Guidebook for GIS Spatial Analysis Practice
Guidebook
For
GIS Spatial Analysis Practice
Course Title: Spatial Analysis Using GIS
Applies to GIS Specialty in Semester 2 of Grade 3
Course Author: LIN Guangfa
Total Period in Hour: 40
Practical Period in Hour: 20
Credit Hour: 2
Compiled by: LIN Guangfa & YANG Liping
Department of Geo-Informatics, School of Geography, FJNU
May 2007
60
Guidebook for GIS Spatial Analysis Practice
Past experience indicates that those who do not attend both Lectures & Practices are likely to have problems.
60
Guidebook for GIS Spatial Analysis Practice
Table of Contents
Part One: General Requirements for a Practice 4
1. Be Ready for a Practice 4
2. Notice 4
Part Two: Introduction of Each Practices 5
Practice 1 Spatial Interpolation Using ArcGIS 6
Practice 2 Spatial Analysis with Raster Data 21
Practice 3 The Shortest Path Analysis 34
Practice 4 Spatial Analysis with Vector Data 43
Practice 5 Surface Modeling of Human Population Using ArcGIS 51
Part Three: Report Structure 58
1. Contents of a Report 58
2. Format of a Report 59
60
Guidebook for GIS Spatial Analysis Practice
Part One: General Requirements for a Practice
1. Be Ready for a Practice
1. Please read the corresponding sections of the guidebook BEFORE you come into the lab.
2. Be definitude to the aims, requirements and contents of an experiment. It is each student’s responsibility to download and read the lecture notes BEFORE the lab practice starts. The lab conduct assumes you have understood all the corresponding terms, principles and methods about the experiment.
- It is highly recommended that you also download the dataset and debug the software need for the practice BEFORE the experiments.
2. Notice
1. Pay attention to the demonstration and explanation of your teacher, and do stand by the disciplines of the class. Clean up your table and shut off the computer after the experiment.
2. The tips in the guidebook are giving you some reference points only, there are not military orders for you all. So, if something strange happened to you during an experiment, please try to solve them yourself first, then turn to your classmates or teachers. Note down the phenomena and result of the experiment in detail.
3. Don’t delete the files in the system partition. Keep in mind that not to save your data on the partition C and D of the hard disk that is protected by the restore card.
4. Please collate and analyze the notes and data of practice, and hand up your report in time. Or you will forget the process of the experiment soon.
5. The file of your practice report should be named in format as experiment number plus your student identity or group number, such as “Practice 1 of 109042004001” or “Practice 5 of Group 1”.
6. A seminar class will be provided after each practice. Students are grouped to prepare for the class with a power point file. All of you are encouraged to report your results, discussions and conclusions and take part in the Q & A session.
Part Two: Introduction of Each Practices
PRACTICE 1 SPATIAL INTERPOLATION USING ARCGIS
PRACTICE 2 SPATIAL ANALYSES WITH RASTER DATA
PRACTICE 3 THE SHORTEST PATH ANALYSIS
PRACTICE 4 SPATIAL ANALYSES WITH VECTOR DATA
PRACTICE 5 SURFACES MODELING OF HUMAN
POPULATION USING ARCGIS
60
Guidebook for GIS Spatial Analysis Practice
Practice 1 Spatial Interpolation Using ArcGIS
1. Aims
By the end of this practical you should:
l Understand the basic principles of common used spatial interpolation methods include Inverse Distance Weighted (IDW) and Kriging.
l Be able to use the tools in ESDA to find out the distribution feature of datasets.
l Understand the meaning of each parameter in the operation process; find out the relationship between parameters and the result of the interpolation.
l And, know how to apply these methods to solve a practical question with ArcGIS software step by step.
2. Data Requirements
A dataset of 1413 points with elevation in Xiamen city will be provided in SHP format.
3. Environment of Experiment
Hardware: Computers
Software: Arcview3.3, ArcGIS9.0
4. Contents of the Experiment
l Using IDW and Kriging method to create surface of terrain in Xiamen by interpolating the given 1413 points with elevation.
l Adjusting the parameters in the methods and software employed to find out how these parameters are working.
l Comparing these two methods in detail.
5. Tips of Practical Exercise
(1) IDW Spatial Interpolation Method in Spatial Analyst Module of ArcGIS9.0
Open package ArcGIS9.0 à Add elevation points data: xiamendao.shp à Add Spatial Analyst module, as Fig.1-1 shows.
Fig. 1-1 Add Spatial Analyst Module
Note: after adding the Spatial Analyst module, if the toolbar is still not available, please open tools\extensions dialog as Fig.1-2 shows. Tick the checkbox before Spatial Analyst then close it.
Fig.1-2 Make Spatial Analyst Module Activated
Open Spatial Analyst à Interpolate to Raster à IDW, as Fig.1-3 shows.
Fig.1-3 Open IDW Dialog in Spatial Analyst Module
IDW parameters concrete setting showed in Fig1-4.
Input Points: the points data to be interpolated, in this practice was xiamendao.shp elevation points data;
Z value field: the field contained attributes value to be interpolated; in the practice was the GC field with elevation data of xiamendao.shp data;
Power: the exponential for interpolation, generally 2 is better. Yet, for different regions, it is necessary to try more and get an optimal exponential for its interpolation.
Distance: search radius.
Minimum number of points: if there is not enough points searched within the search radius to interpolate, the search radius will be extended until enough points were found.
Fig.1-4 IDW Related Parameters Setting
Output cell size: the cell size of the raster of interpolation result. The smaller the value is, the more precise the interpolation result is. Yet, it is unnecessary to make the value too small, because this will cause data redundancy, and the computation speed will be lowered.
IDW interpolation result
Fig.1-5 IDW Spatial Interpolation Result
As Fig.1-5 shows, the point’s elevation data has been interpolated into surface data.
IDW spatial interpolation considered break line: create a new polyline layer in ArcCatalog, and then add the new-created polyline layer into ArcMap.
Open IDW dialog, tick Use barrier polylines checkbox, browse the new-created break line data, as Fig.1-6 shows.
Fig.1-6 IDW Spatial Interpolation Parameters Setting Considering Break Line
IDW interpolation result took break line into account is showed in Fig.1-7, the red line is a break line.
Fig.1-7 IDW Interpolation Result Considering Break Line
IDW spatial interpolation method in Geostatistical Analyst module of ArcGIS9.0: Add the geostatistical Analyst module. Then open the module, click the Create Subsets option, showing in Fig.1-8.
Fig.1-8 Create Subsets
Choose the Data to Create Subsets.
Fig.1-9 Create Subsets Dialog
Divide the source elevation point data into two parts, one used to Interpolate, the other used to validate the interpolation precision, by moving the valve showed in Fig.1-10.
Fig.1-10 Create Subsets by Moving the Valve
IDW in Geostatistical Analyst Module
Fig.1-11 Geostatistical Wizard
Choose the IDW method in the test box under the Methods label first. Then set the related parameters as Fig.1-12 shows.
Fig.1-12 Choose IDW Method in Geostatistical Wizard
Set the parameter in the IDW dialog
Fig.1-13 IDW Parameters Setting
Check the validation
Fig.1-14 Cross Validation
Fig.1-15 Validation
The IDW Interpolation Result
Fig.1-16 IDW Interpolation Result
(2) IDW Spatial Interpolation Method in ArcView
Open the ArcView3.3 Software package à Add the Spatial Analyst Extensive Module followed the Route: File\extensions à Then tick the Spatial Analyst Module à Make sure to change the unit in the View Properties dialog into meters, as Fig.1-17 showing.
Fig.1-17 Change Units in View Properties
à Add practice data (xiamendao and barrier line) à Use the command: Surface\Interpolate Grid (make sure that the xiamendao layer is kept active at this moment) à set the related parameter as Fig.1-18 shows.
After input the Value of the output Grid Cell Size, mind to input an enter after entering the cell size value, then the numbers of Rows and Columns will be filled in automatically.
Fig.1-18 Parameter Setting
Then set the parameter in the dialog of Interpolate Surface
Fig.1-19 Interpolation Method Choice
The Interpolation Result without the Barrier line:
Fig.1-20 Interpolation Result Without Barrier Line
Taking barrier line into account set the barrier line parameter.
Fig.1-21 IDW Barrier Line Parameter Setting
The Interpolation Result Considered the Barrier line:
Fig.1-22 The Interpolation Result Considered the Barrier Line
(3) Interpolation Using Kriging in ArcGIS9.0
Common Steps of Kriging Interpolation: Data Preparation à Data Analysis and Transform à Applicability Appraisement à Calculate the Empirical the Variogram à Fit a Model and Make Interpolation à Judge the Result & Adjust Parameters. Firstly, you can activate the Kriging method by click the Spatial Analyst module (Fig. 1-23).
Fig. 1-23 Toolbar of Kriging in ArcGIS9.0
The most important step in the process is to find out a proper Variogram for the given interpolation question. The other parameters are similar to those in IDW. So, The following text will focus on the method of finding out a Variogram (Fig. 1-24).
Fig. 1-24 Select a Semivariogram Model for a Kriging Method
A Semivariogram model is definite by selecting the type of function and advanced parameters (Fig. 1-25).
Fig. 1-25 Advanced Parameters in Kriging Variogram
ArcGIS9.0 provides a very good tool for us to Explore Spatial Data Analysis (ESDA, Fig. 1-26) that can help use to find out the distribution feature of the sample data, which is used to select a semivariogram model.
Fig. 1-26 Tools Used for Explore Spatial Data Analysis
Because of the complexity of computing semivariogram for each sample points, ArcGIS sets a restriction that the number of sample point is not more than 300. So, you have to create a training subset of not more than 300 points from the total 1413 points randomly (Fig.1-10). And then use the subset samples to compute semivariogram (Fig. 1-27). ArcGIS provides the following functions to choose from to model the empirical semivariogram: Circular, Spherical, Tetraspherical, Pentaspherical, Exponential, Gaussian, Rational Quadratic, Hole Effect, K-Bessel, J-Bessel and Stable. Experientially, you will observe the graph of semivariogram and compare it with the graph of functions listing above and select the most similar one. And, then measure to work out the parameters major range, partial sill and nugget.
Fig. 1-27 Semivariogram Graph of the Training Samples
6. Practice Result
With the knowledge and skills learned from the above, comprehend the principles and thought of interpolation seriously. Then create DEM of the study site by using the interpolation technology.
7. Practice Report
According to the practice aims, finish the practice report. The emphases of grade assessment are the discussion on parameters and compare between the two methods of interpolation employed.
60
Guidebook for GIS Spatial Analysis Practice
Practice 2 Spatial Analysis with Raster Data
1. Aims
By the end of this practical you should be able to:
l Perform spatial analysis for raster data (DEM grid data in this practice mainly) with Hydrology module.
l Build a spatial model to extract vegetation information from the ETM+ image according to the knowledge obtained from the AOI sampling. And, then extract the forestland parcels from the image though raster spatial analysis in ArcGIS.
l Understand the principle of spatial analysis for raster data through the practice.
2. Data Requirements
Xiamen Island ETM image (2001): xmd_etm1234.img;
Xiamen Island DEM data (15m*15m): Xmd_tingrid15.
3. Environment of Experiment
Hardware: Computers;
Software: ArcGIS9.0, ERDAS8.7.
4. Contents and Tips of the Practical Exercise
4.1 The Main Contents for this Practice
(1) Extract river and stream network and perform rainfall flow direction analysis with Hydrology module.
(2) Extraction forest information based on knowledge (Assume that the part of vegetation, with which the DEM value is greater than 150m, is the forest.)
4.2 Main Steps for the First Content.
(1) Build None-sink DEM
Open Hydrology module in ArcMap: Open Tools/Customize in ArcMap, then choose Toolbar option. Check whether having Hydrology Modeling module or not in Customize dialog. If have, tick the checkbox before it; or click the button showed in Fig. 2-1, then browse the DLL file (included in the installation disk directory): esrihydrology_v2.dll, and the default file directory is:
C:\ Program Files \ ArcGIS \ Developer Kit ¥ samples \ Spatial Analyst \ Hydrologic Modeling \ Visual Basic
Fig. 2-1 Add Hydrology Module in ArcMap
Then, check the check box before Hydrology Modeling in Customize dialog, Hydrology Modeling would appear in ArcMap interface.
Open the Hydrology module in ArcMap, and choose Fill Sinks. Browse DEM data for input surface, e.g. xmd_tingrid15; for output raster, could use the default, it is just the temporary data and generally stored in temporary folder of the Operation Systemic defaults, the output name is Filled Sink1.
If you want to keep the temporary data in your own customized folder, open Spatial Analyst ¥ Option, set General ¥ Working as your customize folder.
(2) Flow Direction Analysis
The direction of flow is determined by finding the direction of steepest descent from each cell. This is calculated as: change in z value / distance * 100.The distance is determined between cell centers. Therefore, if the cell size is 1, the distance between two orthogonal cells is 1, and the distance between two diagonal cells (i.e. when the direction, which is the neighboring cells towards the center cell, is 2, 8, 32, 128, as Fig. 2-2 shows) is 1.414.
Fig. 2-2 Flow Direction Calculation Between Center Cell and Neighbor Cells
Open Hydrology / Flow Direction in ArcMap; set Input Surface as the none-sink DEM built in step 1. If default before, it is Filled Sink1. Click ok, then get Flow Direction raster data: Flow Direction1, comprehend what the value of Flow Direction raster cell represents.