Environmental Engineering Reference
In-Depth Information
ecologically meaningful information and mapping it over large
areas using both direct and indirect approaches, and reusing the
data for prioritizing areas for biodiversity conservation in urban
planning and management contexts.
One example of a modeling approach using remote sensor
data for mapping biodiversity components is the modeling of
the spatial distribution of Natura 2000 habitats across Europe
(Mucher et al ., 2009). The Natura 2000 sites are protected under
the Habitats Directive (92/43/CEE) and represent habitats with
high biodiversity values; the spatial distribution of these habitats
outside the protected areas was, however, unknown. Datasets
derived from remote sensor data, such as CORINE land cover
(European Environment Agency, 2000) and GTOPO30 (NASA,
2009) were used in combination with other data, such as the
distributions of indicator plant species. The data were combined
through spatial distribution modeling using expert knowledge
and machine-learning methods. The result was a series of maps
of 27 rather specific habitats across Europe; these maps are very
valuable for biodiversity studies, including urbanization studies.
Recently, spatial prioritizing models have been used in urban
areas to identify urban forests that are of high priority for conser-
vation and species richness (De Wan et al ., 2009). In Australia,
prioritization methods have incorporated species-specific con-
nectivity into multi-species conservation planning within the
urban context (Gordon et al ., 2009). These prioritization tools
could be used for strategic decision-making by land-use planners.
In the Stockholm region, long-term research has been con-
ducted concerning methods for integrating biodiversity issues
into urban planning and management, ranging from mapping
detailed vegetation types and habitat structures (Lofvenhaft,
Bjorn and Ihse, 2002) to spatial ecological modeling and the
development of planning process tools (Lofvenhaft, Runborg
and Sjogren-Gulve, 2004; Gontier, Balfors and Mortberg 2006;
Mortberg, Balfors and Knol 2007; Gontier, Mortberg and Balfors,
2010; Zetterberg, Mortberg and Balfors, 2010).
These methods have been applied in case studies in several
real-world planning situations within the region (Mortberg and
Ihse, 2006; Mortberg, Zetterberg and Balfors, 2009; Mortberg
et al ., 2010a). All case studies used remote sensor data (in
combination with other data) for spatially complete coverage of
vegetation and land cover parameters. Vegetation data derived
by manual interpretation of CIR aerial photographs (Lofvenhaft,
Bjorn and Ihse, 2002), covering the Stockholm municipality and
including the city center, were used in two of the case studies.
The other three case studies covered larger areas and therefore
had to rely on less detailed data derived from the classification of
Landsat images (Holmgren et al ., 2000).
One of the case studies concerned Hanveden, a large forest-
dominated suburban and periurban area in south Stockholm.
Three municipalities, Haninge, Huddinge and Botkyrka, are
responsible for the area. Furthermore, large areas are owned by a
fourth municipality, the city of Stockholm. Part of the rationale
of the study was concern about policy changes imposed by this
big land owner. The overall aim of the project was to create a
plan and a common strategy for nature conservation, recreation
and multi-objective forestry within Hanveden. With respect to
planning, there were three main scenarios: urban exploitation of
land parcels; intensive commercial forestry; and a management
plan for thewholeHanveden areawithbiodiversity and recreation
objectives.
The biodiversity targets were derived via a participatory pro-
cess, involving stakeholders, and aimed at finding priority habitat
types and ecological profiles for the study area. The targeted
habitat types were coniferous forest and deciduous forest with
several ecological profiles for each. Habitat networks for the
priority ecological profiles were derived through GIS-based habi-
tat modeling, using both empirical models (Gontier, Balfors
and Mortberg, 2006, Mortberg, Balfors and Knol 2007; Gon-
tier, Mortberg and Balfors, 2010) and expert models (Mortberg
et al ., 2010a; Zetterberg, Mortberg and Balfors, 2010). The study
resulted in habitat networks that were composed of reproduction
areas and connectivity zones for invertebrates, as well as areas
likely to supply suitable conditions for breeding birds (Fig. 20.4).
The results were used to evaluate the scenarios. The urban
exploitation scenario would have strong negative impacts on
biodiversity targets. The forest management scenarios examined
forest growth 40 years ahead with different management plans
for different forest stands (see Fig. 20.4). The commercial forestry
scenario would have relatively strong negative impacts on conif-
erous forest targets, given that hardwood deciduous forest was
conserved in this scenario. The scenario with forest management
adapted to biodiversity needs would have positive effects on
biodiversity targets, as expected. The results provided decision
support for joint strategies for the municipalities with respect to
multi-purpose forestry with integrated biodiversity, recreation
and economic objectives. The method has great potential for
incorporating biodiversity issues into the creation and evalua-
tion of development and management scenarios. Furthermore,
the study indicated that there may be particular opportunities for
management to enhance biodiversity close to urban areas that
may not be available elsewhere.
Conclusions
Urbanization hasmajor effects on biodiversity and this, in combi-
nation with the tendency for cities to be leading centers of policy
and innovation, means that implementation of, for example,
biodiversity-informed planning tools, infrastructure guidelines,
education and public awareness will have influence at wider scales
than within the cities themselves. For such planning tools and
guidelines, remote sensing has considerable potential as a source
of information on biodiversity at landscape and regional spatial
scales.
Remote sensor data is increasingly being used in biodiversity
studies, which take advantage of recent rapid advances in data
capture, data interpretation and classification methods for deriv-
ing ecologically meaningful information, and spatial ecological
modeling. Important biodiversity research directions and moni-
toring schemes involve systematic combinations of remote sensor
data and field studies. Furthermore, there are great possibilities
for quantifying the urbanization gradient using remote sensor
data; this will be useful for further research into urban ecosystem
processes and the impacts of urbanization on biodiversity. How-
ever, urban biodiversity studies are far fromusing the full capacity
of these rapidly expanding databases and research directions.
In order to reach the full potential of urban biodiversity
studies, high-quality data need to be readily available, derived
data and new methods effectively communicated and shared,
and the necessary inter- and transdisciplinary research must also
incorporate demographic, social and economic issues. Finally,
for future knowledge of how tomanage urban habitats and assure
Search WWH ::




Custom Search