Eco-hydromorphic Characterization of the Louisiana Coastal Region Using Multiple Remotely Sensed Data Sources and Analyses

Size: px

Start display at page:

Download "Eco-hydromorphic Characterization of the Louisiana Coastal Region Using Multiple Remotely Sensed Data Sources and Analyses"

Fay McDowell
5 years ago
Views:

1 National Wetlands Research Center Eco-hydromorphic Characterization of the Louisiana Coastal Region Using Multiple Remotely Sensed Data Sources and Analyses 1Holly Beck, 2 Brady Couvillion, 1 Nadine Trahan USGS National Wetlands Research Center 1 Five Rivers Services, Contracted to USGS NWRC, Lafayette, LA, USA 2 U.S. Geological Survey, Baton Rouge, LA USA U.S. Department of the Interior U.S. Geological Survey

4 Photo by: Brady Couvillion

11 Land Area Change in Coastal Louisiana

12 1,883 square mile decrease

15 Hurricane Impacts

16 Hurricane Impacts

19 Model elements Wetland Morphology Team - Land Change/Relative Elevation Module + - Model/Function Input Output Importance: High thick line Medium med. line Low thin line Understanding: High green line Medium blue line Low Red line Predictability: High solid line Medium dashed line Low dotted line Relationship: Positive Negative Non-linear Background Land Change 2010 Land/Water Base Year Elevation % Time Inundated Fragmentation Operation Order Land Loss Model Time 2 Land/Water (initial) Land Change Rates Avg. Band 5 Reflectance Historic Land Change Vegetation Index Fresh water inflow Sediment Distribution Soils/ Bulk Density Natural Delta Sediment supply Volume of Sediment Retained Avg. fill density Nourishment model Time 2 New deltaic Land area (3 rd Stage LW) Bathymetry Max fill extent Combination Model Compaction Base Year Elevation SLR Eustatic Sea level Water level rise Regional subsidence Relative sea level rise Time 2 LW (2 nd Stage) & Elevation - Inundation model Accretion Look-up table Avg. Elev/Bathy Restoration Measures Shoreline Protection Ridge Restoration Diversions Marsh Creation Benefits Calculation/ Spatial Assign. Shoreline Stabilization Hurricane Protection Time 2 Created land area 4 th Stage LW Quality of the data currently available: Good currently available data / information is acceptable Fair need some additional data Poor- need substantial investments to acquire additional data Wetland Fragmentation Upper trophic model Storm Surge model Final Time 2 Land Area Habitat Switching model

20 Landscape Characterization The work I will describe today aims to describe the biophysical structure and dominant processes of the Louisiana coastal landscape as derived from various sources and analyses of remotely sensed data. This multi-parameter approach enables observation and projection of interrelated and cross-scalar processes.

21 Remote Sensing Remotely sensed data, in combination with ground observations, can provide valuable information with regard to many of these form/process associations. The value of remotely sensed datasets is the spatially variable representation of these parameters.

22 Training Data Training data is the most important part of any remotely sensed assessment. It is of vital importance that the training data is accurate (garbage in/ garbage out) In coastal Louisiana, we are fortunate to have an expansive network of monitoring sites.

Coastwide Reference Monitoring System - CRMS funded by CWPPRA - 390

stations - monitoring sites established both inside & outside of CWPPRA

projects - system allows for assessments at project, basin, & ecosystem

23 Coastwide Reference Monitoring System - CRMS funded by CWPPRA CRMS sites established - several thousand environmental monitoring stations - monitoring sites established both inside & outside of CWPPRA project boundaries - many sites serve as control reference areas for projects - system allows for assessments at project, basin, & ecosystem level - system allows for assessments of projects both individually & cumulatively

24 CRMS Website

25 CRMS Website

26 Methodology Generalized diargam of classifcation methodologies utilized. Illustration courtesy of: Joyce Fry - SAIC/USGS EROS Archive and Data Resources Department

27 Ancillary and Remotely Sensed Datasets commonly used as Independent Variables Remotely sensed imagery (e.g. Landsat TM, MODIS) Particular bands (data representative of a specific range of wavelengths light) are often informative about particular parameters as there are distinctive reflection and absorption patterns associated with specific features. Derivations from spectral imagery such as ratios, indices and transformations Elevation data Land Use/Land Cover data Distance to features

28 Landscape Characterization Can include thematic variables such as Land Use/Land Cover, or continuous variable such as bulk density.

29 Landscape Characterization EXAMPLES

30 Landscape Characterization Land Use Land Cover

31 Landscape Characterization Land Use Land Cover Training Data Chabreck/Linscombe Helicopter Surveys (2007) 7289 points (not including other ) 4289 points excluding water 3914 plots excluding forested wetlands

32 Land Use/Land Cover Data

33 Landscape Characterization Elevation Lidar

34 LIDAR

35 Elevation

36 NED (National Elevation Dataset)

37 Classified Topography Results Vermilion Example

38 Classified Topography Results Mississippi River Delta Example

39 Final Coastwide Data Composite DRAFT

40 Topography/Bathymetry

41 Bulk Density

42 Bulk Density

43 Bulk Density

44 Organic Matter Improvements

45 Organic Matter

46 Summary Importance of multi-parameter characterization of the landscape Importance of spatial variability Importance of using training data to the best of your advantage

48 Questions

49 Wetting/Drying Cycles evident in multitemporal imagery

50 Wetting/Drying Cycles evident in multitemporal imagery

Monitoring Coastal Marshes for Persistent Saltwater Intrusion

Monitoring Coastal Marshes for Persistent Saltwater Intrusion Maria Kalcic, 1 Callie Hall, 2 Jeff Russell, 3 and Rose Fletcher 1 1 Science Systems and Applications, Inc., Bldg. 1105, John C. Stennis Space