Return to Outline Copyright © 2009 by Maribeth H. Price 2-1 Chapter 2 Mapping GIS Data.

Return to Outline

Copyright © 2009 by Maribeth H. Price

2-1

Chapter 2

Mapping GIS Data

Return to Outline


2-2

Outline• GIS Concepts

– Map scale– Ways to map data– Classifying numeric data– Displaying rasters

• About ArcGIS– Map documents and data frames– Using ArcMap– Data frame coordinate systems– Symbolizing features– Symbolizing rasters

Return to Outline


2-3

Map scale

Return to Outline


2-4

How maps portray the world

Point features

Line features

Polygon features

Annotation features

Return to Outline


2-5

Map scale

• Ratio of distance on the map to distance on the ground

• Dimensionless: cm or inches or mm…

1 cm on map = 100,000 cm on ground

Return to Outline


2-6

Talking about map scale

• A large denominator gives a small fraction a small scale map. It shows a large area.

• A small denominator gives a larger fraction a large scale map. It shows a small area.

1--------

50,000,000

1--------

500,000

1--------5,000

Return to Outline


2-7

Generalization and Scalehttp://encarta.msn.com/map_701515760/portsmouth.html

Polygons at one scale may be points or lines at a different scale

Large scale map

Intermediate scale map

Small scale map

Return to Outline


2-8

Scale and precision

1:100,000 scale map3-pt highway symbol1-pt local road symbol

1 inch = 72 points

3 pts = 1 S 100000S = 300,000 pts S = 4166 inches

S= 347 feet

Let S be the size of the highway represented by a 3-point line symbol

So the highway location has an uncertainty of nearly 350 ft due solely to the symbol used to portray it. The local roads have an uncertainty of about 115 feet.

Return to Outline


2-9

Estimating precision from scale

• A 1-pt line is about 0.001 feet

• Map scale / 1000 gives approximate precision in feet for a 1-pt thick line

1:5,000 5 ft

1:24,000 24 ft

1:100,000 100 ft

1: 1 million 1000 ft

Larger symbols would have lower precision.

Return to Outline


2-10

Source scale and display scale

• Most GIS data have an intrinsic scale inherited from the source

• Display scale varies

1:24,000 USGS Topo Map (source scale)

Return to Outline


2-11

Display vs source scale

• Once in GIS data may be displayed at any scale, BUT

• Original scale of the map does impact the precision and accuracy of the data.

Original scale1:25 million

Original scale1:5 million

You should not display or analyze data at scales very different from the original source data.

Return to Outline


2-12

Scale and resolution

• Resolution is the sampling distance of the stored x-y values.

1:5M scale source 1:25M scale source

A larger scale map generally has a finer sampling distance and better spatial resolution. It can represent features with better precision.

Display scale approximately 1:500,000

Too fine a resolution wastes storage space and slows drawing—stores more points than needed at a particular display scale

Return to Outline


2-13

Finding source scale

• Usually documented in the metadata– Scale of original paper map source– Scale or precision at which data were

gathered

Return to Outline


2-14

Ways to map data

Return to Outline


2-15

Map Types and Data Types

• Single symbol maps• Unique values maps• Quantities maps

– Graduated color– Graduated symbol– Dot density

• Nominal data• Categorical data• Ordinal data• Interval and Ratio

data

Return to Outline


2-16

Nominal data

• Names or uniquely identifies objects– State names– Owner of parcel– Tax ID number– Parcel ID Number

• Each feature likely to have its own value

• Usually portrayed on a map as labels

Return to Outline


2-17

Single symbol maps

• Display all features with the same symbol• Combine with labels to portray nominal data

Return to Outline


2-18

Categorical data

• Places features into defined number of distinct categories

• Category names may be text or numeric• Portrayed by different symbol for each category

Return to Outline


2-19

Unique values maps

• Different symbol for each category or value

Geologic unitsVolcano types Road types

Return to Outline


2-20

Types of unique values

Nominal dataUse to show different features

Categorical dataState subregionUse to show patterns

Return to Outline


2-21

Ordinal data

• Places features into ranked categories or along an arbitrary scale– Low, Medium, High

slope– Village, Town, City– Assistant, Associate,

Full professor– Grade A, B, C, D, F

A 0-40%B 40-70%C 70-100%

Portrayed as categories but choosing variations in symbol size or color to indicate increase

Return to Outline


2-22

Interval or Ratio data

• Interval data places values along a regular numeric scale– Supports addition/subtraction– Temperature, pH, elevation

• Ratio data places values along a regular scale with a meaningful zero point– Supports addition, subtraction, multiplication,

division– Population, rainfall, median rent

Return to Outline


2-23

Mapping numeric data

• Interval and ratio data must be divided into classes before mapping

• Mapped using variations in symbol size, thickness, or hue

Return to Outline


2-24

Classed mapsGraduated color map(choropleth map) Graduated symbol map

Return to Outline


2-25

Colors for choropleth maps

• Generally use change in saturation or close hues to indicate increase

• Avoid using too many colors which tend to mask increase

Return to Outline


2-26

Normalizing classed maps

• If the size of the sample impacts the measured value, data should be normalized– By percent of total

• Percent of farms in each state• Percent of mobile homes in

each state

– By another field• Farms divided by area• Mobile homes divided by total

housing units

Number of farms

Number of farms per sq. mile

Return to Outline


2-27

Unclassed mapsProportional symbol map Dot density map

Return to Outline


2-28

Chart Maps

Proportional chart map

Return to Outline


2-29

Symbol psychology

Where is the water?

Where is there less rain?Which towns have more people? What’s there?

Where’s the danger?

Return to Outline


2-30

Choosing symbols

Which one looks more aesthetic?Which one is easier to understand?Which one shows the roads better?

Return to Outline


2-31

Classifying numeric data

Return to Outline


2-32

Ways to classify data

• Choose number of classes• Variety of different classification methods

– Jenks Natural Breaks– Equal Interval– Defined Interval– Quantile– Standard Deviation– Manual (set your own)

• Best methods vary depending on data distribution

Return to Outline


2-33

Common data distributions

Value

Num

ber

of

sam

ples

Normal

Uniform

Skewed

Bimodal

Return to Outline


2-34

Jenks Natural Breaks

•Exploits natural gaps in the data•Good for unevenly distributed or skewed data•Default method, works well for most data sets

Class breaks

Return to Outline


2-35

Equal Interval

Specify number of classesDivides into equally spaced classesWorks best for uniformly distributed data

Return to Outline


2-36

Defined interval

User chooses the class sizeData determines number of classesWorks best for uniformly distributed data

Return to Outline


2-37

Quantile

Same number of features in each classMay get very unevenly spaced class rangesResults depend on data distribution

Return to Outline


2-38

Geometrical Interval

Multiplies each succeeding class boundary by a constantWorks well for normal and skewed distributions

Return to Outline


2-39

Standard Deviation

Shows deviation from meanUser chooses units e.g. 0.5 standard deviationsAssumes data are normally distributed

Return to Outline Copyright © 2009 by Maribeth H. Price 2-1 Chapter 2 Mapping GIS Data.

Documents

map scale slide

feet map scale

larger scale map

original scale

intrinsic scale

scale http

scale source1

map scale ratio of distance