San Joaquin Experimental Range, CA, USA

HYPERMED project

Species traits assessment to monitor tree health condition over open MEDiterranean forest ecosystems for a Hyperspectral imager

SCOPE & OBJECTIVES

The HyperMED project focuses on the study of oak woodland savannas in Mediterranean climate, which usually represent a limited number of endemic and fire-tolerant tree species, and some used to serve for pastures. Such tree-grass ecosystems are threatened by agriculture and urbanization expansion, and become critical areas due to their sensitivity to abnormal increase of drought and forest fire periods.

The objective is to demonstrate the ability of two hyperspectral satellite missions, CNES-CHIMERE and NASA-SBG, to estimate and monitor over time biophysical and biochemical variables (namely LAI: Leaf Area Index, CHL: CHLorophylls, CAR: CARotenoids, EWT: Equivalent Water Thickness, LMA: Leaf Mass Area) as species traits able to witness the plant health condition or physiological processes such as photosynthesis, transpiration, nutrient allocation, growth rate and decomposition. Accurate estimation of these species traits requires a good signal-to-noise ratio (especially for CAR having a narrow sensitivity spectral range), a spatial resolution adapted to the ecosystem specificities (e.g. limitations due to the large encountered range of canopy covers), and an appropriate temporal revisit to follow their seasonal and interannual variability (i.e. phenological cycles). CHIMERE and SBG sensors may be complementary to deal with this issues since they have different spatial resolution (CHIMERE @8m targeting the tree scale, SBG @30m observing a larger and more diverse scenery), same spectral characteristics (hyperspectral 0.4-2.5 µm with 10nm resolution) but different signal-to-noise ratio (CHIMERE’s lower to SBG’s).

DATA & METHODS

Two grass-oak-pine woodland savannas are selected, Tonzi Ranch (TZ) and San Joaquin Experimental Range (SJER), located on the foothills of the Sierra Nevada mountain range in California, USA. They both have a cattle grazing activity. TZ is a private owned land and carry instrumentation platforms managed by Biomet Lab, some included in the Ameriflux/Fluxnet and Phenocam networks. SJER is hosted by the U.S. Forest Service where NEON (National Ecological Observatory Network) established one of their terrestrial site for further samplings and observations. They are studied as reference sites for the HyspIRI mission. The two studied broadleaf tree species are Quercus Douglasii (deciduous) and Quercus Wislizeni (evergreen).

The complete used dataset includes species trait measurements in the field and laboratory performed by CSTARS and AVIRIS-Classic and AVIRIS-Next Generation hyperspectral airborne acquisitions @2m and @18m performed by NASA JPL from which were simulated CHIMERE and SBG images.

The general methodology relies on the development of a hybrid inversion method based on simulated spectral databases or LUT (Look-Up-Table) from leaf (PROSPECT) and canopy (DART) radiative transfer models and the building of design of experiments with OpenTURNS python library, and the use of LUT-based or machine learning regression algorithms (MLRA).

RESULTS & WORK PROGRESS

For the sakes of generalisation over the two sites presenting a similar ecosystem type, a simple forest representation composed of a flat lambertian background and four “lollipop” trees was chosen for radiative transfer modelling. Three major criteria impacting the estimation of species traits have been further studied:

  • the canopy cover (CC),
  • the 3D structure of the tree,
  • the spatio-temporal variability of the background.

The first criterion is based on the low vegetation signal mixed with the high contribution within a pixel of the scene background when the CC is low (i.e. for the most open areas of the forest). Some work showed the limitations in processing CC equals or lower than 10% and better accuracies are obtained from 30% CC.

The second criterion is related to the low tree LAI of the oaks for both sites, highlighting the importance to model properly the presence of wood (e.g. trunk, branches). A very simple modelling composed of a cylindrical trunk proved to be sufficient to estimate accurately LAI and leaf pigments with AVIRIS @18m, but not EWT and LMA. Terrestrial LiDAR measurements were acquired by Crystal Schaaf’s lab crew and postprocessed to reconstruct the 3D shape of wooded elements. Adding it to the forest mock-up showed that EWT and LMA can be retrieved with AVIRIS at high spatial resolution @2m. The upscaling of the results to CHIMERE/SBG, and the comparison with AVIRIS @18m, demonstrated that performances to estimate the gap fraction (derived from the combination of CC and LAI) is similar between the three sensors whatever the CC, also similar for pigments but mostly over the densest canopy areas (CC > 80%), and improvements are needed for EWT and LMA.

The third criterion is under study.

SCIENTIFIC PRODUCTION

Ancillary publications

Workshops and conferences

GENERAL INFORMATION

Duration: 2019-2022

Funding: CNES (APR TOSCA)

Principal Investigator: Karine Adeline (ONERA); Co-investigators: Xavier Briottet (ONERA), Jean-Philippe Gastellu-Etchegorry (CNES-CESBIO), Susan Ustin & Margarita huesca (CSTARS, University of California, Davis), Dennis Baldocchi (Biomet Lab, University of California, Berkeley); Other contributors: Sidonie Lefebvre (ONERA), Nicolas Lauret (CNES-CESBIO), Crystal Schaaf (University of Massachusetts, Boston), Thomas Miraglio (PhD student), Mathilda Porterie (PhD student), Jean-Victor Schmitt (intern)

Recruited non-permanent staff:

  • Louis Chauvet (intern)
  • Romain Demoulin (intern)

Search