In recent decades, human alterations
of the earth's surface have been recognized as one of the planet's most
significant cumulative global environmental changes. Increasing population
and per capita income suggest that this trend will continue in coming
decades. In countries such as the US this process manifests principally
as suburbanization. Yet our understanding of the specific causes of
US suburbanization and associated consequences is limited because we
lack a systematic baseline description of the location, extent, timing,
and rates of land use- and -cover changes. Thus it would be enormously
helpful for researchers (both social and natural scientists), policy-makers,
and other interested parties to be able to access such data.
In September 2006, two geography
professors at Clark University (Colin Polsky and Gil Pontius) recognized
this need when discussing avenues for collaborative research with two
ecologists (Chuck Hopkinson and Wil Wollheim) affiliated with the Plum
Island Ecosystem (PIE) project, which is part of the United States’
National Science Foundation’s (NSF) Long-Term Ecological Research
(LTER) program. That discussion formed the basis for a successful, $1.4M
grant proposal to NSF’s Coupled Natural-Human Systems competition,
submitted in November 2006 and awarded in May 2007. This project uses
the premise described above about suburbanization to examine social
dynamics, ecological dynamics, and coupled socio-ecological dynamics.
However, such investigations cannot progress without a core, integrating
dataset. This User’s Manual
details how our project is producing its core, integrating dataset,
which is a <1m resolution, parcel-level classification of land-use/-cover,
by using remotely-sensed imagery in a collection of towns in the northern
Boston suburbs. Our methods address two types of breakthroughs in mapping.
First, the methods offer fixes for some of the most common problems
in accuracy assessment by recommending novel statistical measurements
and extensions of conventional methods. Second, the methods introduce
techniques of “virtual fieldwork” that have just recently become
possible via very fine-resolution (and no-cost) imagery in virtual globes
such as Google Earth. This Manual shows how virtual fieldwork
can be superior to ground-based fieldwork in terms of money, time and
quality of the science. The overall goal of this
Manual is to catalyze the production of similar datasets by other
researchers in other places. As of October 2008, we are currently working
towards this end with colleagues at three additional LTER sites: Baltimore
Ecosystem Study (BES), Central Arizona-Phoenix (CAP), and Florida Coastal
Everglades (FCE). We thank our colleagues at
the BES LTER project, especially Morgan Grove, Jarlath O’Neill-Dunne,
and Weiqi Zhou. Their assistance and guidance have proven extremely
valuable. The authors of this Manual include, in alphabetical
order, Clark University students Jenner Alpern, Albert Decatur, Nick
Giner, and Rahul Rakshit. Alpern and Decatur will graduate with BA degrees
in May 2009; Giner and Rakshit are PhD students. We deeply appreciate
their dedication, energy and attention to detail. –Colin Polsky and R. Gil Pontius, Jr. This material is based on work supported by the National Science Foundation and under NSF Grant No. BCS-0709685 (Colin Polsky, Principal Investigator). Any opinions, findings, conclusions, or recommendations expressed in this material are those of the authors and do not necessarily reflect those of the National Science Foundation. |  Prof. Colin Polsky  |
|
|
|
Home >
Roots |