European Commission - Actual canopy

This page provides access to test cases that are based on detailed measurements of the spectral, architectural (and in some cases also illumination related) properties encountered in a variety of existing vegetation canopies. Potential RAMI participants are to treat the datasets and descriptions that are provided on the experiment description pages below as actual 'inventory data', that is, they should mimick the standard practices that are being used when actual field measurements are matched to the set of input parameters required by their model(s). The following test sites have been implemented for RAMI-V:

This figure exhibits Järvselja pine stand (summer), Estonia).
@Copyright

This figure exhibits Ofenpass pine stand (winter), Switzerland.
@Copyright

This figure exhibits Järvselja birch stand (summer), Estonia.
@Copyright

This figure exhibits Wellington citrus orchard, South Africa.
@Copyright

This figure exhibits Järvselja birch stand (winter), Estonia.
@Copyright

This figure exhibits Lombardy short rotation forest, Italy.
@Copyright

The following publications describe the above test cases in greater detail:

Morsdorf F., B. Kötz, E. Meier, K.I. Itten, B. Allgöwer, (2006) Estimation of LAI and fractional cover from small footprint airborne laser scanning data based on gap fraction, Remote Sensing of Environment, Volume 104, Issue 1, 15 September 2006, Pages 50-61.

Kuusk, A., Kuusk, J., and Lang, M., (2008) A Database of Optical and Structural Data for the Validation of Radiative Transfer Models. Technical Report, 26-30 May 2008 Rhodes, Greece, ESA SP-660, 10 pp.

Kuusk, A., Lang, M., Kuusk, J., Lükk, T., Nilson, T., Mõttus, M., Rautiainen, M. , and Eenmäe, (2008) A dataset for the validation of refle ctance models. Proc. 'The 4S Symposium - Small Satellites Systems and Services', Tartu Observatory, 2008, 52 pp. Available on-line at the following URL, http://www.aai.ee/bgf/jarvselja_db/jarvselja_db.pdf.

Morisette J., Privette J. L., Baret F., Myneni R.B., Nickeson J., Garrigues S., Shabanov S., Fernandes R., Leblanc S., Kalacska M., Sanchez-Azofeifa G. A., Chubey M., Rivard B., Stenberg P., Rautiainen M., Voipio P., Manninen T., Pilant D., Lewis T., Iiames J., Colombo R., Meroni M., Busetto L., Cohen W., Turner D., Warner E.D., Petersen G.W., Seufert G. and R. Cooke, (2006) International LAI Product Intercomparison: Initial Results, IEE Transactions on Geoscience and Remote Sensing, 44/7-1, 1804-1817.

Stuckens J., B. Somers, S. Delalieux, W. W. Verstraeten, P. Coppin, (2008) The impact of common assumptions on canopy radiative transfer simulations: A casestudy in Citrus orchards, JQSRT, doi:10.1016/j.jqsrt.2008.09.001.

Somers B.,S. Delalieux, W. W. Verstraeten,P. Coppin, (2009) A conceptual framework for the simultaneous extraction of sub-pixel spatial extent and spectral characteristics of crops, Photogramm. Eng. Remote Sensing, 2009, 75, 57-68.

Zenone T., Migliavacca M., Montagnani L., Seufert G. and Valentini R., (2008) Carbon sequestration in short rotation forestry and traditional poplar plantation, In Proceedings of the Short Rotation Crops International Conference, Minneapolis, USA, August 18 - 22, 2008 (http://www.cinram.umn.edu/srwc/cs1es.html).

Individual trees were generated with graphtal and xfrog:

B. Lintermann and O. Deussen, (1999) Interactive Modelling of Plants, IEEE Computer Graphics and Applications, Vol. 19, No 1, pp 2-11.

Missing inventory data

When generating the spectral, architectural and illumination conidtions of the various actual canopy scenes above several structural and spectral information were not always available from the inventory data. These included: detailed crown shape descriptions, information about the amount of woody content, branch numbers and angles, as well as the wood area index. On some occasions foliage shape, dimensions and distribution in a crown, as well as foliage orientation distributions were not available. In addition, the directionality of the incident radiation was not known in all wavelengths of interest, neither were the scattering behaviours of foliage, wood and background components known. Transmission properties of foliage elements had not been measured for all species and no information regarding the spatial variability of the various spectral and structural properties was known.

The RAMI forest scenes available via the above links are thus reconstructions that use as much as possible the available inventory information, filling any information gaps with plausible alternatives (if these where available) or else resorting to assumptions that do not compromise the gathered field-data (e.g., spatially uniform background brightness, lambertian scattering probabilities, etc).

The various testcases that are defined within this phase of the RAMI initiative can be visualized on their respective description pages:

Actual canopy

Missing inventory data

Parametric