Using radar for wetland mapping Soil moisture and wetland classification Slides modified from a presentation by Charlotte Gabrielsen for this class.

Southeast Arizona: Winter wet period From C. Wang et al. 2004, Remote Sensing of Environment 90:178-189

Flooding in Pakistan in 2009: Envisat SAR-based image.

Inundated land under forest canopy in Australia mapped using radar – water and tree trunks cause strong double backscatter. See Australian Wetlands, Rivers and Landscapes Centre for details: (http://www.wetrivers.unsw.edu.au/2010/09/remote-sensing-of-flooding-in-arid-australia/)

Importance of Soil moisture Commercial purposes—agriculture and irrigation Crop-yield: pre-planting moisture; irrigation scheduling; de-nitrification Sediment transport: runoff producing zones Climate studies Meteorology Evapotranspiration: partitioning of available energy into sensible and latent heat exchange Hydrology Rainfall runoff: infiltration rate; water supply Productivity often limited by either nutrient or moisture availability

TRADITIONAL METHODS TO MEASURE SOIL MOISTURE Feel method Electrical resistance Neutron probe Soil tension Neutron probes contain a pellet of americium-241 and beryllium. Am-241 emits alpha particles, which collide with Beryllium and produce fast neutrons that return slow neutrons to sensor when they collide with H, which is mostly the H in water molecules. http://cals.arizona.edu

Dielectric constant Dielectric constant: Measure of the capacity of a material to store a charge from an applied electromagnetic field and then transmit the energy The higher the dielectric constant, the less permeable a material is to radar signal The more water there is in soil, the stronger the return signal (all other things being equal)

Dielectric constants for common materials From Ken’s lecture presentation Davis and Annan 1989; Geological Survey Systems 1987

Dielectric Constant and effect of soil type Soil dielectric constant is directly proportional to its moisture content De Jeu et al. 2008 Relationship between soil dielectric constant and soil moisture; relationship is almost linear, except at low moisture contents Variation of dielectric constant as a function of moisture content in sand and clay

soil moisture and radar signal (backscatter) Radar signal has a linear relation with soil moisture Zribi et al. 2005

Moisture content Moisture increases dielectric constant Identical materials can look different in radar imagery due to differences in moisture content Brightness (reflectance) of plants and surfaces increase with moisture content Radar can penetrate deeply into dry materials but not into wet materials (quantifies 0 – 15 cm moisture)

Mapping soil Moisture Timing is important because soil moisture is very dynamic Environmental conditions and precipitation history are important Can’t estimate in frozen soil Depth of penetration depends on soil moisture and incidence angle and is important to understand for modeling

SMAP: soil moisture active passive (Launched Jan. 31, 2015!) Provided global soil moisture and freeze thaw info with 2-3 day return time at coarse (1-3 km) spatial rez. Radar failed July 2015. NASA looking for replacement Passive microwaves are emitted by materials on earth (just like FIR), but they are low energy so need big pixels to get good SNR. Passive microwave emission allows measurement of temperature. Active microwaves can have smaller pixels because higher signal and can indicate water. http://smap.jpl.nasa.gov

detection of wetlands Soil moisture is a good indicator of wetland presence High temporal resolution radar can help characterize wetland hydroperiod

Remote sensing of wetlands Wetlands defined by… Presence of water at, above, or near ground surface Hydric soils Vegetation species adapted to or tolerant of wet soil Remote sensing can be used to map and monitor two of the defining wetland features, vegetation type and surface water/wet soils Historically, wetlands have been one of the most difficult ecosystems to classify using remotely sensed data

Wetland ecosystem services Flood control Groundwater replenishment Sediment and nutrient retention and export Water purification Reservoirs of biodiversity Recreation and tourism Climate change mitigation and adaptation Wetlands and climate change – flood control, carbon storage (peat), etc.

Traditional methods for wetland mapping (NWI) Aerial photography interpretation or single date optical imagery common source for wetland maps Err more by omission than commission May not represent dynamic nature of wetlands www.owrb.ok.gov

Traditional methods for wetland mapping NWI often underestimates proportion of water on the landscape Halabisky et al. 2011

Sar wavelengths Commercially available SAR satellites: L-band (~23 cm wavelength) Useful for mapping forested and high biomass herbaceous wetlands C-band (~5.7 cm wavelength) Most useful in forests during leaf-off condition and for sparse canopies X-band (~3.5 cm wavelength) Useful for detection of water level changes

sar Polarization Horizontal send and receive (HH) polarization is most useful for detecting wetlands Cross polarizations (HV) can discriminate woody versus herbaceous vegetation types due to their sensitivity to biomass Vertical send and receive (VV) polarization is sensitive to soil moisture and flood condition

Surface inundation (flooding) Can increase radar backscatter in forests “Double-bounce” scatter: energy emitted from sensor bounces off the water surface away from the sensor and then off a tree trunk back to sensor But can decrease backscatter in open areas due to specular reflectance

Scattering of c-band energy Bourgeau-Chavez et al. 2009

Radar and optical data fusion Can improve wetland characterization by combining multiple remotely sensed data types Uses multiple sensors Multi-temporal data Fusion of sensors operating in different frequencies (thermal, optical, lidar, radar, infrared) measure various aspects of wetlands for improved mapping accuracy

Corcoran et al. 2011

Study area: Plains and prairie pothole region, U.S.A www.plainsandprairiepotholeslcc.org

Stressors in the PPR Climate change Ag/Development Truncated hydroperiod Reduced connectivity Images: deq.mt.gov

Prairie pothole region research Using a combination of optical and radar data to map wetlands in wet and dry years to obtain information about wetland hydroperiod For ephemeral wetlands, soil moisture is important By determining wetland extent, aim to predict hydroperiod and wetland location for future climate change scenarios Projections can be used to support ecological and landscape-based vulnerability assessments and climate change adaptation planning.

summary Radar data has broad application for the detection of soil moisture and mapping. Fusion of radar data with optical data can improve map accuracy. Use of radar sensors for wetland mapping has implications for an operational wetland hydrology monitoring solution.