URPL-GP.4651

Advanced GIS: Topics in Spatial Analysis

Spring 2018

Wednesdays, March 21 – May 2

Lecture: 6:45 – 8:25 p.m.

Lab: 8:35 – 9:35 p.m.

25 W. 4th St. Room C-6

Sean Capperis

Office Hours: By appointment

Prerequisites: URPL-GP.2618, CUSP-GX.3007, or equivalent.

Course Description

Sophisticated analysis of spatial data has become increasingly important to understand emerging issues and support decision-making in urban planning and policy. GIS software facilitates our ability to answer questions about spatial relationships: presence, absence, access, exposure, dispersion, and connectivity. This course will focus on the GIS tools that measure conditions and manipulate data by spatial relationship. We will seat each technique in a specific, practical application in urban planning/policy, and discuss how each technique’s underlying concepts are related to the problem we are trying to solve. We will explore the following concepts and tools: the modifiable areal unit problem, aggregation, neighborhood definition, and polygon apportionment; spatial distributions and cluster identification; and the network data model and network analysis. We will apply those tools across many domains: neighborhood indicators, demographic and socioeconomic conditions, land use regulation, environment, resilience, public safety, and transportation. Lab exercises and assignments will largely focus on current issues in New York City and reinforce visualization and data management practices introduced in earlier classes. The course will culminate in a student-defined, portfolio-quality final project.

Course Objectives

  1. Understand various spatial relationship concepts, as well as how and when to apply them
  2. Identify and account for common methodological challenges using GIS in urban settings
  3. Critique and defend the selection of analysis methods and spatial relationship models
  4. Use the following new techniques and tools effectively: neighborhood definition, polygon apportionment, hot spot analysis, Network Analyst, and Model Builder
  5. Further develop the ability to use spatial evidence to make a sound argument about an issue relevant to planning/policy
  6. Develop deeper fluency in cartography, data management, and spatial problem solving
  7. Gain more exposure to commonly used spatial data in New York and other U.S. cities

Class Format and Preparation

Each session is made up of two periods: a lecture from 6:45 to 8:25 p.m. and flexible lab from 8:35 to 9:35 p.m. The lecture balances a lecture/discussion that covers key concepts with a hands-on exercise that demonstrates and applies that week’s concepts and tools to a real-world issue. The lab provides time to more deeply explore lab exercises, begin work on graded assignments, and to get individual or small-group feedback from the instructor.

Required Materials

Software

We will be using Esri ArcGIS Desktop for all in-class instruction and exercises. You can use ArcGIS in NYU computer labs or install it on your personal computer, but it requires Windows. (Mac users can run ArcGIS with some effort. You will need to use Bootcamp or Parallels; I myself use Bootcamp.) You may obtain a free educational license for ArcGIS from the NYU Data Services website. When installing and using ArcGIS, request the “ArcGIS for Desktop Advanced” product and ensure that the extensions 3D Analyst, Network Analyst, and Spatial Analyst are installed and licensed.

Data Storage

You will need to maintain a data library that we will use for weekly in-class exercises. Please have a USB drive or cloud storage available for every class session. I will send more details on how to build your data library prior to our first class meeting.

Texts

There is no required textbook for this class. All readings are available online, and links are provided in the course schedule below.

If you need a refresher on basic GIS techniques, especially as they are employed in urban planning and policy, I recommend GIS Tutorial 1: Basic Workbook by Wilpen Gorr and Kristen Kurland. This text is optional.

NYU Classes

All announcements, resources, and assignments will be delivered through the NYU Classes site. I may modify assignments, due dates, and other aspects of the course as we go through the term with advance notice provided as soon as possible through the course website.

Academic Integrity

Academic integrity is a vital component of Wagner and NYU. Each student is required to sign and abide by Wagner’s Academic Code. Plagiarism of any form will not be tolerated since you have all signed an Academic Oath and are bound by the academic code of the school. Every student is expected to maintain academic integrity and is expected to report violations to me. If you are unsure about what is expected of you, please ask.

Academic Integrity in the Context of This Course

Unless otherwise specified, all assignment submissions must be the result of new, original, individual work. You are not permitted to share data or models that you create for assignments/projects with your colleagues, use data created by your peers in an assignment/project, or turn in the work of someone else, including data, maps, and text, as your own. Additionally, you may not submit any work for this class that you completed to fulfill requirements of other classes, past or current. However, with my permission, you may build or innovate on your past work in an original way using techniques learned in this class. I do permit and encourage you to consult your peers, me, and outside resources (including NYU Data Services) for advice using GIS techniques and data.

Henry and Lucy Moses Center for Students with Disabilities at NYU

Academic accommodations are available for students with disabilities. Please visit the Moses Center for Students with Disabilities (CSD) website at click on the Reasonable Accommodations and How to Register tab or call or e-mail CSD at (212-998-4980 or ) for information. Students who are requesting academic accommodations are strongly advised to reach out to the Moses Center as early as possible in the semester for assistance.

NYU’s Calendar Policy on Religious Holidays

NYU’s Calendar Policy on Religious Holidays states that members of any religious group may, without penalty, absent themselves from classes when required in compliance with their religious obligations.

Student Resources

NYU and the Wagner school provide additional resources—particularly for writing and data use—that you may find helpful for this class. Your writing abilities play an important role in your evaluation in this class. See Wagner’s writing skills workshops and, particularly if you are not a Wagner student, its memo on writing memos.You may also want to avail yourself of the NYU Writing Center. You may find NYU Data Services particularly helpful with ArcGIS software support (beyond the scope of this class), data manipulation outside of GIS, and finding data that may be useful for your final project.

Assignments and Evaluation Policies

Your final grade is comprised of the following components:

Class participation (5%): Because we will be doing advanced work, and there is no authoritative text for the course, it can be tricky to fully grasp the material without working together in class. Generally, if you attend class and participate to the best of your ability, you can expect to receive full credit. Some conflicts are unavoidable (illness, family care, capstone client meetings, etc.), so I will permit one absence without penalty. Please notify me in advance of absences due to religious holidays; these have no penalty. If you fall behind in an exercise and cannot catch up, you can still receive full credit as long as you work with your neighbor.

Problem set (15%): This assignment will evaluate your understanding of concepts covered in sessions 1 and 2. This is required of all students.

Memos (2 at 15% each): You must choose to complete two of three available memo assignments. You may choose a memo assignment based on your interest in the technique, topic area/data set, or both.

Final project (50%): In the final project, you will form and answer an urban planning/policy question requiring the use of spatial analysis techniques. You will first submit a proposal by the start of session 6. Similar to memo assignments, the final project will require a longer memo (2 pages) that describes the issue/challenges, research question, methodology (including data sources, GIS tools, and any other methods) and results, and discusses results critically, placed in the context of the original issue. The memo must also include display maps and/or tables that support the memo text. You may complete the final project individually or in pairs. I will share further final project specifications in class.

All assignments should conform to the following policies:

Assignment requirements: All assignments must conform to the assignment guide and display map style guides posted on NYU Classes.In addition to a write-up, all graded assignments will require you to submit display maps, output data, or both.

Deadlines and submission: All assignments are due by the start of class on the specified date. Except in case of emergency, I will not accept late assignments, because we will discuss the solution in class. Students should submit all assignments electronically via NYU Classes.

Software: Some analysis techniques that we will cover can only be done in Esri ArcGIS (e.g., network analysis). However, if you are more comfortable using other packages like QGIS for common GIS functionality (e.g., data processing, cartography), you are welcome to do so.

Grading criteria: I score memo assignments using the criteria below. Each criterion receives a score of 0-4: advanced understanding of course objectives, 4; competent work, showing basic understanding of concepts, 3; inadequate, showing significant flaws or gaps in understanding of material, 2; incomplete or ungradable, 0. A score of 3 or higher indicates understanding of course objectives and expectations.

  • Methods (50%): The use of methods and tools demonstrate understanding of the underlying concepts, principles, and application of tools to the situation. The student designed a process appropriate to the challenge or question, selected the right tool for the task, sequenced tools properly and efficiently, entered the correct parameters, selected the correct input data, and output the correct results.
  • Display map design (25%): The maps tell a clear story and present an appropriate amount of spatial context. The layers shown in the maps support the narrative in the written deliverable. Symbology, extent, layer order, and hierarchy are appropriate to the features in the map. Please refer to map style guide for expectations.
  • Written content (25%): Written materials deliver a clear, coherent, well-structured argument, and directly address the question posed in the assignment prompt. The student uses appropriate evidence (visual and numerical) to support their conclusions. Methods are described clearly and accurately. The tone is appropriate for the audience.

For the final course grade, numerical scores are translated to letter grades using the following scale: A: 3.71-4; A-: 3.51-3.7; B+: 3.31-3.5; B: 2.71-3.3; B-: 2.51-2.7; C+: 2.31-2.5; C: 1.71-2.3; C-: 1.51-1.7; F: 0-1.5

Learning Assessment Table

Graded Assignment / Course Objective Covered
Class participation / n/a
Problem set / 1, 2, 4, 6, 7
Memos / all
Final project / all

Course Overview

Week / Date / Topic / Deliverable
Week 1 / March 21 / Introductions
Advanced polygon aggregation / Pre-class assignment
Week 2 / March 28 / Polygon apportionment
Week 3 / April 4 / Cluster analysis / Problem set
Week 4 / April 11 / Network analysis / Memo 1 (if completing)
Week 5 / April 18 / TBA / Memo 2 (if completing)
Week 6 / April 25 / Command line and ModelBuilder / Memo 3 (if completing)
Final project proposal
Week 7 / May 2 / Open lab
May 11 / No class / Final project due at 5:00 p.m.

Detailed Course Overview

Note: Readings and pre-class assignments are subject to change.

March 21 – Session 1

Advanced polygon aggregation – Demographic/socioeconomic data as neighborhood indicators

Concepts / Tools
Tobler's First Law
Modifiable areal unit problem (MAUP)
Ecological fallacy / Select by location (review)
Summary statistics (review)
Dissolve (review)
Spatial join (review)

Required readings:

  • Lloyd, Christopher D. 2010. “Key Concepts 3: Spatial Data Analysis.” Spatial Data Analysis: An Introduction for GIS Users. Oxford: Oxford University Press. 43-64. [ebook link; focus on sections 4.6, 4.9, and 4.10; also focus on concepts rather than math]
  • Koschinsky, Julia. 2014. “Beyond Mapping: Spatial Analytics and Evaluation of Place-Based Programs.” In Strengthening Communities With Neighborhood Data, edited by G. Thomas Kingsley, Claudia J. Coulton, and Kathryn L. S. Pettit. Washington, DC: Urban Institute. [ebook link; note: section “Spatial Data, Methods, and Tools for Evaluation” is optional]

Optional readings:

  • Kingsley, G. Thomas, Claudia J. Coulton, Kathryn L. S. Pettit. 2014. “Advances in Analytic Methods for Neighborhood Data.” In Strengthening Communities With Neighborhood Data. [link]
  • Van Ham, Maarten, and David Manley. 2012. “Neighbourhood Effects Research at a Crossraods. Ten Challenges for Future Research.” Environment and Planning A 44: 2787-2793.[link; focus on challenges section, and challenges 7-8 in particular]

In-class exercise: Advanced aggregation – redefining concentrated poverty

Due: Pre-class assignment

March 28 – Session 2

Polygon apportionment – Demographic/socioeconomic data in custom study areas

Concepts / Tools
Pycnophlactic principle
Apportionment algorithms
Continued discussion of concepts from session 1 / Intersect (review)
Field calculator (review)
Calculate geometry (review)

Required readings:

  • Beale, Laurie. Apportioning Population Between Areas. 2012. (presented at Esri International User Conference) [link]
  • NYU Furman Center. 2015. “Focus on Density.” State of New York City’s Housing and Neighborhoods in 2014. [link] – focus on section “4. Population density in New York City is associated with certain positive neighborhood amenities and largely unrelated to many negative attributes,” pp. 18-20

In-class exercise: Demographics in the Greenpoint-Williamsburg rezoning

April 4 – Session 3

Clustering – Public safety incidents

Concepts / Tools
Spatial distribution
Spaitial heterogeneity
Spatial autocorrelation
MAUP (revisited) / Create fishnet
Hot spot analysis (Getis-Ord Gi*)

Required readings:

  • Lloyd, Christopher D. 2010. “Key Concepts 3: Spatial Data Analysis.” Spatial Data Analysis: An Introduction for GIS Users. Oxford: Oxford University Press. 43-64. [link; focus on sections 4.7 and 4.8, review section 4.9]
  • Esri. ArcMap [documentation].
  • How Hot Spot Analysis (Getis-Ord Gi*) works [link; “Calculation” section is optional]
  • How Optimized Hot Spot Analysis works [link]
  • Review Koschinsky 2014 [session 1].

Optional readings:

  • McGlone, Daniel. 2014. “Analyzing Philadelphia Crash Data.” Azavea Atlas. [link]
  • Eck, John E., Spencer Chainey, James G. Cameron, Michael Leitner, Ronald Wilson. 2005. Mapping Crime: Understanding Hot Spots. Washington, DC: U.S. Department of Justice. [link; see chapters 1 and 2]

In-class exercise: Detecting crime clusters

Due: Problem set

April 11 – Session 4

Networks – Infrastructure walksheds

Concepts / Tools
Network data model
Network distance / Creating a network dataset
Creating service areas
Generalize
Feature vertices to points

Required readings:

  • Trodd, Nigel. 2005. GeoImaging & GeoInformatics. “Network Analysis.” [link; focus on concepts rather than algorithms, especially in Dijkstra discussion]
  • Esri. ArcGIS Network Analyst Tutorial. [do not complete the exercises]
  • About the ArcGIS Network Analyst extension tutorial [link]
  • Exercise 1: Creating a network dataset [link]
  • Exercise 5: Calculating service areas and creating an OD cost matrix [link]
  • Exercise 9: Choosing optimal store locations using location allocation [link]

Optional readings:

  • Wellington, Ben. 2016. “Almost One Third of NYC is Not Patrolled by the Closest Precinct House.” I Quant NY. [link]
  • Esri. ArcGIS Network Analyst Tutorial. Exercise 2: Creating a multimodal network dataset [link]
  • Oh, Kyushik, and Seunghyun Jeong. 2007. “Assessing the Spatial Distribution of Urban Parks Using GIS.” Landscape and Urban Planning 82: 25-32. [link]

In-class exercise: Finding proximity to parks in Brooklyn

Due: Memo 1 (if completing)

April 18 – Session 5

Topic TBA

Concepts / Tools
TBA / TBA

Required readings: TBA

In-class exercise: TBA

Due: Memo 2 (if completing)

April 25 – Session 6

ModelBuilder – Replication

Concepts / Tools
Replication / ModelBuilder
Command line

Required readings: None

In-class exercise: ModelBuilder for Neighborhood Aggregation

Due: Memo 3 (if completing), final project proposal

May 2 – Session 7

Open Lab – Final project work session

Friday, May 11 at 5:00 p.m. – Final Project Due