Overseas Visit Report
Dr Fred Prata
CSIRO Atmospheric Research
EUROPTO Symposium
Barcelona, Spain
21-25 September, 1998
This meeting consisted of ten separate conferences held concurrently. The conferences were:
I would guess between 500 and 1000 people attended the symposium; given the breadth of coverage of the meeting I was surprised that it was not better attended.
I presented two papers in the conference on "Sensors, Systems and Next Generation Satellites." The first paper titled "Shortwave Calibration of The ATSR-2 at Australian Land Sites", summarized our results during the recent ACEX campaign at Uardry and Amburla, and conducted in collaboration with the University of Nottingham, England. Our results were in fact quite good (in terms of experimental procedure and errors) and demonstrated that the ATSR-2 shortwave channels can be confidently believed at an accuracy better than 5% in all channels and better than 2% in some channels. These estimates agree with other vicarious techniques that have been utilised for the ATSR-2, such as examining the reflectances from the tops of deep convective clouds, and monitoring reflectances from the Libyan desert site. The use of various instruments in the field, such as the Multi-Filter Rotating Shadowband Radiometer (MFRSR), the CCD Spectrometer, our own radiosonde launching equipment and the routine CIGSN field data (pyranometers, pyrgeometers, meteorological data), provided us with a comprehensive data-set from which to draw our conclusions. Since we did not measure ozone profiles we used data from the GOME instrument also housed on the ERS-2 satellite. The other aspect to this work was the use of a radiative transfer model to generate top-of-the-atmosphere reflectances from our ground-truth measurements. We chose Modtran 3, but are now investigating 6S and other models to examine the limitations of the modelling approach in this kind of vicarious calibration method. The results from this study have been submitted to SPIE for publication in the proceedings and I would be happy to send reprints of the paper to anyone interested.
The second paper that I presented was on behalf of Ian Grant (and his collaborators at the USGS) titled: "Lunar Calibration of Geostationary Visible-Band Imagers". This paper followed two other papers also on the topic of lunar calibration; one concerning SeaWifs calibration and the other concerning ground-based lunar photometry. The work that Ian has been doing is, in my opinion, a very nice way to do visible calibration because the moon is radiometrically a very stable target, and has enough brightness variation across its surface to permit a range of reflectances to be observed. Of course, the major difficulty with using the moon is that it has an inaccurately known BRDF (at the moment) and some satellites never view the moon or only view infrequently (e.g. SeaWifs). Nevertheless, the ground-based program being conducted by the USGS and the Northern Arizona University will eventually deliver a lunar photometric model from which moon's reflectance can be determined for arbitrary viewing and illumination conditions. Ian has cleverly circumvented the need for the photometric model by noting that the ground-based measurements at Arizona view the moon at almost the same angle as the GMS-5 imager. This should allow an immediate calibration, provided the atmosphere over Arizona is clear on the night of the observations when the moon is near full within the GMS-5 viewing frame. One such occasion has occurred and Ian's paper discussed the results from this coincidence. The paper was well received and will also be published in the SPIE proceedings. Please contact Ian directly for further information about his work (email: Ian.Grant@dar.csiro.au).
Apart from the important task of giving good presentations, answering questions and being involved in subsequent discussion, I made plans to listen to -carefully chosen presentations (recall that the conferences were concurrent) and I also took the opportunity to meet with European colleagues working on problems of interest to the CSIRO remote sensing community. I attended all of the "Sensors " conference (held on only one day), as much as possible of the "Satellite Remote Sensing of Clouds " conference (held over 3 days), and parts of the conferences on "Spectroscopic Atmospheric Environmental ", "Remote sensing for Geology " and "Remote Sensing for Agriculture". It is fairly obvious, I think, that these last two conferences are relevant to our work here in Australia, but you may wonder why I attended the "Spectroscopic .." conference. The title of the conference is somewhat misleading as the majority of the talks focussed on infrared spectroscopy and the growing use of Fourier Transform Interferometry in applied remote sensing. Thus there were talks on remote sensing of pollution using FTIR, FTIR spectroscopy of hydrocarbon emissions (related to natural gas leaks), studies of exhaust gases using infrared spectrsocopy, and some mathematical papers discussing retrieval techniques, atmospheric gas spectroscopy and instrumental considerations. I have a strong interest in FTIR and can see that, just like in the visible part of the spectrum, high resolution information in the infrared will lead to some new insights in atmospheric phenomena.
The conference on "Satellite Remote Sensing of Clouds " was also full of interest to me, and the talk by Bob Cess on the ongoing saga of anomalous solar absorption was perhaps the most interesting. Still unresolved, the problem of why models and experiment do not agree on the amount of shortwave radiation that is absorbed by the atmosphere in both clear and cloudy-sky conditions, is challenging the minds of many atmospheric researchers around the world. Some firmly believe that the models are incorrect (a "missing Physics ?!"), while others believe that the data are not good enough to make conclusions, or that the experiments have not been set up with sufficient rigour to provide the necessary data. Cess is of the former camp, and his talk suggests that it is in the near infrared region where our knowledge of solar absorption may be deficient. He presented data from a large network (the Canadain network) of pyranometers assembled to be coincident with satellite data (ERBE) to show that the absorption anomaly was present in these data. He also presented more aircraft data. Other papers in this conference included a paper on cirrus property retrieval using ATSR-2 data, the effects of water and ice cloud on out-going longwave radiation and some papers addressing the issue of heating rates in the far infrared region.
In the "Remote Sensing for Geology " conference I listened to a paper that purported to have discovered a 'Lost City' in Honduras. Apart from the fact that I had not realised it was lost, the paper did not convince me that the authors had found anything but some clutter in a radar image ! When asked if they had any corroboration for their findings, they simply stated that next year they are planning an expedition to the region. I was disappointed by this talk. There were some good talks in this session though, including a talk on volcano hazard mapping in the Philippines, a paper utilizing remote sensing and GIS applied to restoration of open-pit mining areas (some relevance to Australia here, I think), and a paper on the University of Hawaii's thermal infrared hyperspectral imager.
There were Poster Sessions at this meeting, convened after the main presentations were complete and supplemented with an informal cocktail party. These poster sessions are always worthwhile, because there is usually much better discussion and they offer the chance to engage the author in a detailed question and answer session. I spent much time discussing a poster by Juan Cuenca (University of Valencia) who had devised an apparatus to measure viewing effects on remotely sensed thermal infrared data. His goniometer, with some slight adjustments, could permit a measurement of the thermal BRDF (or BEDF as David Jupp likes to call it) and he seemed keen to do this in the future. His main interest was to look at zenith viewing angle effects and compare his experimental measurements with thermal infrared measurements from the ATSR-2, which views out to around 55 degrees from nadir.
During the week I had various discussions with Phil Slater (cal/val), Simon Hook (NASA/JPL collaborator), Jose Sobrino, Cesar Coll and Juan Cuenca (University of Valencia; thermal infrared interests), a represntative from Kestral Inc. (building the MightySat II hyperspectral imager), Bob Barnes (SeaWifs lunar calibration) and Peter Read (GERB project team, RAL, UK).
In summary, the meeting proved to be very worthwhile and covered quite a broad spectrum of topics. I have the abstract book and have ordered the SPIE proceedings for the conferences on "Sensors, Systems and Next Generation Satellites IV" and "Remote Sensing of Clouds and the Atmosphere". Proceedings for the other eight conferences are being produced and are available for purchase from SPIE. Please contact me if you require any further information about the meeting or the publications arising from it.
Dr Fred Prata
CSIRO Atmospheric Research
PMB 1
Aspendale, Vic 3195
Tel: (03) 9239 4681
Fax: (03) 9239 4444
Email: Fred.Prata@dar.csiro.au
27 October, 1998