|
USER COMMUNITY
Currently, in the State of Missouri there are approximately 1000
active mines of which more than 700 are open-pit mining operations.
Due to lack of funding for the Missouri Department of Natural Resources’
(DNR) Land Reclamation Program, the branch in charge of monitoring
all mining activity in the state, no aerial photography has been
acquired for years. This lack of funding has required on-site monitoring
of all mines in the state, performed by a limited staff, on a yearly
basis. But, as a result of minimal staffing, the re-visit period
may extend well beyond 2 to 3 years.
An educational mini-workshop was held to familiarize the partners
with the satellite technologies, its capabilities, and limitations.
This allowed the users to see and therefore structure their needs
based on the resolutions that we were proposing to work. These included
the ETM and Hyperion platforms. From this and subsequent meetings
several areas for potential application of the imagery data to possibly
address their needs in monitoring the extent, health, condition,
and possible age of the mining site were determined or eliminated.
For example, the monitoring of the extent of a mine operation with
these resolution groups was not accurate enough for utility unless
the operation was grossly making infractions of its planned activity.
On the other hand, when viewing the hyperspectral nature of Hyperion
they were interested in knowing if the sensor could pick up acidity
in the soil or some other marker that would allow mitigation or
modification of a reclamation activity. Other areas of interest
include vegetative state and status imagery (tied to permitting
and regulations regarding reclamation of mined lands to a vegetated
state), contextual impacts (streams, vegetation, lakes in the surrounding
environment affected), and historic mining activity discovery from
the Landsat archive (Is the site eligible for certain types of federal
funding for reclamation?).
PRODUCT DEVELOPMENT
Because of their limited knowledge of remote sensing capabilities,
it was determined that an introduction to remote sensing capabilities
was needed to establish a list of viable needs and wants before
the project continued. During the second week in October, the team
from the MO DNR traveled to Columbia for an introduction to remote
sensing capabilities presentation. This presentation was aimed at
giving examples of imagery used and what information their resolutions
would allow us to obtain. In this meeting we explored the possible
uses of Hyperion and Landsat with the group present to develop a
better working knowledge of what would actually help them down the
road. Upon completion of this meeting, we were able to compile a
list of 8 prioritized reclamation sites to study in depth.
Based on these 8 sites, several parameters (e.g. biomass, vegetation
type) were decided upon to study at each. Because each of these
sites is at a different state of reclamation, certain parameters
were deemed more prudent to study. While one site might be near
the end of reclamation, a focus on biomass and vegetation replacement
is necessary. Another site might be farther from reclamation, towards
the end of mining operations, and a focus on soil chemistry would
be more appropriate. In table 1, each of the 8 sites is listed along
with the parameters requested by MO DNR for our focus. For each
of these 8 sites, historic satellite imagery has been gathered in-house
to allow us a temporal look at how each site has changed through
its development. Sensors utilized for this phase include Landsat
MSS, TM, and ETM+.
After importing and building header files for all archival imagery,
each of the 8 study sites were clipped to provide image coverage
over only the necessary area. For each of these scenes, a preliminary
unsupervised classification was performed using isodata classification
in ENVI, and clustered down to 5 classes consisting of water, bare/urban,
cropland, forest, and transitional cropland. The images below in
Figure 2 illustrate the change over one of the study sites, Thomas
Hill Energy Center, from 1980 to 1999.
Utilizing current remote sensing technologies, a monitoring protocol
will be established to effectively prioritize site visits by a small
number of individuals resulting in a more efficient monitoring of
mined land reclamation projects. In support of NASA’s mission
to increase the use of the Earth Observation System (EOS) sensors,
Hyperion hyperspectral data will be used in conjunction with Landsat
ETM+ data. We will also explore the use of Hyperion to garner knowledge
about mine mineralogy, topsoil chemistry, and vegetation establishment
on reclaimed mines. We will also work towards a better understanding
of the intended synergy between Landsat and Hyperion as they acquire
imagery simultaneously. Work on change detection will also be discussed
to highlight what similarities exist between both sensors in regards
to visibility and classification accuracy.
Targeted Imagery products for Synergy IV Mined Land Assessment
and Monitoring
1. Imagery showing potential acidity in the soil (or some other
marker) that would allow mitigation or modification of a reclamation
activity. (Hyperion)
2. Imagery showing vegetative state and status imagery (tied to
permitting and regulations regarding reclamation of mined lands
to a vegetated state), (ETM)
3. Classified imagery showing contextual impacts (streams, vegetation,
lakes in the surrounding environment affected), Some interest in
examining the lead-mining areas in Missouri due to its toxicity
and long term affects on humans. (Hyperion)
4. Historic imagery showing active mine sites prior to EPA mining
regulations from the Landsat archive so as to determine if the site
eligible for certain types of federal funding for reclamation. (MSS)
Figure2: Preliminary classification of Thomas Hill Energy Center
using isodata unsupervised classification.
EXPERIENCE
USER COMMUNITY
An introductory meeting was held with the Missouri Department of
Natural Resources (MO DNR) to identify needs and wants for the initial
phase of the project. In this meeting, the following key ideas were
listed. These monitoring ideas were raw and based solely on their
limited knowledge of remote sensing capabilities.
o Key Dates to study would be:
o Early 1970’s – EPA formation
o 1990 – Law Change in the State
o 2002 – Current activities
o Monitoring Possibilities:
o Area of mine expansion
§ To monitor licensed area vs. actual mine area
o Buffer Zones for Streams near Coal Mines
o Track Pit Advancement
§ Topsoil replacement (270 days to complete)
§ Initial seeding dates for cover replacement
o Road Construction
o Roughness Monitoring
§ Graded vs. Non-Graded
o Soil Chemistry
§ Acidity
§ Metals Content
o Prove Abandoned Land
Because of their limited knowledge of remote sensing capabilities,
it was determined that an introduction to remote sensing capabilities
was needed to establish a list of viable needs and wants before
the project continued. During the second week in October, the team
from the MO DNR traveled to Columbia for an introduction to remote
sensing capabilities presentation. This presentation was aimed at
giving examples of imagery used and what information their resolutions
would allow us to obtain. In this meeting we explored the possible
uses of Hyperion and Landsat with the group present to develop a
better working knowledge of what would actually help them down the
road. Upon completion of this meeting, we were able to compile a
list of 8 prioritized reclamation sites to study in depth. Figure
1 illustrates the geographic location of these sites throughout
the State of Missouri.
Based on these 8 sites, several parameters (e.g. biomass, vegetation
type) were decided upon to study at each. Because each of these
sites is at a different state of reclamation, certain parameters
were deemed more prudent to study. While one site might be near
the end of reclamation, a focus on biomass and vegetation replacement
is necessary. Another site might be farther from reclamation, towards
the end of mining operations, and a focus on soil chemistry would
be more appropriate. In table 1, each of the 8 sites is listed along
with the parameters requested by MO DNR for our focus. For each
of these 8 sites, historic satellite imagery has been gathered in-house
to allow us a temporal look at how each site has changed through
its development. Sensors utilized for this phase include Landsat
MSS, TM, and ETM+.
After importing and building header files for all archival imagery,
each of the 8 study sites were clipped to provide image coverage
over only the necessary area. For each of these scenes, a preliminary
unsupervised classification was performed using isodata classification
in ENVI, and clustered down to 5 classes consisting of water, bare/urban,
cropland, forest, and transitional cropland. The images below in
Figure 2 illustrate the change over one of the study sites, Thomas
Hill Energy Center, from 1980 to 1999.
Utilizing current remote sensing technologies, a monitoring protocol
will be established to effectively prioritize site visits by a small
number of individuals resulting in a more efficient monitoring of
mined land reclamation projects. In support of NASA’s mission
to increase the use of the Earth Observation System (EOS) sensors,
Hyperion hyperspectral data will be used in conjunction with Landsat
ETM+ data. We will also explore the use of Hyperion to garner knowledge
about mine mineralogy, topsoil chemistry, and vegetation establishment
on reclaimed mines. We will also work towards a better understanding
of the intended synergy between Landsat and Hyperion as they acquire
imagery simultaneously. Work on change detection will also be discussed
to highlight what similarities exist between both sensors in regards
to visibility and classification accuracy.
|
