Change Outstrips Adaptability: Monitoring Land-Use
Y.Q. Wang, Assistant Professor
Department of Natural Resources Science
Y.Q. Wang earned a BS in physics from China's Northeast Normal University, an MS in remote sensing and mapping from the Chinese Academy of Sciences, and MS and PhD degrees from the University of Connecticut, both in natural resources management and engineering. He was an assistant professor at the University of Illinois at Chicago before coming to URI in 1999. He conducts research in terrestrial remote sensing and was selected as one of 59 recipients of the prestigious Presidential Early Career Award for Scientists and Engineers in 2000.
Scientists are often challenged to produce evidence
that the Earth's ecosystems are being degraded. There are three indicators that
can show that ecosystem conditions have deteriorated during the last 30 to 40
years: a rise in atmospheric CO2, a rise in atmospheric
NOx, and changes in land-use patterns. But it is land-use change and the loss
of natural habitat that will have the greatest impact on biodiversity and is
most likely to be irreversible. Biological diversity, or biodiversity, comprises
myriad species of living things on Earth, their genetic material, and the ecosystems
in which they live. The rapid, anthropogenic loss of biodiversity is one of
the most important global environmental concerns today. Many habitats are disappearing
quickly. If biological diversity were evenly distributed, then the latest satellite
information would be sufficient to create better plans for conservation of natural
areas. But biological diversity is not evenly distributed. With remote sensing
technologies, we will be able to project where likely stresses will occur based
on land-use patterns of the past and where our efforts over the next decade
will do the most good. Then we can try to reverse the changes in land use that
are most harmful to biodiversity.
With the Chicago Wilderness, a partnership of
more than 90 organizations united to protect, restore, and manage natural lands,
I used Landsat satellite data in support of the development of a Regional Biodiversity
Recovery Plan for the metropolitan Chicago area. In the Department of Natural
Resources Science at URI, I have been working with satellite data to study the
impact of land-use and land-cover changes on the forest ecosystem in Rhode Island.
In the metropolitan Chicago area, unplanned urban
growth and suburban sprawl are among the main contributors to accelerated degradation
of natural communities, and to the deterioration of human living conditions.
Because the remaining natural areas of highest integrity are small and isolated,
their long-term viability depends on proper management of the surrounding environments.
Satellite images enable us to analyze threats to natural areas and the efficacy
of conservation measures across space and across time. Our primary tools are
vegetation maps with sufficient detail to register changes in land use and land
cover during the past 25 years and to highlight the most severe threats to a
region's natural remnants. With some indication of what growth patterns are
likely to be in the future, perhaps we can intervene before change outstrips
adaptability.
The Landsat Program is the longest running enterprise
for the acquisition of imagery of Earth from space. The first Landsat satellite
was launched in 1972; the most recent, Landsat 7, was launched in April 1999.
The instruments on Landsat satellites have acquired millions of images. The
images, archived at Landsat receiving stations around the world, are an important
resource for global change research. The multispectral capabilities of Landsat
remotely sensed data allow observation and measurement of biophysical characteristics,
while the multitemporal and multisensor capabilities track changes in these
characteristics over time. In this project, we used Landsat images acquired
on October 28, 1972; May 31, 1985; and October 10, 1997 (see figure). To obtain
land cover information, we initially selected 150 samples of land-cover types
as training areas for supervised classification. After running the classification,
land managers and ecologists verified the results and refined the training signatures
as needed to improve accuracy.
From the original land-cover map, we developed
vegetation maps of the region and delineated lands owned and protected by municipal,
federal, and state agencies. We classified the 1972 and 1985 Landsat images
and obtained the land-cover information for corresponding years. These banks
of geographically referenced data make it possible to quantitatively track regional
trends over time.
Dramatic increases in urban land dominate the
changes in land cover detected from 1972 to 1997. Between 1972 and 1985, urban
land increased by 14.5 percent; between 1985 and 1997 by 30 percent; and between
1972 and 1997 by 49 percent. Most of the suburban land expansion came at the
expense of agricultural lands, which experienced a decrease of 37 percent over
the 25 years. Surprisingly, more than one-fifth of natural area was converted
to urban use during that period.
A significant change in land cover has been the
increase in unassociated vegetation (abandoned agricultural fields and degraded
natural lands) in the absence of appropriate management and ecological restoration.
Rampant urban sprawl results not only in wholesale loss of natural and degraded
lands, but also in extreme fragmentation and isolation of the remaining natural
areas in suburban areas.
We completed up-to-date vegetation maps with
sufficient detail to be useful for biodiversity recovery. These maps, and the
baseline land-cover maps from which they were derived, serve as benchmarks against
which we measure progress toward the landscape-level goals of the Recovery Plan.
The Chicago Wilderness project examined changes
in land use and land cover during the past 25 years, and highlighted the most
severe threats to the region's globally important natural remnants. The 25-year
window chosen for the project, and the three temporal "slices" within
it, revealed dramatic land-use trends, commensurate with the coarser scale of
regional resolution. Qualitative comparisons provide striking evidence that
wholesale conversion is not the only process at the landscape level that threatens
the natural communities of the region. Fragmentation, isolation, and the quality
of the ecosystem in which natural lands are located can be equally important.
For example, many protected sites once were surrounded by open space such as
agricultural land, a relatively hospitable environment for some species (e.g.,
grassland birds). Through the accelerated transition of this cover type to urban
structures, the integrity and viability of wild isolates have been compromised.
Quantitative comparison is particularly telling in the case of one of the most
severe threats to regional biodiversity: urban and suburban sprawl and its consequences.
Poorly planned development contributes more heavily
to ecological degradation than does sheer population growth. Between 1970 and
1990, the surface area of metropolitan Chicago's developed land increased by
49 percent while its population grew by four percent. The Northeastern Illinois
Planning Commission projects that the region's population will increase by 25
percent during the next 25 years. At these rates, the accompanying conversion
of open land would be catastrophic. Combined with the visual power of land-cover
maps, the quantitative results of the change-detection project argue strongly
for reforming land-use policies.
A powerful outcome of this work is its potential
as an instrument for conservation. By presenting complex concepts from the Recovery
Plan in striking graphic form, the land-cover maps and quantitative measures
of change speak eloquently to the public.
In another project, funded by the Rhode Island
Agricultural Experiment Station, we are processing and analyzing Landsat remote
sensing data acquired in 1972, 1985, and 1999 for Rhode Island. It is evident
that population growth has increased in coastal regions. Worldwide, 61 percent
of the population lives near a coast. In the United States, more than half of
the population lives in coastal counties, including one of the densest corridors
in the Northeast—Washington, DC to Boston. According to the National Resources
Inventory conducted by the U.S. Department of Agriculture, Rhode Island experienced
a six percent increase in urban and built-up lands in just five years (1992–1997).
Urban growth rates show no signs of slowing. Some of the consequences of urban
sprawl include a loss of natural vegetation and open space, and a decline in
the spatial extent and connectivity of forests, wetlands, wildlife habitat,
and agricultural lands. It is essential to study land-use and land-cover change
and to understand the response of ecosystems to disturbances. With historical
satellite data, we will be able to extract information about what our state
looked like 30 years ago, examine the changes and their consequences, and help
guide land-use decisions that will shape the future.