Masachusetts Institute of Technology - Department of Urban Studies and Planning


11.521
Spatial Database Management and Advanced Geographic Information Systems
11.523
Fundamentals of Spatial Database Management

Metro Boston Modeling, PostGIS, QGIS, and Geospatial Services

22 March, 2018


Administrative

Overview

 


Discussion of Tuesday's Test


Review parts of our snowed out Lab #6: http://mit.edu/11.521/labs/lab6

Illustrate use of QGIS and PostGIS

select z.casenumber, c.stcntytr
  from gis.lonlat5m z, gis.msa5_tr90m c
 where st_intersects(z.geom,c.geom)
 order by casenumber 
-- limit 10;
-- half-second to do all 1520 variances
select z.casenumber, c.stcntytr, "H061A001" medhval
  from gis.lonlat5m z, gis.msa5_tr90m c, bzoning.msa5_tr90_data d
 where st_intersects(z.geom,c.geom)
   and c.stcntytr=d."STCNTYTR"
 order by casenumber;
select z.casenumber, c.pid_long 
  from gis.lonlat5m z, gis.ebos_parcels05f c
 where st_intersects(z.geom,st_transform(c.geom,26986))
 order by casenumber;
-- gis.ebos_parcels05f is in Mass SP NAD83-feet
-- takes 2 minutes with no spatial indices

    
    

Modeling Metropolitan Growth and (Re)Development

As indicated in Lab #6, 'Scenario 360' is part of a modeling tool called Community Viz that is a substantial 'extension' to ArcGIS and that was orginally financed by the Orton Foundation. Community Viz is a little like Model Builder. That is, you use it to develop a sequence of ArcGIS spatial data processing steps that model changes to the built environment and urban activities. However, Community Viz goes further than Model Builder in providing analytic formulas for modeling complex interactions and wizards and tools to help in building and running the model. Community Viz also focuses on scenario modeling - that is, playing out the long term consequences of growth and development in the face of various constraints and interactions regarding land use, accessibility, environmental constraints, and the like.

The ongoing MetroFuture regional planning effort for metro Boston has been using Community Viz to model Boston devleopment out to 2030. The Metropolitan Area Planning Council (MAPC) is the organization that has built the MetroFuture model. We have their model but cannot currently run Community Viz on Windows-7 with ArcGIS 10.

Here are the 6 steps from Lab #6 regarding our exploration of Community Viz using the tutorials and the MetroFuture model for Boston.

(1) Skim the scenario 360 tutorial in the 'proj11' portion of the class locker: "K:\proj11\metrofuture\Scenario_360_Tutorials.pdf"

(2) Read the Your Guide to "Winds of Change" document in the class locker: K:\proj11\metrofuture\explore_scenario_woc.pdf This document explains one of the four scenarios that MetroFuture modeled using Community Viz. The 'indicators' and 'drivers' on the last few pages are the assumptions and key relationships in the Community Viz model that explain the results of modeling the effects of a 'winds of change' strategy for metro Boston growth out to 2030.

(3) * Since the MetroFuture model is quite complex, start first with one of the Community Viz 'tutorial examples'. BEFORE RUNNING ANY COMMUNITY VIZ MODEL, be sure to copy the entire CVFiles directory tree to a local, writeable drive.

(4) * Start Community Viz - Scenario 360 from the Start/Community-Viz/Scenario-360 menu on the WinAthena PC. Once ArcGIS comes up, you will see a new toolbar for scenario-360 and a dialog box asking you to choose the saved analysis you would like to open. Browse to your local copy of Communityville and select it. You will end up with an ArcMap window something like the one below:

Screen shot of Community Viz startup with Communityville example.

Note the new icons that appear in the panel on the left. In order to see the Scenario 360 toolbar, you may need to turn on th Scenario 360 toolbar in the View/Toolbars and Tools/Customize

The Communityville model calculates and visualizes the buildout that would be allowed for a hypothetical subdivision depending upon the zoning and environmental constraints that are imposed. Rerun the model after adjusting the assumption about minimum distance from the 'bird nest'. Note the reduction in allowed development that is computed and displayed as the distance-from-nest assumption is adjusted. Explore the diagram and formulas that codify the spatial and mathematical relationships.

(5) * If Start/Community-Viz is missing, you may need to reboot the machine. Also, after Communityville has been opened inside of ArcGIS, you may see a message indicating that ChartFX is not working. In this case, exit from ArcGIS since CommunityViz will not work. The machine needs to run an 'msi' installation file in: C:\Program Files\CommunityViz\Scenario 360\ChartFX\ChartFX.msi Unfortunately, you do not have permission to run this file (otherwise you could just double-click on the file to run it.) Depending upon the order of consideration of bootup files, these charting tools occasionally do not work. If you reboot the machine, the 'msi' file may get run properly during the bootup. After running ChartFX.msi, go back to step (4) to restart CommunityViz.

(6) * To run the MetroFuture model, copy the entire directory: K:\proj08\CVFiles_metro to a local, writeable directory. Then start CommunityViz as before (or start ArcGIS, make sure the CommunityViz extension is loaded and viewable, and choose scenario-360/analysis to browse to your local MetroFuture_BCCS_may2007 location).

(7) * Explore the effects of changing some of the assumptions for water or type of housing. (Rerunning the model after you have changed these assumptions takes some time, but much less than changes in many other assumptions.)

* The CommunityViz extension to ArcGIS is not currently installed on any CRON machines. These steps are indicated to give you a sense of how one runs the model with ArcMap after installing the Community Viz software.

The MetroFuture Community Viz model

 

 

The "Future Urban Mobility" project in Singapore

Intro to Internet GIS and Geospatial Web Services

Class Project

Today's class is the last meeting for the 11.523 portion of the semester. The remainder of the semester is the 11.524 portion of the class and will be focused on the class project plus a few exercises to help us utilize geospatial services. Right after Spring Break, Eric Huntley will introduce Leaflet and CartoDB tools for building web mapping services using shapefiles from our ArcGIS analyses and a 'backend' Postgres database.  The remaining weeks of the semester will be focused on the class project.  On Tuesday, April 10 in the W31-301 computing lab we will discuss class project possibilities and begin narrowing our focus.  That week, we will discuss class project possibilities with staff from the Metropolitan Area Planning Council (MAPC) and other Singapore researchers with whom we will be collaborating. For the class project, we will be focusing on Boston issues concerning metropolitan growth and community development (most likely related to current efforts to examine regional changes in vehicle ownership and miles driven) plus some aspect of our ongoing 'Future Urban Mobility' project in Singapore.

The Boston projects build on the work we have done during the past several years to examine (a) the implications of MetroFuture's alternative development scenarios for metro Boston, (b) the effect of built environment on vehicle miles traveled, (c) travel patterns in the recent MassDOT household travel survey and a special survey journey-to-school issues for grade schools in Massachusetts, and (d) the viability of energy efficiency improvement strategies in Cambridge (using electricity and gas meter readings combined with the building and assessment data that we have seen in our labs).

The Singapore projects build on current efforts of the 'future urban mobility' interdisciplinary research group at the Singapore/MIT Alliance for Research and Technology (SMART) Centre. Past projects include building a hedonic price model for the effect of amenities and accessibility on housing prices, analyzing travel patterns evident in the EZLink (tap in / tap out) transit cards, and developing, utilizing and evaluating local activity indices for downtown shopping districts.  This year, we are likely to focus on analyzing and visualizing housing relocation and vehicle ownership patterns predicted by the agent-based microsimulation platform (SimMobility) that is under development.

For the project, we will have available a number of datasets beyond those that we have used in 11.523 exercises. These data include detailed parcel, transit, land use and zoning data in both Boston and Singapore; multi-year, geocoded estimates of annual mileage for every registered private passenger vehicle in Mass; and various household travel surveys, taxi and transit fare card transactions, and real estate sales data. We continue to use PostgreSQL and also its GIS extensions (PostGIS) that can be added to a Postgres installation so that spatial proximity questions (like distance from, contained within, etc.) can be included in SQL queries. Use of QGIS in place of ArcGIS will also be an options for certain project purposes.


Home | Syllabus | Lectures | Labs | CRN | MIT
Created by Joseph Ferreira; last modified: 21 March 2018 [jf]