A brief report on the Second International Asia-Pacific Symposium on Remote Sensing of the Atmosphere, Environment, and Space, 9-12 OCT 2000, Sendai, Japan

Kai Yang, CSIRO Exploration and Mining, North Ryde

After the first conference two years ago in Beijing, the second SPIE’s International Asia-Pacific Symposium on Remote Sensing of the Atmosphere, Environment, and Space was held during 9-12 Oct 2000 in Sendai, Japan.  With the support of a travel grant from COSSA/EOC, I attended the conference and orally presented a paper (co-authored with Jon Huntington, Tom Cudahy and Keith Scott) on geological applications of spectrally mapping white mica composition in the Hyperspectral RS session.

The four-day conference attracted about 150 (just guessing here, as no delegates list available, as complained by many attendees) scientists and engineers from various parts of the world.  There were a few Australians listed on the programs, and I met Mervyn Lynch of Curtin University.  The conference themes were very much diversified, covering topics from Radar, Lidar, ocean color sensors, Landsat, to hyperspectral, and from land, ocean to atmosphere.  With multiple sessions proceeding in parallel, I often had difficulty in deciding which session or talk to go to.  Anyway, most of the time I stayed in the session on Hyperspectral RS of Land and Atmosphere, and so here I am briefly reporting about this session.

The first day Plenary Session highlighted a few key-note speeches.  John Cunningham introduced the general concept, the goals, the challenges, and the current status of the National Polar-Orbiting Operational Environmental Satellite System (NOPESS) project.  HOPESS is an ambitious project, which involves three US government agencies (NASA, NOAA and Defense) and four nations.  I cannot go into any details here, but the slides of his talk are now on their web site already.  Another speaker in the Plenary Session was M. Tanaka (Tohuku Institute, Japan), on the remote sensing of global carbon circle.

Just like many SPIE organised conferences, instrumentation was one of the main topics (3/4 day were allocated in the Hyspectral RSLA session).  The presented instruments included the EO-1/Hyperion sensors (by M. Folkman et al. of TRW), which now many people know about.  Also presented, and interesting, were three talks, by Revercomb of Univ of Wisconsin; Puschell of Raytheon Electronics Systems; Bingham of Utah State Univ.; USA), on the imaging fourier transform spectrometer (FTS), which is to be used for sensors of the EO-3 (?) project.  FTS was claimed to be the future of spaceborne earth observation technique, because it met the requirement for selectable spatial and temporal resolution.  Also presented were instrument calibration issues for the sensors of AVIRIS, AIRS (on the EOS-Aqua), FY-II (Chinese weather satellite), and a few others.

As to the applications, there were some interesting work presented.  These included the work by E. Winter (University of Hawaii) with Airborne Hyperspectral Imager (AHI).  AHI is an airborne thermal infrared instrument with 30 bands in the 8-12 microns wavelength region.  Both geological and vegetation mapping were presented.  Winter also briefly mentioned their software (N-Finder), which is based on unmixing of pixel spectra and tested on both geology (Cuprite, AVIRIS data) and vegetation (one dataset from TRW).

Also noticed was that Laurie Richardson’s (Florida International Univ) work with AVIRIS data on identifying and classifying phytoplankton assemblages in coastal waters.  For the first time, I learnt that she took the unmixing approach for mapping (in ENVI) and selected end-members directly from the image data, though I saw some results of her work many times from someone else’s presentation.  What also interested me was that when working with data of multiple flight lines Richardson mosaicked the lines first, and then processed the combined dataset, the way we found hard to proceed when the combined dataset is too big to be handled by ENVI.

Another good paper was presented by Cassady of Boeing Co. on detecting agricultural crop stress with AVIRIS data.  Of the few red-edge parameters compared, he found the one proposed by Ray Merton worked best!

Though not really fit in the hyperspectral theme, two Chinese scientists presented rather interesting results using the data from China’s FY-1C meterological satellite (10 bands, AVHRR-like), which was lunched last year, to determine spatial and temporal changes in snow cover in northwest China.

There were also several presentations on ground-based hyperspectral work.

During the conference, a half-day tour was scheduled to take all delegates up to the crater on the peak of a volcanic mountain about 3-hours drive from Sendai.

During the conference, I met and talked with Prof QX Tong of Institute of Remote Sensing Applications, Chinese Academy of Sciences about collaborations in geological hyperspectral remote sensing in China.  We both again believed that such collaborations were useful particularly for bringing the planned ARIES to the attention of the potential Chinese users.

The location of the next SPIE’s Asian-Pacific remote sensing conference in 2002 has not been decided, but cities under consideration are Hong Kong and Shanghai.