Global: Ramsar Theme of the Treaty Enforcement
Services Using Earth Observation (TESEO) Project, European Space
Agency (ESA)
Dr. Frank Ahern
Project Partner, TESEO
Atlantis Scientific
Ottawa, Canada
Web: https://www.atlsci.com/TESEO/indexteseo.htm
and http://earth.esa.int/teseo
I. Introduction
TESEO - Treaty Enforcement
Support using Earth Observation - is a project of the European
Space Agency (ESA), which aims at exploring the potential of Earth
Observation (EO) technology to support in the near future the
implementation of international environmental treaties. This activity
will, additionally, be an important ESA contribution to the detailed
definition and early implementation of the Global Monitoring for
Environment and Security (GMES) , an initiative of the EC to coordinate
and expand the use of remote sensing for environmental treaties,
natural disasters, and humanitarian aid. The overall objective
of TESEO consists in exploring the potential of EO, with a focus
on future technology, to support key environmental areas of particular
interest for establishing European policies for the implementation
of the relevant fundamental environmental treaties
- The Ramsar Convention on Wetlands;
- The Kyoto protocol to the UN Framework Convention on Climatic
Change;
- The UN Convention to Combat Desertification; and
- International Convention for the Prevention of Pollution from
Ships (MARPOL 73/78).
The Ramsar TESEO project
is being performed by a team of scientists, technologists and
knowledge managers, led by Atlantis Scientific, Inc. of Ottawa,
Canada. We are researching and developing Earth Observation applications
which can be used for improving support for the management and
sustainable use of wetlands within the context of the Ramsar Convention.
By the end of the project,
the team will have completed:
- Thorough investigation of the information needs of wetlands
managers by working with end users.
- Study of the capacity of EO technology (present and future
satellites, models, algorithms and data fusion techniques) to
fulfill those needs.
- Exploratory studies of the most appropriate tools and techniques.
- Selection of a few prototype products which provide reliable
information in a way that can become operational, and responds
to the users' needs.
- Creation and validation of prototype products.
- Recommendations to the ESA for future satellite missions that
will deliver information of value to this community.
Three test sites are involved
in the study: Mer Bleue in Ontario, Canada; Doņana in Andalucia,
Spain; and Djoudj wetland near the delta of the Senegal River
in Senegal.
2. Description of results
Our team has conducted
an exhaustive review of the national and international bodies
involved in the implementation of Ramsar Convention, and identified
those that may benefit from EO products and services. We identified
the actions, provisions and objectives within the Ramsar Convention
where EO may contribute. We collected and analysed end user requirements
in terms of information products & services. We obtained a
group of end users that expressed their interest in participating
in the project. These include the Estación Biológica
de Doñana (managers of the Doñana wetland in Spain),
National Capital Commission of Canada (managers of the Mer Bleue
wetland) and World Resources International. We entered into a
partnership with the Centre de Suivi Écologique in Senegal
to jointly create and test prototype products of the Djoudj wetland
in Senegal. We reviewed the cost of EO applications compared to
competing sources of information. We created a "user-needs"
web site with a survey asking about information requirements of
wetlands managers that could be fulfilled more conveniently with
EO products. We presented the information needs and products in
an easy-to-read chart showing key selection criteria and recommendations.
Exploitation of EO technology
for wetlands management consists of the following steps:
- Creation of a base map which includes not only the wetland
area, but a regional zone of influence, where human activities
can affect (either beneficially or detrimentally) the natural
functioning of the wetland within the protected area. The base
map will normally contain political boundaries, infrastructure
(roads, railroads, canals, settlements, power distribution networks,
and other relevant infrastructure), political and administrative
boundaries, and a current and accurate depiction of the hydrology
of the wetland and area of influence
- Creation of a baseline inventory. Some of the most important
types of information include land cover (including detail on
vegetation) and land use, at a level of detail adequate to address
the essential management issues.
- Capture of essential information on changes that impact the
management of the wetland. Land cover and land use change are
almost always essential information layers. Many wetlands are
hydrologically dynamic. Information about the extent of open
water and flooded vegetation is key to wetland managers
- Creation of products which "tell the story" to the public,
and to politicians, about the state of the protected wetland,
its recent history, threats to its sustainable ecological functioning,
and constraints imposed by external realities.
- Creation of products that present alternative management scenarios,
and the likely consequences of each scenario. These products
will provide the resource managers and decision makers with
the information they need to weigh competing interests and reach
informed decisions.
We conducted four exploratory
studies aimed at exploiting the potential of recently available
and future space-borne sensor data. In Mer Bleue, Canada, we investigated
the use of multiploarized and fully polarimeteric radar data to
distinguish water extent from inundated vegetation, and classify
gross vegetation types. We also found that radar repeat-pass interferometry
holds more promise than was expected for mapping coherence, which
can indicate gross vegetation characteristics, and for mapping
subtle changes in elevation.
Again at Mer Bleue, we
merged high spatial resolution (1m) panchromatic data (airborne
photography, re-sampled to simulate the IKONOS sensor) with Landsat
colour to produce a highly-interpretable visual product which
can be excellent for baseline mapping, land cover and land use
classification, and change monitoring, particularly vegetation
changes (Figure
1).
At Ebro Delta in Spain,
two exploratory studies were carried out:
- A study of the changes in the delta resulting erosion, deposition,
and sediment transport by marine currents;
- A study of anthropogenic modifications to Canal Vell, an internal
marsh in the Ebro Delta which is ecologically significant as
a breeding area for several species of birds.
These exploratory studies
were particularly important demonstrations of the value of the
30 year archive of Landsat data. The studies showed that Landsat
archived and current data can be used for extremely accurate studies
of trends in ecological conditions, using photointerpretation
and digital classification techniques. A comparison of the advantages
and disadvantages of the various kinds of imagery for Ramsar-specific
needs can be found in Table
1 and Table
2, respectively (this requires Adobe Acrobat reader).
In terms of relating this
information to the needs of wetland managers, the wetland manager
is used to dealing with the wetland which he or she can visit,
touch, smell, and so on. A photo or an image can be regarded as
a first abstraction from that "real world." A map derived
from that photo (e.g., a topographic or thematic map) is a further
abstraction from the real world. A table of areas derived from
that map is yet another abstraction. As one moves farther away
from the real world, information is in one sense less believable,
and less useful for many applications since not only is some information
lost, but the information retained may not be of direct relevance
to the potential user. This is especially the case for those dealing
with management and monitoring of a resource.
Thus, focusing on the
users' perspective, what we wish to do is create the best possible
first abstraction or image of the real world, which clearly represents
the real world as the user understands it. At this stage, our
literature review suggests that optical and radar imagery combined
with ancillary data (such as those described above) and integrated
in a geographic information system (GIS) is best. Then we want
to ensure that we produce tools with which the user can extract
required information (areas, lengths of ecotones, etc). Based
on past experience we believe that if we skip this first step
and go directly to a derived product farther down the abstraction
chain, the users will be less prone to accept what is produced
and will be less likely to buy into the use of the product, especially
if there are any errors in the first derived products that they
see. We want them first to understand that the image contains
valuable information and then, once they reach that understanding
(which they will), we help them extract it with the tools developed
or drawn from other sources.
3. Conclusions
The team has summarized
the requirements for an operational system with the four adjectives:
Reliable, Robust, Affordable, Repeatable. The project web site
was used as a mechanism for disseminating progress, and also for
posting a user survey in three languages: English, French and
Spanish. We received 13 responses from our web-based survey from
Canada, Greece, Netherlands, U.S., India, Turkey, Malaysia, UK,
Botswana and South Africa. Although the number of responses was
small, the respondents showed enthusiastic interest and unanimous
willingness for further participation. The responses were from
private companies, non-government organizations, National and
local governments, and private consultants. Respondents were mainly
wetlands managers and researchers. The respondents' organizations
are mainly involved with Ramsar sites, either directly managing
wetlands or in inventorying wetlands.
In answer to the question
about types of information they need, survey respondents identified
the following main categories of information:
- Identification and physical description of wetlands.
- Change in vegetation, land use, environmental pressures,
dominant vegetation, invasive species, water quality and quantity,
preferably on a 2-5 year update frequency.
- Water quality information.
We also created a list
of innovative products and services based on EO technology which
respond to the user needs analysis and the survey results. We
then asked our end user collaborators to comment on the priorities
of each product or service from his/her perspective. These products
and services are described in Product Description Sheets.
We evaluated all of the
48 proposed products and services according to the following criteria:
- Technical feasibility for EO based products.
- Priority identified by the questionnaire of our end users.
- Practical advantages.
- Contribution to the needs of the Ramsar Convention.
- Contribution to users (our assessment based on knowledge of
the technology).
- Novelty.
The result of this evaluation
was the selection of the following products which we intend to
prototype at our test sites of Mer Bleue, Doñana and Djoudj:
- Water cover and water-cover change;
- Vegetation cover and vegetation cover change;
- Land use and land-use change;
- Exchange of information with other Ramsar sites and with the
Ramsar Convention Bureau.
A design for these products
was created. Production of prototype products is in progress,
and the prototype products will be shown at COP VIII in Valencia
(November 2002).