The Best Project
A Breakdown of the Neighborhood
The easternmost point of Inman Park is abutted by Little Five Points, which has been called Atlanta’s Haight-Ashbury, with its bars, restaurants, independent record stores and clothing shops (as well as an American Apparel), the Variety Playhouse music venue, and a transient population that shares sidewalks with curious suburban teenagers and city dwellers. The northernmost properties of Inman Park run along picturesque, pedestrian- and bike-friendly Freedom Park, as well as the Carter Center property. To the west, construction and revitalization is constant, with new condos, restaurants and other commercial properties, and development of the Beltline pushing the limits of property value and land purpose. The southern border runs along traffic-heavy DeKalb Avenue and, just south of that, industrial and MARTA train tracks. A green corridor of parks and walking trails divides the neighborhood. Springvale Park, within Inman Park and divided by Euclid Avenue, was designed by the sons of Frederick Olmsted and before that was the site of a key Civil War battle in Atlanta.
Amid this diverse landscape, climbing and barking among the lumbering foliage of oaks and pecans and magnolia, lines and fences, straying from the vacant industrial construction zones, the Eastern gray squirrel has continued to thrive – albeit with a few extra worries. Living in Inman Park, a squirrel is still susceptible to those old-acquaintance predators the hawk, the falcon, and the owl, which regularly patrol the area skies. But in a “civilized” urban environment like Inman Park, the squirrel must also contend with humans and their byproducts: human developments which destroy food sources, squirrel-zapping power lines, predatory canines and felines, and (perhaps of most concern to squirrels) killer automobiles.
It is an attestation to the gray squirrel hardiness and mating prowess that it survives in such an environment. Along with counting the Eastern gray squirrel, the Inman Park Squirrel Census was curious about how the animal made due in this rough-and-tumble environment.
It should also be mentioned that Inman Park was a top choice for the Census because several members of the team live in or near it, and it is not that big – .59 square miles. The Census team wanted an area that could be covered in a few short weeks, on foot, with support from the community, which was received.
Before embarking on the count, a number of approaches to counting squirrels in the urban wild were considered by the Census team.
The most scientific and conclusive that was suggested: setting non-kill traps, capturing all or most of the squirrels in Inman Park, and tagging each and every one – or at least enough to get a sense of the sheer numbers. While this method would no doubt provide the most accurate count, a few factors deterred the Census team from taking it on. For one, it was a lot of work. It was estimated early on that the number of squirrels in the neighborhood was “in the hundreds,” and the idea of capturing and tagging so many squirrels was, while appealing from the storytelling standpoint, not realistic. There were also concerns that squirrels might be injured when trapped, or that Census takers might be injured (bitten, scratched, psychologically tormented) while dealing with trapped squirrels. But the big wrench in this approach was the cold and startling fact that the city of Atlanta requires a permit for citizens to handle squirrels. At least, that’s what the Census team heard from pretty reliable sources.
Let’s assume it is true. Clearly, this permit deal is often ignored. One Census taker, in fact, knew a boy in nearby Candler Park who kept a juvenile squirrel as a pet. And the Census team, during their work, came into contact with a much more vigilant and violent populace of Inman Park – the squirrel haters – who bragged about shooting squirrels out of their trees (as if the squirrels know which humans own which trees), trying to run them down in cars, even firing bottle rockets at them. Perhaps these are the people for whom the supposed permit deal is written. Or perhaps the law is written to keep people from performing wildlife censuses in the city. Either way, for this reason and others noted, the Census team decided against the capture-and-tag method.
Another counting methodology suggested (brought up by several people, surprisingly): tagging squirrels without capturing them but instead by shooting them with paintballs. This approach conjures more questions than it probably deserves. For instance, hunting braggadocio aside, would it really be that easy to shoot squirrels with paintball pellets, particularly when they are in a tree, 100 feet above, lying flat on a branch, in high wind?
And, would a paintball traveling at a welt-producing velocity accidentally kill a squirrel that a Census taker was only intending to count? And, assuming a squirrel was shot but not killed, would the paint left on its hide remain there until all the other squirrels in Inman Park were shot by paintballs, so that an accurate count could then be performed? And, how many paintballs and paintball guns would be needed for such an enterprise? And, of course, would Inman Park residents think it was a-okay for people to walk around their neighborhood carrying assault-weapon-style guns? This idea was shelved.
Ultimately, the Census team settled on a methodology inspired by other science-based wildlife counts, including squirrel counts in North Carolina and Great Britain.
How It Worked
- Using a map of the Inman Park neighborhood, the Census team divided the area into walk-able 100-meter-by-100-meter hectares, or quadrants, using alphabetical and numeric coding to give each quadrant a grid point reference (i.e., A-1, H-3, S-14, and so on). The idea was, Census takers would count the number of squirrels in each quadrant, one hectare at a time. A total of 151 quadrants were created in this exercise, and an official Inman Park Squirrel Census Map was created.
- A test count was performed by the Census team and volunteers in November of 2011. Carrying maps of their assigned quadrants and tally sheets, the Census takers spent a weekend walking the neighborhood and counting the number of squirrels they saw in a total of 35 hectares, taking 20-30 minutes in each location during squirrel peak-activity times. Problems with the tally sheet were noted. Challenges of counting squirrels were discovered. Things were fine-tuned.
- It was decided that the actual Inman Park Squirrel Census would take place in the spring of 2012. During that period, Census takers would count squirrels in each and every quadrant during the most active periods of mid-morning and late afternoon/early evening. In other words, two counts would be performed for every quadrant, a total of 302 counts, over the course of several weeks.
- On April 1, 2012, Inman Park Squirrel Census headquarters was set up at a rented home in Inman Park. The official map was pinned to one wall. Tally sheets, pencils, clipboards, paper badges that read “Inman Park Squirrel Census” with the silhouette of a squirrel, and bottled water were handed Census takers. And the Census commenced.
How Did the Census Team Know It Was Not Counting the Same Squirrel More Than Once?
When the Census team told the curious about the Census, one of the most common questions was, “Why are you doing this?” (See above.) Another common question: “How do you know that you’re not counting the same squirrel?”
Before answering this question, it’s interesting to note the concern with which the question was often asked. This is a world where all kinds of atrocities are committed; and even in the context of “counting squirrels,” there are plenty of reasons to express unease. The Census team wondered, for instance, if these same people who were so worried about a squirrel being counted twice showed the same concern when they passed squirrel road kill. Which is worse, after all – to be counted twice, or to be killed by a car? And yet, here they were, suddenly preoccupied with the idea that a Squirrel Census count might include a few duplicate squirrels, the implication being that if this was the case, the entire Squirrel Census, which they were not taking part in but had only heard about three minutes prior to the question, would be rendered wholly inaccurate and unworthy of their scientific stamp of approval. In short, isn’t it fun to get people concerned about something they never previously considered?
So, how did the Census team know it was not counting the same squirrel twice, three times, four times, beyond? The answer: Actually, the Census team couldn’t say, with 100-percent certainty, whether or not they were counting the same squirrel more than once. However, one of the strengths of the methodology used by the team was that it, if not eliminated, mitigated the chance that duplicate squirrel counts would be performed. This is due to the size of the quadrants, and the time spent in each.
The size: The fact is, squirrels tend to stick pretty close to their nests; they make their nests, in fact, close to their main food sources and don’t have much reason to venture more than a few hundred feet outside of this area. Unless they are looking for a mate. But even then, they don’t have to go far. And the Census team didn’t count during mating season. Which is to say, if you see a squirrel in a particular yard on a particular day, the chances are very low, even nonexistent, that you will see that same squirrel a half a mile away the following day. In other words, the size of the quadrant was large enough to contain the squirrels and all of their movements. Naturally, there were some squirrels that lived right on the border of two or, in extreme cases, three or four quadrants. Were they counted by Census takers in every quadrant? We can’t say for sure.
The time spent in each quadrant: one hectare, 100 meters by 100 meters, is not the smallest parcel of land. It takes about 20-30 minutes to truly navigate it and count the number of squirrels in it. Then it’s time to go. In other words, if a Census taker spent an hour in the hectare, he might come across the same squirrel he had counted before and think, “Did I count this one already?” But in just 20-30 minutes, the chances of that were lowered considerably.
Further, there were rarely so many squirrels in each quadrant that a Census taker lost track of the number of squirrels present. The highest number of squirrels counted in one quadrant was 21. But the mean number of squirrels for each quadrant was 5.7. And it was easy to tell them apart. In fact, let’s completely forget double-counted squirrels. The Census team is certain that many more squirrels were not counted. Squirrels are pretty talented at hiding – in their nests, in trees, inside attics.
Ultimately, the Census aimed to report the “squirrel abundance” of Inman Park – an accurate estimate of the total number of squirrels present during the spring of 2012. To reach this end, the Census team used a well-known density estimation equation first proposed in wildlife counts in, we think, the 1950s. It takes into account possible counting discrepancies to arrive at a final – and accurate, we feel – number.