1/25/2000 Introduction Descriptive Statistics.
Sampling Populations Ideal situation - Perfect knowledge Not possible in many cases - Size & cost Not necessary - appropriate subset  adequate estimates.
TYPES OF STATISTICAL METHODS USED IN PSYCHOLOGY Statistics.
PCB 3043L - General Ecology Data Analysis. OUTLINE Organizing an ecological study Basic sampling terminology Statistical analysis of data –Why use statistics?
The Central Tendency is the center of the distribution of a data set. You can think of this value as where the middle of a distribution lies. Measure.
Categorical vs. Quantitative…
Distributions of the Sample Mean
Chapter 7 Sampling Distributions Statistics for Business (Env) 1.
Experimental Research Methods in Language Learning Chapter 9 Descriptive Statistics.
Extent and Mask Extent of original data Extent of analysis area Mask – areas of interest Remember all rasters are rectangles.
1 Descriptive statistics: Measures of dispersion Mary Christopoulou Practical Psychology 1 Lecture 3.
Central Tendency & Dispersion
Sociology 5811: Lecture 3: Measures of Central Tendency and Dispersion Copyright © 2005 by Evan Schofer Do not copy or distribute without permission.
What’s the Point? Working with 0-D Spatial Data in ArcGIS
So, what’s the “point” to all of this?….
Chapter Eight: Using Statistics to Answer Questions.
PCB 3043L - General Ecology Data Analysis. PCB 3043L - General Ecology Data Analysis.
PCB 3043L - General Ecology Data Analysis.
Business Statistics, 4e, by Ken Black. © 2003 John Wiley & Sons. 3-1 Business Statistics, 4e by Ken Black Chapter 3 Descriptive Statistics.
Technical Details of Network Assessment Methodology: Concentration Estimation Uncertainty Area of Station Sampling Zone Population in Station Sampling.
Measurements and Their Analysis. Introduction Note that in this chapter, we are talking about multiple measurements of the same quantity Numerical analysis.
166-7: Modeling Out Crossing Probabilities for Maize in Iowa Daryl Herzmann Jeff Wolt Raymond Arritt Mark Westgate Susana Goggi Iowa State University.
Chapter 6: Descriptive Statistics. Learning Objectives Describe statistical measures used in descriptive statistics Compute measures of central tendency.
Rain gauge Lathiya milan. Methods of Measuring Rainfall: Remote Tipping bucket rain gauge -The bucket tips when precipitation of 0.2 mm,
Technical Details of Network Assessment Methodology: Concentration Estimation Uncertainty Area of Station Sampling Zone Population in Station Sampling.
Measures of Dispersion Measures of Variability
Spatial statistics Lecture 3 2/4/2008. What are spatial statistics Not like traditional, a-spatial or non-spatial statistics But specific methods that.
Exploratory Data Analysis
Regression Analysis.
PCB 3043L - General Ecology Data Analysis.
Daryl Herzmann and Raymond Arritt
Numerical Measures: Centrality and Variability
Descriptive Statistics
Spatial statistics Topic 4 2/2/2007.
Lecture 6 Implementing Spatial Analysis
Basic Statistical Terms
Basic Training for Statistical Process Control
Basic Training for Statistical Process Control
Presentation transcript:

Representativity of the Iowa Environmental Mesonet Daryl Herzmann and Jeff Wolt, Department of Agronomy, Iowa State University The Iowa Environmental Mesonet (IEM) collects data from disparate sources. Combining these data sources into accurate products requires an understanding of the spatial characteristics of the observation network. This work investigates the spatial representivity of the IEM following methodology described by: Dubois, G., 2000: How representative are samples in a sampling network? J Geographic Info Decision Anal, 4, Motivation Individual NetworksNetwork Groups Procedure The IEM’s component networks were analyzed individually and then grouped to investigate the statistical benefits of combining these networks. Computations were completed using spatial capabilities of ESRI’s ArcGIS, PostGIS relational database, and Python software. The NAD83 UTM Zone 15 North map projection was used to calculate distances between observation points. Preliminary Results & Future Work The clustering of stations near urban areas and impacts of the dense SchoolNet negatively impact the spatial statistics of representativity for Iowa. This ‘problem’ is clearly illustrated in the CR analysis as the “rich get richer” and the “poor get poorer” with the addition of tier 2 and tier 3 observing networks. Fundamentally, the ability to resolve statewide phenomena will be limited by the undersampled areas of Eastern Iowa. Future work will include the examination of sub-areas of the state to ascertain the resolvable spatial scales. Coefficient of Representativity The Coefficient of Representativity (CR) was proposed by Dubois (2000) as a measure to combine the Thiessen polygons and the distance to the nearest neighbour (NNI). The CR is a product of two terms, the first (A) is a simple ratio of the area of the Thiessen polygon (S Th ) to the mean Thiessen polygon (S m ): total area divided by number of sampling points; the second (B) is a ratio between the squared distance between nearest neighbors and the same mean Thiessen polygon. Tier 1 clearly shows the under sampled (values less than 1) portions of the state, which are NE Iowa and parts of SW Iowa. Adding the RWIS network in tier 2 increases localized clustering at the expense of other areas where observations become even more sparse. Tier 3 shows the affect of adding 60+ SchoolNet sites in Central Iowa, since the clustering goes up and the sparse areas get even sparser due to the increased number of stations. Thiessen Polygons Thiessen polygons have been a traditional staple of the spatial statistician. Each constructed polygon contains exactly one measurement point while having the property of all points within the polygon being closer to this measurement point than any other point. Isolated measurements will therefore have larger polygons than clustered measurements. Histograms of the area covered by these polygons provide insight into network homogeneity. Areas in a network that are under sampled can quickly be seen by looking for the largest polygons. The largest polygons can be seen in Northeast and Southern Iowa. The histogram for Tier 1 also nicely details the collection of smaller polygons around 2,000 km**2 along with a second mode of larger polygons around 5,000 km**2. Unfortunately, this bi-modal distribution is not greatly helped by the addition of the RWIS sensors in Tier 2. The clustering of the RWIS sensors simply splits some of the larger polygons, but a couple of rather large polygons in Northeastern Iowa still exist. For Tier 3, the clustering of sites in Central Iowa with the SchoolNet network creates a large number of smaller polygons and doesn't help to eliminate the larger polygons in Eastern Iowa. The average polygon size on the maps are only interesting in the fact that they indicate what the ideal size would be if all of the sites were spread out equally. Nearest Neighbor Index The NNI is a statistic comparing the mean minimum distances between sampling points to those that are expected by chance over some sampling area. Index values greater than 1 indicate dispersion while values less than 1 indicate clustering. The ideal value is 1. The significance of the departure from unity is represented by the Z statistic. Values of Z greater than 2 indicate that the network is not randomly oriented in space (clustering or dispersion exists). The NNI for the ASOS network is a relatively high value (1.58) which indicates dispersion of the sampling points. This spread was one of the main reasons the IEM was created, since the baseline ASOS network is poorly representative at even a 50 km scale over the state. The NNI value of 0.95 for the combination of the AWOS and RWIS networks to the ASOS network (Tier 2) shows the benefit of the IEM collaboration. The NNI for the Tier 2 network (0.95) is very close to unity and indicates that the network has a nearly uniform distribution. The addition on the SchoolNet and ISUAG networks into the Tier 3 has no effect on the NNI. Results Interpretation of NNI Morisita Index The Morisita index investigates how much clustering occurs when a sampling network is broken up into regular cells. If each cell has the same number of observation points inside of it, then the index should indicate uniformity in space at that scale. The equation is as follows: where Q is the total number of cells, n i is the number of samples in i th cell, and N is the total number of sampling points. This chart indicates the scales at which clustering (values larger than 1) are ocurring. At the smallest scales of km, clustering is evident in the Tier 2 and Tier 3 networks. This makes practical sense considering the co- placement of RWIS sites near cities with ASOS/AWOS sites in Tier 2 and the overall clustering of SchoolNet sites in Tier 3. An illustration of the Morisita Index An observation network (left) is broken up into grid cells of equal size. Cells which have more than one observation (clustering) will increase the index value. This research is funded by the USDA Biotechnology Research Assessment Grants Program