February 7, 2004.

February 7, 2004. Follow up from meeting of February 2, 2004.

Electronic maps showing vacant parcels/surface parking and related files obtained from Planning.
Regulatory mechanisms are different for parcels within the DDA from those not within the DDA. Thus, the initial focus is on parcels within the DDA.
There are five existing zoning categories that support premiums: C1A, C1A/R, C2A, C2A/R, and C2B/R.
Taking these points into consideration, along with environmental and cultural factors, a map was made of the vacant/surface parcels lying within the DDA that

did not lie within any existing historical district
whose centroid did not lie within the Allen Creek floodplain (with the thought that those parcels that did intersect the floodplain, but whose centroid lay outside the floodplain, might be candidates for development on the substantial non-floodplain portion--perhaps with public benefit coming from donating the floodplain portion to continuous greenspace, or some such environmental amenity).

The animated map below shows the building footprint file displayed previously, in shades of pastels reflecting building height in stories; the file of red parcels that comes in represents the set of all vacant/parking parcels in the five premium categories within the DDA--only those that do not lie also in an historic district or in the floodplain.

The following set of Virtual Reality files, with the sun in the south in a "high" position and with terrain underlying the street pattern based on City contour files, was created based on assigning

VR1: 300% to all parcels in red, lot line to lot line (back and forth; side to side)
VR2: FAR to all parcels in red by zoning category, lot line to lot line (back and forth; side to side)
VR3: FAR premiums as "by-right" values to all parcels in red by zoning category, lot line to lot line (back and forth; side to side)
VR4: 600% to all parcels in red, lot line to lot line (back and forth; side to side).

The set of VR files is organized roughly according to amount of height, from least to most. The first two files are quite similar to each other; the second one has two buildouts that are one story shorter than in the first one (in C1A); it has more than that that are one story taller than in the first one (in C2A). The last two files (third and fourth) are even more similar to each other. They are identical everywhere except that there are two buildouts, in the S. University area, that are 400% on the third one, and 600% on the fourth one.

The amount of square footage in the red (assuming multiplying the parcel footprint area times the number of stories) is as in the following table--square footage includes only parcel area...not streets:

VR1. 1238694.372

VR2. 1318163.76

VR3. 2465865.642

VR4. 2477388.744

Screen captures of the "closer view" of each VR are shown as an animation below--to experience being able to walk around, the user must go to the actual VR, though.

C1A: 200 by right, 400 with premiums

C1A/R: 300 by right, 600 with premiums

C2A: 400 by right, 600 with premiums

C2A/R: 300 by right, 600 with premiums

C2B/R: 300 by right, 600 with premiums.

Use of the FAR means that if a building does not cover the entire parcel, that then it can be taller, in order to gain the same area. Thus, if a building covers 1/2 of a parcel it can be 1/(1/2)=2 times as tall as if it covered the entire parcel. If a building covers 75%, or 3/4, of a parcel, then it can be 1/(3/4)=4/3 as tall as if it covered the entire parcel. The pattern is generally one involving a rectangular hyperbola (see link to December 21a article in Solstice). Expressing coverages in percentages, rather than by using setbacks, might make matters easier to understand (although of course one can be translated to the other). Policy involving setbacks will determine how any "by right" values play out in the actual landscape, on a zoning category by zoning category (or other category) basis, or on a parcel by parcel basis (in the case of PUDs and other special zonings). So, no assumptions involving setbacks or anything of that sort were made in this initial case.

One natural first step might be simply to use what is already present in code and allow the "premiums" values in these categories to become the "by right" values for these categories. Then, there would be no more "premiums." A new zoning category might be the result: with one piece describing mixed use on parcels in the DDA that do not lie in the floodplain and do not lie in an historic district, and another piece describing similar parcels in a 'walk-to-work' zone beyond the DDA. As in the Ann Arbor Zoning Ordinance book, there might be various considerations addressing Maximum usable age of Lot Area, Required Setback Line Dimensions, Maximum Height, Minimum Gross Lot Size.
The step that is most important for this sort of mapping effort, beyond determining accuracy which will take time, is to get information involving actual breakpoints that tie to construction costs (that Beal mentioned last time and that he's mentioned previously). With those in hand, then it makes sense from the standpoint of making efficient use of time and computing resources to calculate a number of different height scenarios.
Just beyond that, further study might lead to maps with

digitized land areas smaller than entire parcels, extruded according to some policy involving setbacks or placement or design
similar analysis involving areas within the DDA but not of these five zoning types and outside the DDA within the buffers--need

census block data that Chandra Hurd uses
Planning Department assessment of what constitutes an "unbuildable" lot
aerial from Merle Johnson showing the broader area containing the buffers
zoning categories other than these five

premiums or other incentives in various locations, perhaps in the buffer area outside the DDA?

Miscellany

The red parcels fill in much of the DDA; still much remains blank

field-checking of all maps is critical.
considering other zoning categories needs to be done--these five are a start (but a big chunk); some of the blank area is not really blank, but is simply in a zoning category different from these five
the database has been spot-checked for duplication; however, a final database will need to have each parcel checked for possible duplication of entries. Duplicate entries may not show in the map but if present, will cause errors in overestimation of the amount of square footage available. If large parcels are duplicated, the error could be substantial.
Checking of paper maps against database needs to be done in a thorough manner; spot-checking has been done.

Similar strategies can be employed on parcels outside the DDA;

Define a 'walk-to-work' zone, using .25 mile buffer (rationale from bus-stop ideas--how far someone will walk to a bus stop)
Planning Staff also provided DDA buffer zones
Numerous folks have commented on sun pattern differences in shadows between north-south and east-west streets.
Kelbaugh has supplied some thinking on setbacks and so forth.

Sources include:

Files provided by City of Ann Arbor Planning Department, Karen Hart and Chandra Hurd
Files and images provided by City of Ann Arbor, Merle Johnson
City of Ann Arbor, Zoning Ordinance Book
Conversation with Doug Kelbaugh
Attendance at task force meeting of February 2, 2004.

Sandra L. Arlinghaus

Solstice: An Electronic Journal of Geography and Mathematics, Institute of Mathematical Geography, Ann Arbor, Michigan.
Volume XV, Number 1.
http://www.InstituteOfMathematicalGeography.org/