Spatial Data Analysis with R

Note

Please note that these pages for the course I used to teach are provided for reference only. I am no longer at UMass, but if you wish to reach out regarding anything on the site you may email me at michaelfrancenelson@gmail.com.

Welcome to the course!

Instructor: Michael France Nelson

The course meets Mondays and Wednesdays 11:15AM - 12:05PM in Holdsworth Hall room 211.

Lab sessions are on Wednesdays at 12:20PM - 2:15PM in Holdsworth Hall room 211

Office Hours:

• Thursdays 2-3 PM
• Other times by appointment (email me)

Office hours are in my office (Holdsworth Hall room 311)

Course Readings

The main course text is

• Spatial Ecology and Conservation Modeling: Applications with R. (2018). Robert Fletcher and Marie-Josée Fortin

Note: you may access the text for reading online through UMass libraries. You are limited to 300 accesses.

Hardcover copies are available for purchase.

Course Units

Unit 1: Introductions

Overview and Objectives

In this first unit, we’ll:

• review essential concepts in statistics
• review essential concepts GIS/geography
• set up our course software
• begin working with spatial data in R

Main Readings

• Introduction: F&F Ch. 1
• Scale: F&F 2.1 – 2.2, 2.4
• First R examples: F&F 2.3

Note: F&F refers to our course text ‘Spatial Ecology and Conservation Mapping’

Review and supplemental readings:

• R Beginnings: F&F Appendix, pp. 489 - 505
• Spatial Data in R: F&F Appendix, pp 505 - 511
• Note that the code for importing ‘landbird’ is on page 496
• Check the Wiki for other resources, and be sure to add any other useful resources you might find!

Slide Decks

• Deck 1

Labs

• Lab 0: Software Setup
• Lab 1: Spatial Data Basics - Vector Data

Unit 2: Land Use/Land Cover and Scale

Overview and Objectives

In this unit, we’ll:

• further develop our understanding of scale
• refine and expand our concept of distance
• explore land cover data concepts
• explore multiple conceptual models of land cover

Main Readings

• Scale: F&F 2.1, 2.2, 2.4, 2.5
• Additional scale R examples: F&F 2.3
• Land cover concepts: F&F 3.1, 3.2, 3.4
• Working with land cover data in R: F&F 3.3
• Distance concepts: F&F 7.2.1, 7.2.2, 9.2.1 - 9.2.3
• Network concepts: Pautasso, M., Moslonka-Lefebvre, M., and Jeger, M.J. (2010). The number of links to and from the starting node as a predictor of epidemic size in small-size directed networks. Ecological Complexity 7, 424–432.

Review and supplemental readings/materials:

Interplay of spatial and network distances:

• Harwood, T.D., Xu, X., Pautasso, M., Jeger, M.J., and Shaw, M.W. (2009). Epidemiological risk assessment using linked network and grid based modelling: Phytophthora ramorum and Phytophthora kernoviae in the {UK}. Ecological Modelling 220, 3353–3361.
• Zhong, C., Arisona, S.M., Huang, X., Batty, M., and Schmitt, G. (2014). Detecting the dynamics of urban structure through spatial network analysis. INTERNATIONAL JOURNAL OF GEOGRAPHICAL INFORMATION SCIENCE 28, 2178–2199.

Land cover state and transition models:

Landfire Project

Slide Decks

• Deck 2: Scale, LULC

Labs

• Lab: Spatial Data - Rasters and Scale
• Lab: Land Use/Land Cover

Unit 3: Point Patterns and Descriptive Spatial Statistics

Overview and Objectives

In this unit, we’ll:

• develop our intuition about spatial pattern and probably
• explore methods of quantifying spatial pattern

Main Readings

• point patterns: F&F 4.1, 4.2, 4.4, 4.5
• point patterns in R: F&F 4.3
• A point pattern example: Flügge, Anton J., Sofia C. Olhede, and David J. Murrell. “The memory of spatial patterns: changes in local abundance and aggregation in a tropical forest.” Ecology 93.7 (2012): 1540-1549.

Review and supplemental readings/materials:

Example of point-process analysis

• Urban lead contamination: Walter et al. 2005. Spatial point-process statistics: concepts and application to the analysis of lead contamination in urban soil.

Some spatial statistics references

• Peter J. Diggle. Statistical Analysis of Spatial and Spatio-Temporal Point Patterns, Third Edition. 3rd ed. Chapman & Hall/CRC Monographs on Statistics & Applied Probability. Hoboken: Taylor and Francis, 2013.
• Illian, Janine, Antti Penttinen, Helga Stoyan, and Dietrich Stoyan. Statistical Analysis and Modelling of Spatial Point Patterns. Vol. 70. John Wiley & Sons, 2008.

Slide Decks

• Deck 3a: Concept review, prep for point patterns
• Deck 3b: Point Patterns and Descriptive Spatial Statistics
• Deck 3c: Point Pattern Significance and Basic Interpolation

Labs

• Point Patterns lab

Unit 4: Spatial Dependence

Overview and Objectives

• Review key general statistical concepts and learn about their spatial counterparts:
  • common assumptions required for inference
  • model diagnostics
  • independence, non-independence
  • autocorrelation and correlation structures
  • how to quantify dependence
  • how to proceed when assumptions are not met
  • Introduce spatial interpolation concepts: variograms, Kriging, etc.

Main Readings

• Spatial dependence/autocorrelation: F & F sections 5.1, 5.2
• Quantifying spatial dependence: F & F sections 5.3 - 5.5
• Chapter 6

Review and supplemental readings/materials:

• Interpolation of urban air pollution: Schneider et al, 2017. Mapping urban air quality in near real-time using observations from low-cost sensors and model information.
• Hatna, Erez, and Itzhak Benenson. “The Schelling Model of Ethnic Residential Dynamics: Beyond the Integrated - Segregated Dichotomy of Patterns.” Journal of Artificial Societies and Social Simulation 15, no. 1 (2010): 6.
• Biswas, Shekhar R., Rebecca L. MacDonald, and Han Y. H. Chen. “Disturbance Increases Negative Spatial Autocorrelation in Species Diversity.” Landscape Ecology 32, no. 4 (April 1, 2017): 823–34. https://doi.org/10.1007/s10980-017-0488-9.

Slide Decks

• Deck 4a: Spatial Autocorrelation/dependence
• Deck 4b: Autocorrelation 2: kriging, misc. topics

Labs

• Lab: Spatial Dependence and Autocorrelation

Unit 5: Spatially-Aware Regression

Overview and Objectives

The primary objective of this unit is to give a broad overview of regression modelling techniques that can account for spatial dependence in various ways.

Topics include:

• Quantifying spatial autocorrelation in regression models
• Spatial lag models
• Modeling variance/covariance

Main Readings

• F+F Chapter 6

Review and supplemental readings/materials:

• Bivand, Roger S., Edzer J. Pebesma, Virgilio Gomez-Rubio, and Edzer Jan Pebesma. Applied Spatial Data Analysis with R. Vol. 747248717. Springer, 2008. http://link.springer.com/content/pdf/10.1007/978-1-4614-7618-4.pdf.

Slide Decks

• Deck 5a: Spatial Autocorrelation/dependence
• Deck 5b: Classes of Spatial Models
• Deck 5c: Examples of Spatial Models

Labs

• Lab 6: Spatially-Aware Regression

Unit 6: Additional Topics & Main Projects

Main Readings

Topics and readings to be chosen by students

Review and supplemental readings/materials:

Slide Decks

Labs

Main Projects

You can read about the course Main projects by clicking on this link

Course Materials

You can find links to the lab assignments, lecture notes, and other course materials in the tabs below.

Syllabus

You can check out the Syllabus here.

Lab Assignments

• Lab 0: Software Setup
• Lab 1: Spatial Data - Vectors
• Lab 2: Spatial Data - Rasters and Scale
• Lab 3: Land Use/Land Cover
• Lab 4: Point Pattern Analysis
• Lab 5: Spatial Dependence and Autocorrelation
• Lab 6: Spatially-Aware Regression

In-Class Activities and Tutorials

• In-Class Vector Data Operations
• In-Class Raster Data Operations
• In-Class Raster Data Operations 2
• In-Class Point Pattern Analysis 1
• In-Class Point Pattern Analysis 2
• In-Class Interpolation
• In-Class Vector Data with sf
• In-Class Centroids, Tessellations, and Neighborhoods
• In-Class Random Walks
• In-Class Brownian Bridges Intuition
• In-Class Home Ranges

Lecture Slides

• Deck 1: Introductions
• Deck 1a: Main project info and R vector operations
• Deck 2: Scale, LULC
• Deck 3a: Concept review, prep for point patterns
• Deck 3b: Point patterns
• Deck 3c: Point patterns - significance and basic interpolation
• Deck 4a: Spatial Autocorrelation/dependence
• Deck 4b: Autocorrelation 2: kriging, misc. topics
• Deck 5a: Spatial Autocorrelation/dependence
• Deck 5b: Classes of Spatial Models
• Deck 5c: Examples of Spatial Models
• Deck 6a: Potpourri - ggplot2, SDMs, NetCDF
• Deck 8a: Space Use

Data Files

Course data files will live here:

• aqs-daily-ozone-2017.zip
• ca_border.GPKG
• ca_counties.GPKG
• CA_Counties_TIGER2016.zip
• CA_ozone_2017.GPKG
• dvnp.GPKG
• grsm_sp.Rdata
• ia_counties_tiger_2020.zip
• Iowa_County_Boundaries-shp.zip
• nv_border.GPKG
• nv_counties.GPKG
• NY8_utm18.zip
• ozone.csv
• PRISM_ppt_30yr_normal_4kmM3_annual_bil.zip
• PRISM_tmean_30yr_normal_4kmM3_annual_bil.zip
• temperature.zip
• terrain.nc
• vector_practice.RData
• wy_counties.zip

R and Other Course Resources

R Refresher/New R Learner Resources

DataCamp

Enrolled students will use some of DataCamp’s materials.

You can use the DataCamp invite link in the Moodle site to create your free account. This will give you access to the materials we’ll use in this course.

YaRrr! The Pirate’s Guide to R

A really excellent guide to learning R by Nathaniel D. Phillips

Warning: contains pirate-speak and terrible jokes!

Spatial R Resources

Spatial Data Science with R

An amazing resource with lots of info about working with spatial data and spatial analyses in R.

Moran’s I Tutorial

A basic introduction to Moran’s I analysis in R

Other Spatial Data Courses

Applied Spatial Analysis for Public Health

This site outlines an 8 week online course on Applied Spatial Analysis for Public Health using R

Intro to GIS and Spatial Analysis

A great online text for an intro GIS/spatial analysis course. This is not R-specific.

Understanding Map Projections

An entire book by ESRI all about map projections!