Aquatic Case Studies

Dr. Sofia Wikström
AquaBiota Water Research
Svante Arrhenius Väg 21A; S-104 05 Stockholm, Sweden
Phone: +46 8 161012
Email: sofia.wikstrom@aquabiota.se

Resultaten från WP A1 kommer att analyseras vidare med avseende på i vilken utsträckning de utvecklade metoderna ger den noggrannhet och tillförlitlighet som krävs för att de ska kunna användas operationellt. Projektet syftar till att utvärdera metoderna utifrån befintliga behov av kartering och övervakning av akvatiska miljöer, definierade i samarbete med relevanta användare av metoderna. Utvärderingen kommer att genomföras med hjälp av fältinventeringar samt efterföljande analys och tolkning.

On the basis of the methods developed in WP A1, the classification data (delivered from WP A1) will be further examined to determine whether it can produce the degree of specificity and confidence required. The project will develop and validate methods in a process which includes field documentation, subsequent analysis and interpretation. The validation will concern the performance characteristics of the methods related to specifications and requirements of potential users, defined by the current needs for mapping and monitoring in Swedish aquatic ecosystems.

1    Purpose and Scope
The general objective of this work package is to achieve operational methods for the use of airborne bathymetry laser and digital aerial photos for environmental mapping and monitoring in aquatic ecosystems. Specifically, the project will:
    (1) Evaluate the methods and outcomes produced by WP A1 from a user perspective (e.g., the needs of the County Administrative Boards and municipalities) and develop guidelines for the operational use of the methods. The work will be done in close collaboration with stakeholders and users of data, in order to ensure that the developed methods correspond to current needs for mapping and monitoring.
    (2) Specify and refine requirements on the results produced in project WP A1. The evaluation results generated by WP A2 will be continuously fed into the method development in WP A1.
    In addition, project WPA2 will be responsible for collection of the aquatic field inventory data to be used in the programme. 

2    Contribution to the programme aim
The work package will ensure that the methods and data produced can be used in operational scenarios. The classification and mapping methods and subsequent data analysis will be systematically evaluated from a user perspective. The project is a vital part of the aquatic part of the programme in the aspect of ensuring that fully operational automatic or semi-automatic methods will be attained. The evaluation results will also be delivered to WP T1 for the preliminary development of an integrated map in coastal zones.

3    Background, Theory and Methods
A conceptual framework for assessment and monitoring of environmental functions in marine and freshwater areas is needed both for natural and anthropogenic systems. The Swedish national monitoring program as well as monitoring on the European level needs high quality data to identify both the present state and the long term changes including effects of climate change. Data is also needed for the reporting according to a number of EU directives and policies such as the Water Framework Directive, the Flood Directive, the forthcoming Soil Protection Directive and Marine Strategy, and the Habitat Directive, among others.
    The usable depth range for data from airborne laser scanning and digital imagery is similar to the depth range of the euphotic zone, which is the depth that is exposed to sufficient sunlight for photosynthesis to occur. This zone includes high biological values and a dominant part of the species and habitats that have presently been given priority for conservation are found in these shallow areas (Naturvårdsverket 2007). For instance, aquatic vegetation forms important and species-rich habitats that are only found in the euphotic zone. Today, biological investigations of these shallow underwater habitats are mainly performed by experienced scuba divers, sometimes supported by underwater video (Kautsky 1999, 2004). This method provides detailed information on the local habitats and has been shown to be efficient for monitoring of changes in depth distribution and species composition. The major drawback is that the data from diving transects are spatially limited, which restricts their usefulness for mapping and monitoring of distribution changes at a large spatial scale. An increasing number of successful examples indicate the potential of remote sensing techniques to facilitate mapping and monitoring of coastal and freshwater ecosystems, including mapping of habitats and monitoring of water quality, as well as the spread of invasive species (e.g., Andréfouët et al. 2008, Goetz et al. 2008).
    Due to the technical difficulties involved in mapping aquatic habitats, predictive habitat modelling has been used as a complement to field measurements in producing maps of marine habitats and species (e.g., Bekkeby & Isaeus 2008, Isaeus et al. 2007, Naturvårdsverket 2008). However, the quality of predictions from habitat models is frequently restricted by the low quality or resolution of data on bathymetry and bottom substrates (Naturvårdsverket 2008). Recent developments in laser scanning show that use of laser data can be an efficient method for bathymetric mapping of shallow areas and also a promising method to obtain information on seabed type (e.g., hard/soft substrata, presence of vegetation; e.g., Méléder et al. 2007, Piel & Populus 2007, Wang & Philpot 2007). If bathymetric data from laser scanning and possibly also interpreted maps of bottom substratum and vegetation could be integrated in predictive habitat modelling, this could strongly improve mapping of species or values that cannot be seen directly in remotely sensed data.
    Task 1 - Definition of user cases. The project will define and evaluate a series of user cases, in cooperation with users. The user cases will cover defined needs for mapping and monitoring of aquatic environments for the Swedish environmental goals and reporting according to EU directives. Some examples are coastal habitats (e.g. shallow bays, sublittoral sand banks, hard/soft substrata with/without benthic communities, fish spawning habitats) and particular species (e.g. key species or threatened species such as Zostera marina). Initially, a user group will be formed with representatives from the County Administrative Boards and other users, where laser and aerial imagery data is available. The group will give input to the current need of data for mapping and monitoring in their specific region, which will help the identification of; (1) the key parameters and ideas of results for the classification, and (2) requirements on the generated output from WP A1. An initial report will be compiled from this analysis and will serve as a reference for the method development in WP A1 and A2. The report will be revised and updated with input from the user group, SEPA, the reference group and external stakeholders, and in collaboration with WP A1 based on statistical separability of vegetation types.
Task 2 - Collection and interpretation of field data. The main part of the aquatic field inventory data for the EMMA-programme will be provided by County Boards and municipalities, but additional data will be collected within the project. Field surveys will be conducted simultaneously with the collection of additional laser data and aerial imagery (WP A1). Diving and underwater video will be used to collect data on bottom substratum and vegetation. Two separate datasets will be collected, one for development of the classification (input to A1) and one for the habitat modelling (Task 2). Both datasets will include data for training of methods/models and data for subsequent validation. Diving and video inventories will be performed according to the standardised method for monitoring of benthic vegetation on the Swedish east coast, i.e. transect inventories (Kautsky 1999, 2004). Furthermore, the water optical properties (spectral absorption, scattering and backscattering) will be measured with state-of-the-art instrumentation accessible within the project group (Hobilabs c-Beta and Hydroscat) completed with measurements by external subcontractor (e.g. instrumentation Wetlabs AC-9). Spectral reflectivity of species and substrata will be measured both in situ and in laboratory from samples taken at the diver inventories (routinely performed by instrumentation accessible in the project group). The interpretation and processing of the field inventory data (both the data collected within EMMA and other data) will be a part of the method development in collaboration with WP A1. The inclusion of different datasets will allow comparison of different amounts and types of field data for the quality of the final classifications (collaboration with WP A1).
    Task 3 - Integration of laser data and habitat modelling. The aim of this task is to evaluate whether an integration of data from the laser scanning in predictive habitat modelling can improve mapping of species or other features that cannot be identified directly from the laser data or aerial photos. This will include key species such as Potamogeton perfoliatus, Zostera marina, Fucus vesiculosus and Mytilus edulis and potential habitats for fish recruitment. Models will be built (1) with pre-existing maps of bathymetry and bottom substrate and (2) with bathymetry and bottom substrate from the laser measurements. The quality of the models will be evaluated using an external dataset, i.e. field inventory data that was not used to build the models. The habitat modelling will be done with well tested methods for marine habitat modelling (GAM models; Bekkeby & Isaeus 2008, Sandman et al. in press) and development of modelling methodology will not be a part of the proposed project. The result of the model comparison will be presented in a conference paper. Output maps from the models including laser bathymetry and bottom substratum will also be used in the evaluation of the output for end users (Task 4).
Task 4 - Evaluation of output for end users. The user cases will be evaluated in order to explore if the data produced in WP A1 (classified data from laser scanning, aerial photos and underwater video, alone or combined) can be used directly to identify and map features of interest. The project will also define detection limits for taxonomic detail and coverage ratios. The task will be executed in close collaboration with the user group, by regular group discussions where maps and other output from WP A1 and A2 are presented and discussed. The end-state of each of the user cases will be to evaluate if the data produced can be exploited by stakeholders to fulfil the requirements of regional and national conservation management, EU Habitat directive and Natura 2000. This will result in the development of user guidelines for mapping and monitoring of aquatic environments with laser scanning, aerial photos and underwater video, which will be presented in a scientific paper and a report directed towards potential users (collaboration with WP C).

4    Practical Relevance
The project will demonstrate how the integrated use of laser and digital imagery can be used in practice for monitoring and evaluation of environmental functions in marine and freshwater ecosystems. By close co-operation with end-users on national, regional and local levels, the implementation and usage of such inventory and mapping methods will be facilitated. The methods developed in the project will also contribute both to the near-future monitoring needs and to providing data for environmental research. Furthermore, change detection in marine and coastal areas can be identified and used for mitigation and adaptation measures.

5    References

  • Andréfouët, S., Costello, M.J., Rast M and Sathyendranath S. 2008. Earth observations for marine and coastal biodiversity and ecosystems. Remote Sensing of Environment 112: 3297-3299.
  • Bekkby, T. and Isæus M. 2008. Mapping large shallow inlets and bays – modelling a Natura 2000 habitat with digital terrain and wave exposure models. ICES. Journal of Marine Research 65: 238-241.
  • Goetz, S.J., N. Gardiner and Viers J.H.. 2008. Monitoring freshwater, estuarine and near-shore benthic ecosystems with multi-sensor remote sensing: An introduction to the special issue. Remote Sensing of Environment (in press).
  • Isæus, M., Carlén, I., Wibjörn, C. and Hallén S. 2007. Svenska högarna. Marinbiologisk kartläggning och naturvärdesbedömning. Stockholm, Stockholm county board: 50.
  • Kautsky, H. 1999. Miljöövervakning av de vegetationsklädda bottnarna kring Sveriges kuster. Mimeogr.version 2004.05.13, Institutionen för Systemekologi, Stockholms Universitet, 106 91 Stockholm: 33.
  • Kautsky, H. 2004, Handledning för miljöövervakning, Undersökningstyp Vegetationsklädda bottnar, ostkust 1 Version 1: 2004–04-27, Naturvårdsverket: 15.
  • Méléder, V., J. Populus & C. Rollet 2007. Mapping seabed substrata using Lidar remote sensing. MESH guide to marine habitat mapping, IFREMER.
  • Naturvårdsverket 2007: Skydd av marina miljöer med höga naturvärden. Rapport 5739.
  • Naturvårdsverket 2008: Utbredningen av arter och naturtyper på utsjögrund i Östersjön. Rapport 5817.
  • Piel, S. and Populus J.  2007. Recommended operating guidelines (ROG) for LiDAR surveys. MESH guide to marine habitat mapping, IFREMER,
  • Sandman A., Isaeus M., Bergström U. and Kautsky H 2008. Spatial predictions of Baltic phytobenthic communities: Measuring robustness of Generalized Additive Models based on transect data. Journal of Marine Systems, in press.
  • Wang, C.K. and Philpot W.D. 2007. Using airborne bathymetric lidar to detect bottom type variation in shallow waters. Remote Sensing of Environment 106:123–135.