TERN’s national research infrastructure is being used by scientists from all around the world, including by a scientist from Estonia whose research will help new NASA technology be better utilised to monitor and measure environmental change. Meet the researcher, learn about an innovative method to assess vegetation clumping, and read his independent assessment of the importance of TERN for global satellite product validation.
In 2015, SpaceX launched a revolutionary satellite to monitor space weather, space climate and provide a system of warning Earth of solar magnetic storms. The joint NASA NOAA (National Oceanic and Atmospheric Administration) satellite—called Deep Space Climate Observatory (DSCOVR)—also provides important Earth Observation data.
DSCOVR’s Earth Polychromatic Imaging Camera (EPIC) takes multi-spectral images of the entire sun-lit side of Earth every one to two hours, allowing scientists around the world, including Dr Jan Pisek at Estonia‘s Tartu Observatory, University of Tartu, to monitor the land and how it’s changing.
An enhanced image of Earth captured on 31 May 2021 by DSCOVR’s Earth Polychromatic Imaging Camera (EPIC) (credit NASA)
Jan is particularly interested in new ways to improve satellite-derived measurements of vegetation foliage distribution or ‘clumping’.
Vegetation clumping in forest canopies is important as it affects how sunlight is distributed within canopies and influences photosynthesis and other important land-atmosphere interactions such as water and carbon cycles. Data on clumping are also used in many ecological and meteorological models.
Jan’s latest research, featured in an upcoming Frontiers in Remote Sensing special issue, provides the first assessment of new approach to estimating vegetation clumping from space using the unique observing capabilities of the DSCOVR EPIC satellite and its products.
TERN sites the proving-ground for remote-sensing innovation
The seeds of the current study go back to 2013 when Jan travelled to Australia to utilise TERN’s world-class research infrastructure.
Alongside local scientists, Jan conducted research using TERN’s Wombat Stringybark SuperSite, Whroo Dry Eucalypt SuperSite and Warra Tall Eucalypt SuperSite as part of a global project to characterise and compare three different products that measure vegetation clumping using remote sensing technology.
“When I visited Australia and used the TERN sites in 2013, I saw how professionally the infrastructure is run and how my studies of the unique Australian ecosystems complemented and validated my Europe-based research.
So, I returned in February 2019 to measure a couple of additional TERN sites – Cumberland Plain SuperSite and Tumbarumba Wet Eucalypt SuperSite.
Dr Jan Pisek, University of Tartu
Above, Dr Jan Pisek climbs the eddy-covariance flux tower at the TERN Wombat Stringybark SuperSite to measure the vertical structure of the forest (image courtesy of Zbynek Malenovsky) and at top, conducting research at TERN’s Cumberland Plain SuperSite in Western Sydney.
The new method, which uses a variety of data from DSCOVR including Leaf Area Index (LAI), Normalized Difference Vegetation Index (NDVI) and Fraction of incident Photosynthetically Active Radiation (FPAR), was evaluated using the in-situ measurements taken at the five TERN sites in 2013 and 2019.
The findings of the long-term study indicate that DSCOVR EPIC data can provide accurate measures of clumping and useful data for ecosystem modelling at multiple scales.
“It is very encouraging to show that it’s possible to obtain such good quality results using a different approach and different Earth Observation data that are very much comparable to previous efforts of mapping clumping from space.
This general agreement between different retrieval strategies and input data sources is important for increasing overall confidence, justification, and general validity of clumping information retrieval from space in the future.”
Clumping index: new sensor @dscovr_epic, new approach but similar results-very encouraging for the general validity, compatibility of the foliage clumping information retrievals from space! Spatial representativeness of five @TERN_Aus sites assessed, too:https://t.co/HSyiGgspP9
— Jan Pisek (@jan_pisek) March 17, 2021
TERN sites highly suitable for validation of Earth Observation products
During his work developing the new method, Jan also assessed the suitability of the five TERN SuperSites for the validation of global satellite data.
The five sites assessed had a diverse range of canopy structure from short and sparse to dense and tall forest, but it is their spatial representativeness that makes the infrastructure so valuable for satellite validation.
“I was lucky and also had a unique opportunity to visit and carry out field campaigns at most of the ICOS [Integrated Carbon Observation System] forest ecosystem sites around Europe. The ICOS sites are great sites for satellite data validation too, but the problem is that the European landscape is often too fragmented for proper validation of Earth Observation products, especially at coarser spatial resolution.
In that sense, our recent study showed that some TERN sites are representative of the greater area surrounding them and are highly suitable for validation of Earth Observation products across a wide range of spatial scales, that may be very seldom a case in, for example, Europe.”
- Click here for more information on this research
- Click here for more information on TERN’s Australia-wide network of ecosystem processes monitoring SuperSites.
- TERN’s network of sites is well equipped with a huge variety of monitoring devices to measure ecosystem change across the continent. TERN welcomes co-use of sites by researchers.
Alongside local scientists, Dr Jan Pisek conducted research using five TERN SuperSites – Cumberland Plain (A,B), Tumbarumba Wet Eucalypt (C,D), Whroo Dry Eucalypt (E,F), Wombat Stringybark (G,H) and Warra Tall Eucalypt (I,J) – as part of a global project to characterise and compare three different products that measure vegetation clumping using remote sensing technology.
