The 20th ACRS was held in the Convention and Exhibition Centre in Hong Kong from 21 to 25 November. There were over 300 registrants to the conference, including 129 from China (the largest group!), 39 from Thailand, around 30 from Taiwan, 12 from USA and 7 from Australia.

After a few keynote speeches in the morning of the first day, the conference was divided into several parallel technical sessions, which were agriculture/soil, water resources, land, forest, mapping from space, oceanography/coast zone, geology, environment, GIS, image processing, education and disasters. In addition, there were two special sessions on hyperspectral image processing and 3D measurement and modelling.

In terms of type of sensors, most presented work was related to Landsat TM, AVHRR, SPOT, and RadarSat-1, but only a few was in the hyperspectral field. GIS applications were one of the main topics.

I orally presented a paper, co-authored with Jon Huntington and Joe Boardman, on ARIES geological simulation at Virginia City, Nevada in the special session on hyperspectral data processing. Most presentations in the session were on data processing techniques, but our paper was concentrated on the application. My general impression was that hyperspectral work was weak in the Asian RS community, probably due to the unavailability of data. China is the only country in Asia that has hyperspectral sensors (airborne). Although several Chinese hyperspectral sensors were flagged in two keynote speeches, only a few technical papers on hyperspectral work from the Chinese sensors were presented.

Several workshops were also organised during the conference. I was invited to give a presentation on mineral mapping in the hyperspectral workshop organised by the Institute of Remote Sensing Applications (IRSA), Chinese Academy of Sciences (CAS). There were four speakers on that session, Tong Qingxi (overview), Wang (sensor technology), myself (mineral mapping) and Zhang Bing (data processing and software). I took the opportunity to promote ARIES again, especially its geological application potentials. Though not originally scheduled, Alex Goez was invited to talk for 20 minutes, and he specifically mentioned the relationships between the number of end-members, spectral resolution and spatial resolution, the experiments recently done by his team.

In the geology session, 6 papers were orally presented, only two of them were on mineral exploration.

Trade exhibition represented a best part of the conference. Companies and organisations taking part in the exhibition included some well-known players such as ER Mapper, PCI, RadarSat, and Spot.

One display that particularly interested me was the portable FieldSpecÒ Pro spectrometers, designed and manufactured by Analytical Spectral Devices, Inc. One model of the FieldSpecÒ Pro series operates from 0.35 to 2.5m m, covering the entire wavelength range for hyperspectral remote sensing. Unlike the Australia-made PIMA, the FieldSpecÒ Pro spectrometers work on passive light and do not measure targets in contact, and therefore they are remote sensing instruments. Using a fiber optic input cable, the spectrometers work really fast (a few seconds for a spectrum). More surprising when I saw Prof Alex Goez (Univ of Colorado, Boulder) demonstrating the instruments in the booth. That company has a local agent in Beijing, and three FieldSpecÒ Pro instruments were recently sold in China.

What was not displayed, but relevant in this topic, was a ground-based image spectrometer developed by the group led by Yoshifumi Yasuoka at Tokyo University. From the conversation with Yasuoka, I learnt that the instrument worked in the VNIR wavelength region. It uses AOTF (dispersing filter), and weighs about 10 kg. I have requested more detailed information about this instrument.

The fact that China made the biggest delegate group may be partly due to the geographical proximity of the conference venue to mainland China, but also reflects the great number of remote sensing scientists in China and increasing RS applications countrywide. The conference, therefore became an occasion to see the current Chinese RS activities.

There were lots of presented applications in weather forecast, flooding effect assessment, crop yield prediction, and environment monitoring. There were only a few papers in mineral exploration application.

China quickly adopted the concept of digital earth, which was originally proposed by the Americans, and the government has set it as one of the priority fields for the next century. A keynote speech by Guo Huadong from CAS was on digital earth. Many conference participants were going to Beijing for the first international digital earth symposium to be held three days after the Hong Kong conference.

China’s effort in RS was also reflected by the development of sensors. Chinese RS sensors were brought to attention several times during the conference. Their airborne sensors included hyperspectral ones, L-band SAR and CCD camera, and the spaceborne ones were CBERS (a Landsat type satellite) and Fengyun-1C (a NOAA/AVHRR type meteorological satellite). For the first time, a Chinese hyperspectral satellite system, C-HRIS, was published at the conference.