DEVELOP

1
DEVELOP
A NASA-Sponsored ProgramNASA Ames Research
CenterMountain View, CA Tamarisk Assessment in
Nevada 2004 2005Presented by Dyuti
Senguptaon behalf of the DEVELOP Tamarix Team
Dyuti Sengupta, SJSU Charles Geraci, Univ of
North Dakota Yeva Komandyan, CSULA Karen Cheng,
UC Berkeley Shimon Kolkowitz, Stanford
University Mentors Dr. Jay Skiles, Ms. Cindy
Schmidt NASA Ames Research Center
2
DEVELOP Internship Program

• This student-led / student-run program produces
  earth science-based pilot projects with
  supervision from NASA scientists.
• DEVELOP students use NASA facilities,
  techniques, computers, and technology.
• Research primarily directed towards
  environmental issues, community development,
  management, and/or local policy.

3
DEVELOP 2004 2005 Tamarix

• Project inception March 2004 kick-off meeting in
  Reno
• Outline goals and objectives, partners including
  NRCS, ISFS/USGS, UNR, Pyramid Lake Paiute Tribe
• March fieldwork aided in narrowing project focus.
  Northwestern section of Nevada selected as study
  region for the summer
• Project Goals Vegetation classification focusing
  on Tamarix regions, susceptibility map, and
  biocontrol monitoring

4
Methods

• Field Approach For remote sensing, mostly
  large areas of only Tamarix were considered.
• Tamarix presence/absence polygons were
  collected, with phenological and reproductive
  stages and percent cover being noted.

STUDY REGION
NORTHWESTERN NEVADA Counties Washoe
Pershing Churchill Lyon Mineral
Large areas of Tamarix were main focus in the
field study
5
Methods

• Second field iteration was different from the
  first, evolving with project goals
• Field Work used as the basis for
• Geometric correction of imagery
• Accuracy assessment(s) of imagery
• Field information also used for image
  interpretation and eventual classification

• Landsat Thematic Mapper (TM) imagery was
  selected as the primary source of data for
  objectives
• Two TM Scenes were selected from October 1991,
  1997, and 2003
• To aid in interpretation, imagery from October
  was selected specific to Tamarix phenology (prior
  to leaf drop).

6
Remote Sensing Results

• Basic classification of percentage Tamarix was
  produced based on field information
• Classified map helped in producing more useful
  applications

Land Classification
7
Remote Sensing GIS

• EXAMPLE The incorporation of two GIS layers with
  RS data
• Interpolated water table values (well-depth)
• Elevation values extracted from USGS DEMs

Groundwater Interpolation
Shallow regions
8
Remote Sensing Application
Geographic Stratification

• By masking elevation and interpolated water
  table with classified maps, we arrived at
  possible regions of infestation
• White areas are more likely to fit typical
  Tamarix conditions

9
Remote Sensing Application

• Biocontrol Monitoring

1 pixel 30 sq m
October 1997 pre biocontrol
June 2004 post biocontrol

• Field work data and imagery also yielded an
  opportunity to observe of Diorhabda elongata
  defoliation using NDVI (green regions are NDVI
  values gt 0.15, or vegetation).
• Right shows decrease in vegetation, even in
  June, a high vegetation month. Stands are
  primarily Tamarix sp.

10
Customer Follow-up
Goal Based on student project, bring awareness
to state and federal agencies about technology
that may assist in their invasive species
assessments

• Introductory Remote Sensing Workshop held May 6,
  2005 at UNR
• Participants included various state agencies and
  organizations

Nevada Weed Management Areas
State of Nevada
US Fish and Wildlife
Pyramid Lake Paiute Tribe
11
Current and Future Projects

• Verification of 2004 Tamarix classification using
  high resolution satellite imagery
• DEVELOP continues to train students and assist
  state governments and natural resource management
  agencies this summer by working on a cheatgrass
  coverage map
• Utah customers include the BLM, Utah State Dept
  of Natural Resources

12

• Questions?