Biodiversity Discussion Group
August 21, 2005
Frank Muller-Karger and Hank Shugart
Rapporteurs: Chris Moses and
Overview
H.H. Shugart
Overview of terrestrial ecosystems and fundamental concepts.
Scale:
· Critical issue in terrestrial ecosystems.
· The finer the scale, the more it is determined by plant-level processes
· At larger scale controlled by landscape processes.
· Scale at which transition takes place between two varies with climate.
· This transition zone has implication for sensors’ ability to detect underlying processes.
Structure:
· Structure of vegetation dominates pattern.
· E.g. Mature forest landscape is a mosaic of different age stands.
· Dynamics at plot level are chaotic (various successional stages through time), but at landscape level one finds equilibrium of mixed stand ages.
· In terms of sensors, are moving towards radar and lidar to get a sense of vertical structure of forests.
Questions:
· So, how do we integrate existing technologies?
Modelling:
· Forest simulators range from fine scale interactive mosaic models to non-interactive landscape models that do not incorporate factors such as dispersal. Coarsest scale are homogenous models.
· Need for richer models seen through the variety of structures in forests types with similar LAIs. Same LAI can have forests with radically different respiration, and thus implications for carbon cycling.
F. Muller-Karger
Overview:
Ocean is a habitat that covers ~71% of the globe and is also about 4 km in the vertical dimension on average. Biological systems of ocean generate half of global net primary production. Marine ecosystems are critical to biodiversity, human and ocean health.
Concepts being dealt with in the field:
1) Ecosystem functioning
- Scale, abiotic/biotic interactions
- How to characterize biodiversity
2) Climate change impacts
3) Coastal development and resource use
4) Ecosystem-based management
- Zonation
- Multiple use of areas
Technological Issues
Marine ecology is observation-poor, and poor in modeling paradigms and integrating models with remote sensing.
A number of questions to answer regarding remote sensing needs.
[See slides]
Biodiversity research goals and objectives/future directions:
1) Requirements
2) Observing strategies
3) NASA and CBD
4) Research to operations
Marine Discussion:
F. Muller-Karger
Tiffany Moisan—NASA Goddard
Much of terrestrial realm deals with conservation—should marine area focus on this need as well? Should NASA science help in defining marine protected areas?
Comment: Nature Conservancy is interested in remote sensing of marine protected areas.
Answer: F. Muller-Karger
Less than 1% of EEZ is protected …in most countries, MPA's are just paper parks…how do we make fewer more relevant.
Remote sensing of nearshore areas where MPA's are largely defined is difficult. One concern that the Landsat spectral or spatial resolution or bandset is limited for much aquatic work and understand bottom of ocean factors, such as coral reef attachment. NASA should design a sensor that can deal with this.
Comment: NASA should change specs for LDCM and add an extra blue and blue-green band.
Answer: Question has been raised, but answer uncertain—how many more bands.
Comment (unidentified): Need to deal with four functional components of continental shelf.
2. Don’t know if get more by having more functional groups. Recommend process studies to address simple questions—such as how complex models have to be to address carbon flux—no idea what this is in open ocean.
Oscar Schofield – spatial resolution is key…detritus, phytoplankton, etc. need to be distinguished over continental shelves
F. Muller— NPOESS has no tool fro separating chlorophyll and colored dissolved organi matter because the VIIRS 1 km unit has no bands for this…need those bands and better resolution. Natural fluorescence bands of MODIS (fluorescence line height or FLH) taken out of VIIRS—so don’t have this tool now. Ideally, managers need 30 m resolution VIS imagery for coastal zone management, and a slightly rescoped LDCM would revolutionize coastal resource management and operations nationally and internationally.
Comment (unidentified): Need to look at species effects as well—e.g. keystone effect.
Oscar ?? – is it better to look at phyotplankton functional groups, or ecosystem based management…recommend process studies for simple questions…how complex do models need to be to accurately address predictability of carbon on continental shelves?
TM – species concepts…no beginning no end… it’s all tied together…in the marine we need species concepts
Comment (unidentified): Need to get physics down, but need to get biology as well. Assimilation of satellite data into models is difficult at large scale, not really been tried out in biological modeling at small scales. Spatial resolution is limiting factor in coastal zone. Time frequency of satellite observations may need to increase to match spatial resolution…maybe better than daily
Reply F. Muller-Karger: NOAA and NSF are initiating plans for regional and global observatories. Can satellite systems be planned as integrated to these systems? Can they be linked with observatories in situ? Doesn’t see planning of these observatories to link with satellite sensors. Worry about planning strategy.
Response (unidentified): Don’t know how to assimilate large scale sat data to small scale.
F. Muller-Karger: So, are we asking NASA to play a more active role?
Reply: Yes, to play a more active role in getting involved with planning of observing systems, and better planning and coordination between agencies.
Woody: Like: 1) Focus on why you want to do something. 2) Pointing out where process studies are needed and not just asking for everything in sensors. 3) Looking for opportunities as to whom we can work with (NSF, NOAA, etc.) Job is beyond one agency’s means to achieve.
F. Muller-Karger: Can we communicate to the public a bit better about what NASA is all about i.t.o. earth observation? NASA is effectively only known for moon and mars…should get some PR for oceans and terrestrial biodiversity. Remote sensing is way to integrate observation over all earth.
Woody Turner: Earth science community’s involvement is crucial to getting NASA support. Should push hard to make things better than what they are—even though situation is grim now. Think boldly, despite budget cut situation of NASA. Need to push LDCM and NPOESS to get what we want.
Mandy Cooper /U. Central Florida: Might be beneficial to reach out to traditional biologists/ecologists who aren’t familiar with remote sensing through NSF to get them more involved with remote sensing.
Woody Turner: Thinks that there is some opportunity for joint solicitations with NSF, a more baseline bio/eco science agency.
Bob Knox – technology…Band synthesis from different sensors can work to provide continuity between sensors. On-board processing is improving. It is also to build an imaging spectrometer than it is a multi-spectral sensor of 30 bands.
Woody: Expected people to advocate hyperspectral and lidar/radar sensors.
Stuart Pimm: Thinks NASA will not allow for new sensors if current mission focus on Mars remains. Advocates putting pressure/discussing issues with representatives.
Larry Dysher - Ocean Imaging: Higher spatial resolution gives better performance in ocean environment than higher spectral resolution. Blue bands are key.
F. Muller-Karger: Need two or three visible bands to revolutionize how coastal management is done. can we tweak LDCM and NPOESS to extend utility in coastal environment in the coastal zone management
Aron Moody UNC – NASA could dedicate an “instrument developer” program where engeneers work with scientists regularly to develop what is needed
F. Muller-Karger: Have focused very little on applications. Impression is that NASA applications program is focused heavily on federal partners. Can this be extended to academic or commercial partners.
Woody Turner: Yes, this is true, but can provide wider coverage by going through federal partners, which is why they have gone through this.
F. Muller-Karger: Comments that we are not using the larger infrastructure of the country represented in universities, people outside government to address the nation’s needs. NASA can help coordinate this large infrastructure better.
Woody Turner: Are trying to get away from bottom-up approach, but can expand where opportunities exist beyond federal partners.
Terrestrial Issues:
Stuart Pimm: Issues as relate to CBD. High extinction rate. Bird conservation has reduced that extinction rate several fold, although CBD requires this to be reduced lower. How can extinctions be reduced in other taxa? Have to know more about the 10 percent of the earth’s surface where hotspots are. He would like to see crowns of 1 million sq km of planet’s forests twice a year where hotspots are.
Hank: Agrees—in order to understand habitats of diverse areas, need something that can cover the planet on a 50 m basis that can detect biomass, coupled with a sampler that can give vertical profile. LIDAR/Radar combination could tell much about habitat structure, as well as human impacts. 5-10 m resolution is needed to really understand habitats. This is currently technologically feasible, just needs the will.
F. Muller-Karger: Need better costing policy for high resolution data to allow for long-term data collection planning and to cover large geographical areas routinely.
Nadine Laport??: Very poor coverage in tropical forests, particularly Africa. Possible resolution to have a geostationary satellite over tropical forests, with multiple images per day in order to allow for daily images by looking through cloud windows. Want multiple-scales—broad to fine scale. Doesn’t see need for hyperspectral satellite.
Pat Colmer / NatureServe: Need for higher spatial resolution for vegetation structure, like 10 m. Needed for habitat relationships, soil moisture, fire regimes, etc. Don’t have understanding of these at scale needed currently. Understanding what is threatening biodiversity, and abating threat is good way of addressing how deal with requirements of CBD. Remote sensing needs to allow one to assess whether making progress towards CBD goals.
Tom Smith UCLA: Need information on tropical forest phenology. Tropical forests are dispersed by vertebrates, so understanding phonological patterns at monthly and large scale is important.
Mgugi CSU: What is feasibility for finer resolution on thermal bands (question in relation to soil moisture)?
Woody: ASTER provides higher resolution on thermal bands, but lacks temporal coverage.
Hank: ASTER can be pretty powerful for evapotranspiration driven models.
Missed name: Question on phenology.
Hank: Seems that phenology is needed at canopy scale. Is this needed over full region of interest at longer return interval, or finer scale with less spatial coverage over shorter intervals?
Missed name: Is there competition for resources between terrestrial and marine scientists i.t.o. sensors? Can land data be used for ocean work?
F. Muller-Karger: Land is much brighter than ocean during day. Technologically have to push sensitivity to look at oceans—need narrower bands, and need to deal with atmospheric correction. How can we combine capabilities to deal with whole earth? It is a mistake to continue to compete between communities.
Nicholas Coops: Agree with Lidar sampler—don’t think we need complete coverage with high resolution. Hyperspectral work has allowed people to focus on the spectra of use, which can allow for more targeted multi-spectral sensors to be used.
Woody: Do we now know what we need in terms of spectra to observe?
Coops: Yes, thinks we are close. Issue is now how narrow can we make bands and still have acceptable signal to noise ratios. Adds that think we need to go bigger on lidar footprint than 10m to fit in crown.
Hank: Problem with putting lidar and multi-spectral together relates to cloud. Radar allows clouds to be seen through and give biomass. Need fine and coarse resolution to image and sample.
Name missed: Shouldn’t be competition between two disciplines. Research has shown that both terrestrial and marine need to look at same spectra (e.g. red edge).
VIRS needs a 700 nm badn for class 2 waters
Hank: Good time to transition to integrative part of discussion. There are coupling between river systems/mangroves, etc. and marine ecosystems. There are several places where there are very obvious connection between two systems. What kind of sensing and modeling couples these two systems.
Someone UMD: Greatest changes due to climate change are happening in far northern latitudes. Phenological periods there happen during time of high cloud cover. Second need for geostationary capability over these areas.
Missed name: How are products/images available useful to showing public importance of having these images?
Woody: Images are obviously very powerful in terms of outreach—another group is looking at that now.
Missed name: Geostationary orbits are far from earth, and best resolution available is 1 km. Is there any technically feasibility for improving resolution?
Woody: Bill Emery at Colorado thinks that an off-angle approach and combinations can get this down to 100 m.
Missed name: NASA needs to communicate its unique ability to look at whole world and this relevance to ecological study.
Bob Crabtree?: Comment on lidar versus radar. If had a small footprint lidar, wouldn’t need radar for forest structure. Radar, however, is much cheaper, and can be improved. So, what are we doing to work with Europeans who have put up high resolution radar?
Missed name: Agrees with comments on radar and lidar. Multi-observation INSAR is a very active area, and can give capability for vertical profile (not as good as lidar, but with wider coverage) in several years.
Hank: Dubayah and Saatchi have just got an r2 of 0.89 with fused radar/lidar in Costa Rica—as good as it gets.
Integration:
Missed name: Now is a great time to take advantage of Orion and NOAA and getting involved with them.
A recommendation is to establish a pilot study that examines land use and changes in a watershed and examines impacts on the coastal ocean, including sedimentation changes, water quality effects, etc., and helps develop a set of recommendations for applications.
Fred UMBC: Problem with geostationary is that don’t get coverage of high northern latitudes. Constellation of low earth orbit sats maybe better.
Missed name: Biodiversity cuts across terrestrial and marine. Many solutions for biodiversity will be setting aside protected areas in both terrestrial and marine settings. Satellites give best synoptic view over these things, so NASA can wave biodiversity flag as way of integrating marine and terrestrial.
Volker: Need ability to track dispersal of animals, etc. A whole different suite of sensors than what has been discussed until now.
Woody: Wanted to have that issue raised in this discussion.
Eric Sanderson WCS: Agrees with what Volker has said. But need to remember that animals move over background. Best contribution that NASA can make is 1) figuring out landcover change is, and 2) what consequences of LC change is. Also telling how species and biodiversity are changing with climate. E.g. Current changes in the marine environment—what are effects of this? Seems like a perfect thing for NASA to look at.