Flood modelling and management. Glasgow University. 8 September Paul Shaw - GeoVision

Transcription

1 Flood modelling and management Glasgow University 8 September 2004 Paul Shaw - GeoVision

2 How important are heights in flood modelling?

3 Comparison of data collection technologies GPS - Global Positioning Systems Aerial Photography LIDAR - Light Detection and Ranging SAR Synthetic Aperture Radio Detection and Ranging Multi Spectral Satellite Imagery

4 Comparison of data collection technologies For each technology a summary on Process of data collection Potential sources of error Benefits and concerns General comments

5 Comparison of data collection technologies GPS - Global Positioning Systems Aerial Photography LIDAR - Light Detection and Ranging SAR Synthetic Aperture Radio Detection and Ranging Multi Spectral Satellite Imagery

6 GPS Process Use technique called RTK or kinematic Requires base station and rover Requires radio link Software converts to grid and geoid heights Collect mapping data (as points, polylines,polygons and attributes) May require post-plotting using survey infill techniques

7 GPS Potential sources of error Grid transformation and geoid model Link Satellite azimuth, number, quality and spread (measured as DOP s) - Link Radio range Base station problems

8 GPS Benefits Accurate to 10mm Certainty with ground heights Accurate definition of break lines Data collected in useable format Concerns Slow at 0.2 sq km/day Not allowed in all countries Radio strength Foliage and infrastructure No image Access restrictions

9 GPS General comments Many UK companies - Link In GB OS have active and passive stns Link OS trig and BM system redundant NAVSTAR (US) 29 operational at present - Link GLONASS (Russian) - 8 operational at present (Link) GALILEO (European) operational from in total (Link) Started becoming available in the mid nineties Further details on how GPS works at (Link)

10 Comparison of data collection technologies GPS - Global Positioning Systems Aerial Photography LIDAR - Light Detection and Ranging SAR Synthetic Aperture Radio Detection and Ranging Multi Spectral Satellite Imagery

11 Aerial Photography Process Route planning; weather forecasting; flying permission; possible pre-mark; control network Film processing Possible Scanning Photo/Imagery control Aerial triangulation DEM generation; possible vector digitising; post plotting; integration Ortho production/mosaicing

12 Aerial Photography Potential sources of error Crabbing, fore and aft overlap Lens calibration Film quality Verticality Scale problems through large terrain change Forward Motion Photo control errors

13 Benefits Large areas covered in one flight Height generated from the imagery Infrared available Orthophotos 3D visualisations Automated DEM s Aerial Photography Concerns Flying permission required Seasonal Weather dependency Heights to top of foliage Restrictions by government Matching of orthos Automated DEM s

14 Aerial Photography General comments Was used by the EA replaced by LIDAR in 1998 Increasingly sophisticated on-board systems Approx 10 UK companies Fixed Wing or helicopter, hard copy or digital Large archive at larger scales through BKS, Simmons Aerofilms, OS, OSNI and others Getmapping have GB covered at 1:10,000 Link Imagery 25 per sq km at 1:10,000 scale, heights extra

15 Comparison of data collection technologies GPS - Global Positioning Systems Aerial Photography LIDAR - Light Detection and Ranging SAR Synthetic Aperture Radio Detection and Ranging Multi Spectral Satellite Imagery

16 LIDAR

17 LIDAR Process Route planning; weather forecasting; flying permission; control network Processing last reflection DEM generation; Post survey infill Integration

18 LIDAR Potential sources of error include GPS and IMU calibration IMU and laser head calibration Lazer pulse rate GPS baseline length Level of uncertainty with ground height No break line definition Details available at (Link)

19 LIDAR Benefits Accurate to 50mm Large areas covered in one flight (30 sq km in one hour) Can be flown at night Not dependent on sun Water height Concerns No image Weather dependency Flying permission required Processing time up to 4 months for a few days data collection

20 General comments LIDAR 4 UK companies global directory at (Link) QinetiQ have combined with digital cameras to create ATLAS system (Link) General information about LIDAR can be found at (Link and Link) Used extensively by the EA - archive coverage - Link. Catalogue - Link

21

22 Comparison of data collection technologies GPS - Global Positioning Systems Aerial Photography LIDAR - Light Detection and Ranging SAR Synthetic Aperture Radio Detection and Ranging Multi Spectral Satellite Imagery

23 SAR Process (InterMap) Aircraft has 2 antennae, receive slightly different return signals, phase difference determines height Synthetic aperture digitally enhances the data in real time Rigorous QC processing checks DEM DSM by interpolation Intermap SAR description at (Link)

24 SAR Potential sources of error Shadow Rain shadow Signal saturation Missing data Specular reflection DSM generation Details at (Link)

25 Benefits Can cover 100 sq km in a few hours Generates height (up to 0.3m) and image (up to 2.5m) All weather capability SAR Concerns Flying permission required Gives height to first reflection Accuracy of DSM No break lines - Nextmap products are 5 and 10m grid spacing

26 SAR General comments No UK companies Accuracy comparison by EA available at (Link) Nextmap data for UK has 2 accuracy levels 0.6m for the SE and 1.5m for the rest QinetiQ have just gained reseller status Nextmap archive at Link Intermap global archive and prices at Link

27 Comparison of data collection technologies GPS - Global Positioning Systems Aerial Photography LIDAR - Light Detection and Ranging SAR Synthetic Aperture Radio Detection and Ranging Multi Spectral Satellite Imagery

28 Multi Spectral Satellites Imagery Process Order and collect basic imagery Image control points Rectification DEM generation Possible vector digitising Map completion

29 Multi Spectral Satellites Imagery Potential sources of error Cloud Other atmospheric interference Off-nadir angle of separate image Sun Shadow Incorrect sensor calibration details

30 Multi Spectral Satellites Imagery Benefits No restrictions Large archives Large area coverage Repeatability Concerns Problems with automated DEM s Weather dependency Heights to top of foliage/buildings Height accuracy still at research stage

31 Satellites General comments Wide range of sensors with stereo capability On-going program of launch of new satellites Quickbird (Link), IKONOS (Link), SPOT (Link), IRS (Link). Recent research by a Japanese University has shown 0.5m height accuracy possible Link Quickbird archive at Link

32

33

34

35 Comparison Technology Cost/sq km Delivery Accuracy GPS (No image) 5-10k 50 days/sq km new 1cm Aerial Photography 3-5k 50 days new 4cm Global archive LIDAR (No image) 0.35k 120 days new GB archive 15cm SAR 0.064k Global archive 30cm (Hts only) Quickbird 0.03k 50 days new 50cm Min 100 sq km

36 Summary There is an increasing amount of height data for flood modelling. The skill is to select the correct one for a given application. Considerations include:- How accurate do you want the heights? How regular do you want it monitored? What intensity of information do you want? How much money have you got? How quickly do you want it? but...

37 Other considerations Global warming 10-20cm msl rise last century, estimated 50 cm this century - (Link) Glacial Isostatic Adjustment 2 mm per annum (- in South, + in North) (Link). General Info at (Link) OS recently adjusted UK geoid up to 10cm variance Spring tides up to 12m range - (Link and Link) Tidal loading up to 6cm variance in 1 day

38