A robust leaf area index algorithm accounting for the expected errors in gap fraction observations

Gonsamo Alemu 1, Walter Jean-Michel 2, Chen Jing M. 1, Pellikka Petri 3,4, Schleppi Patrick 5

1 Department of Geography and Planning, University of Toronto, Toronto, ON, Canada
2 Institut de Botanique, Faculté des Sciences de la Vie, Université de Strasbourg, Strasbourg, France
3 Earth Change Observation Laboratory, Department of Geosciences and Geography, University of Helsinki, Helsinki, Finland
4 School of Geography and Remote Sensing, Nanjing University of Information Science and Technology, Nanjing, China
5 Swiss Federal Institute for Forest, Snow and Landscape Research, Birmensdorf, Switzerland

Agric. For. Meteorol. 248 (2018): 197-204

DOI: 10.1016/j.agrformet.2017.09.024


The leaf area index, LAI, representing the physiological and structural functions of vegetation canopies, can be estimated from gap fraction measurements obtained at different zenith angles. Earlier works have provided practical and convenient theoretical solution to retrieve LAI based on the integration of contact numbers (a projected area of leaves on a plane perpendicular to the view or solar zenith angle) over zenith angles as obtained by a linear regression, i.e., LAI = 2(A + B), where A and B are the coefficients of the regression of contact numbers against zenith angles. This graphical procedure is equivalent to the more accurate method of LAI retrieval by integrating gap fraction measurements from nadir through horizon angles. However, using an ordinary least-squares regression on inherently unsteady relationship between contact numbers and zenith angles limited the use of a simple graphical procedure for LAI estimation. In this study, we introduce the use of robust procedure to retrieve regression coefficients (i.e., A and B), and assess the performance of the new procedure using numerically derived hypothetical data, computer simulated and real measurements of hemispherical photographs. Our results indicated, the new procedure not only outperformed the ordinary least-squares solution for graphical procedure, but also outperformed all existing LAI methods We conclude from analyses using numerically derived hypothetical data, computer simulated and real measurements of hemispherical photographs that estimating A and B (where LAI = 2(A + B)) using a robust procedure is a convenient and sufficiently accurate method for estimating LAI from field measurements of gap fractions at different zenith angles.

Keywords: CIMES, gap fraction, hemispherical photograph, leaf area index, least absolute deviations, robust regression