Pan-European distribution modelling of stream riparian zones based on multi-source Earth Observation data
Introduction
In light of current widespread degradation and threats to freshwater ecosystems, the need for systematic assessment of riverine habitats is well-documented (MEA, 2005, Dudgeon et al., 2006, Malmqvist and Rundle, 2002). At European level this is endorsed by a series of key legislative acts; most importantly the Water Framework Directive (2000/60/EC), which introduced the legal obligation for Member States to assess the ecological conditions of river and adjacent land habitats as a basis for effective management policies. Riverine habitats are gaining an increasingly significant role within the new European biodiversity policy and conservation targets for 2020, which are largely based on the implementation of the new European Green Infrastructure (Sundseth and Silwester, 2009). This multi-faceted concept, developed in the mid 1990s in the US, can be defined as a strategically planned and managed network of natural lands, working landscapes and other open spaces that conserve biodiversity, ecosystem values and functions and provide associated benefits to human population (Benedict and McMahon, 2006).
To support research and monitoring activities related to the abovementioned policy framework, the need for information on extent, distribution and characteristics of the riparian zones (from the Latin ripa, bank) is clear. ‘Riparian zones’ refer to transitional areas occurring between land and freshwater ecosystems, characterized by distinctive hydrology, soil and biotic conditions strongly influenced by the stream water (Naiman et al., 2005, Verry et al., 2004). These are not limited to areas associated with floodplain indicators, but they also include portions of upland away from the bank that have some water-land interaction (Gregory et al., 1991); typically, near-slope zones ecologically connected to the watercourse by surface and subsurface hydrology (NRC, 2002). Fig. 1 provides a schematic representation of a riparian zone and main characteristic functions.
Despite the variety of concepts and definitions present in the literature, there is strong agreement with regard to the importance of the riparian systems due to the natural and social services they provide. Riparian zones can encompass valuable natural habitats and are often characterized by high productivity and biodiversity (Whitaker et al., 2000, Knopf and Samson, 1994). They provide reduction of non-point-nutrient and pollution sources via plant uptake, physical filtering and chemical transformation (e.g. denitrification), together with trapping sediment-bound pollutants and waters coming from upstream (Daniels and Gilliam, 1996, Peterjohn and Correll, 1984). Riparian corridors play a major role in maintaining landscape connectivity, functioning as ‘dispersal corridors’ within fragmented landscapes (Gillies and Cassidy Saint Clair, 2008, Machtans et al., 1996). From a hydrological risk perspective, riparian environments supply river bank stabilization and provide resistance to runoff during flood events (Bennett and Simon, 2004).
The majority of research related to riparian zones has focused on permanent and seasonal watercourses, while few studies have focused on small headwaters and ephemeral tributaries (Goebel et al., 2003). Nevertheless, recent research based on surveys of amphibians and plant communities has claimed that these latter watercourse types can also have discernible riparian zones (Hagan et al., 2006, Perkins and Hunter, 2006). We therefore considered relevant to include in this study not only ‘river-floodplain’ systems (Bayley, 1995), but also the riparian networks of minor and ephemeral watercourses.
To the authors’ knowledge, no continental-scale modelling and mapping of riparian zones has previously been performed; most literature describes local-scale analyses, with a few studies extending to a regional or watershed scale. Sutula et al. (2006) developed a method based on remote-sensing data to map potential riparian zones in Southern California watersheds. Their approach was based on purely geomorphological criteria applied to 10 m and 30 m DEMs, together with superimposed NDVI satellite data to report vegetation distribution. Ivits et al. (2009) considered as riparian zones the regions within a 1 km buffer zone from the river network of Andalusia (Spain), and analysed the permanent vegetation present in relation to EU agri-environmental measures. Much literature focuses on smaller scales (e.g. river basins) and on recent techniques, such as LiDAR, to acquire characteristics of riparian vegetation and topography (Johansen et al., 2010, Goetz, 2006).
The generation of a model to derive riparian zones distribution for the whole of Europe being a primary target of the present work, the proposed approach relies mostly on satellite imagery and continental GIS datasets. A large-scale assessment of this kind faces a series of challenges: (i) establishing a trade-off between processing effort/data availability and spatial resolution due to the vast study area and the generally small size of riparian zones, (ii) the high heterogeneity of these environments (e.g. spectral variance, biota, geomorphological setup), (iii) the need to introduce theoretical criteria in order to include functional considerations (NRC, 2002).
Specific objectives of this research are the development of a model to derive the continental distribution of European stream riparian zones, to report broad spatial patterns, and discuss their basic characteristics.
Section snippets
Materials and methods
The methodology followed is structured in three major sequential steps: (i) the selection in the literature of relevant common descriptors to characterize stream riparian zones; (ii) the production of thematic geospatial datasets as required information layers to represent the riparian descriptors; (iii) the implementation of a riparian detection model based on a fuzzy membership approach. All the methodological steps are described in detail in the following sections.
Overview of distribution and characteristics
The model output provides a distribution map1 and basic characterization of the entire continental riparian network. A representation of the model output is illustrated in Fig. 4; riparian zone distribution is indicated using 10-km cells to improve visual observation of its continental extension. Riparian zone distribution strongly depends on the river network density. The hydrographic system, based mainly on CCM
Conclusions
Information on the spatial distribution of riparian zones is crucial for the assessment of ecological functions of riverine environments (Ward et al., 2002), especially in the context of biodiversity conservation and the evaluation of ecosystem services. A novel model to derive the distribution of stream riparian zones is proposed, considering both stream-riparian ecosystems, related to small size and ephemeral streams, and river-floodplain ecosystems of large flood-prone systems. We
Acknowledgements
The Authors would like to thank L. Feyen and P. Kempeneers (JRC) for providing LISFLOOD and Forest Map 2006 data, respectively. We also acknowledge A.C. Cardoso and M. Dunbar for carefully reviewing the manuscript. Data for validation were kindly provided by M. Dunbar (UK CEH), Prof. N. Prat (FEMR, University of Barcelona), J. Barquin (Cantabria University, Spain), Prof. T. Ferreira and F. Aguiar (Tech. University Lisboa). We acknowledge the Spanish Minister of Science and Innovation in
References (66)
- et al.
Analysis of the conflict between omission and commission in low spatial resolution dichotomic thematic products: the Pareto boundary
Remote Sens. Environ.
(2004) - et al.
The use of riparian forest strips by small mammals in a boreal balsam fir forest
Forest Ecol. Manag.
(2001) - et al.
Parameter optimisation and uncertainty assessment for large-scale streamflow simulation with the LISFLOOD model
J. Hydrol.
(2007) - et al.
Estimating the ecological status and change of riparian zones in Andalusia assessed by multi-temporal AVHHR datasets
Ecol. Indic.
(2009) Mapping riparian vegetation along rivers: old concepts and new methods
Aquat. Bot.
(1997)- et al.
Pan-European forest/non-forest mapping with Landsat ETM+ and CORINE Land Cover 2000 data
ISPRS J. Photogramm.
(2009) - et al.
European hydraulic geometries for continental scale environmental modelling
J. Hydrol.
(2006) - et al.
The relative influence of riparian habitat structure and fish availability on otter Lutra lutra L. sprainting activity in a small Mediterranean catchment
Biol. Conserv.
(1996) The role of science in the preservation of forest biodiversity
Forest Ecol. Manag.
(1999)
Coordinating methodologies for scaling land-cover classifications from site-specific to global: steps toward validating global map products
Remote Sens. Environ.
Fluvial geomorphology: a perspective on current status and methods
Geomorphology
Fuzzy sets
Inform. Control
Automatic spectral rule-based preliminary mapping of calibrated Landsat™ and ETM+ Images
IEEE Geosci. Remote Sens.
Understanding large river-floodplain ecosystems
Bioscience
Green Infrastructure: Linking Landscapes and Communities
Riparian Vegetation and Fluvial Geomorphology
Sediment and chemical load reduction by grass and riparian filters
Soil Sci. Soc. Am. J.
Freshwater biodiversity: importance, status, and conservation challenges
Biol. Rev.
Riparian corridors enhance movement of a forest specialist bird in fragmented tropical forest
Proc. Natl. Acad. Sci. U.S.A.
Remote sensing of riparian buffers: past progress and future prospects
J. Am. Water Resour. Assoc.
An ecosystem perspective on riparian zones focus on links between land and water
Bioscience
Do small headwater streams have a riparian zone defined by plant communities?
Can. J. Forest Res.
Comparison of new and existing global digital elevation models: ASTER G-DEM and SRTM-3
Geophys. Res. Lett.
Cited by (49)
Long-term dynamics of soil, tree stem and ecosystem methane fluxes in a riparian forest
2022, Science of the Total EnvironmentContributions of dry rivers to human well-being: A global review for future research
2021, Ecosystem ServicesWetland extent tools for SDG 6.6.1 reporting from the Satellite-based Wetland Observation Service (SWOS)
2020, Remote Sensing of EnvironmentRelationship between ecological condition and ecosystem services in European rivers, lakes and coastal waters
2019, Science of the Total EnvironmentCitation Excerpt :Natural capacity for erosion prevention can be expressed in terms of riparian land area per unit of stream length (km2/km). For this study we used the riparian land density estimated for the Danube river basin by Vigiak et al. (2016), based on the riparian land mapped by Clerici et al. (2013). The service flow of sediment retention can be expressed as sediment load removal afforded by riparian land, namely as the difference of mean annual sediment yields (t/km2/y) that cross any given reach in the absence and in the presence of riparian land.
A spatially explicit patch model of habitat quality, integrating spatio-structural indicators
2018, Ecological Indicators