Effects of canopy structural variables on retrieval of leaf dry matter content and specific leaf area from remotely sensed data

Leaf dry matter content (LDMC) and specific leaf area (SLA) are two important traits in measuring biodiversity. To use remote sensing for the estimation of these traits, it is essential to understand the underlying factors that influence their relationships with canopy reflectance. The effect of canopy structures—particularly stem density (SD), leaf area index (LAI), stand height (SH), crown diameter (CD), and average leaf angle (ALA)—on the relationship between LDMC and SLA with the canopy reflectance were investigated using a canopy reflectance dataset simulated by the invertible forest reflectance model (INFORM) radiative transfer model. The parameterization of the model was based on the range of the field parameters collected in the Bavarian National Park in July 2013 and the configuration of the HYSpex hyperspectral sensor. Strong correlations were observed between the two leaf traits and indices derived from simulated canopy spectra in the NIR and SWIR region ( {math\bf{R}^math\bf{2}} values of 0.87 for LDMC and 0.85 for SLA). Among the tested HYSpex wavelengths, the bands most sensitive to variation were 2298.69 nm for LDMC and 2280.71 nm for SLA. The effects of the stated structural variables on the relationships were best controlled by the modified normalized difference (mND) vegetation index (VI): ( [math\bf{R2275} - math\bf{R1920}]/[math\bf{R2275} + math\bf{R1920} - math\bf{2}{\ast }math\bf{R1520}] ). The structural variables that most affected the relationship were forest SD and CD. The modeling results suggest that the spectral variation due to changes in LDMC and SLA is best captured for stands with math\bf{SD} > math\bf{400};math\bf{trees}/math\bf{ha} and math- f{CD} \geq math\bf{5};math\bf{m} . The influence of LAI and SH on the relationships can be greatly reduced using VIs. We conclude that LDMC and SLA can be accurately estimated from canopy reflectance, irrespective of the heterogeneity of structural variables, providing that canopy cover exceeds 50%.