We found a match
Your institution may have access to this item. Find your institution then sign in to continue.
- Title
Spectroscopic determination of leaf morphological and biochemical traits for northern temperate and boreal tree species.
- Authors
Serbin, Shawn P.; Singh, Aditya; McNeil, Brenden E.; Kingdon, Clayton C.; Townsend, Philip A.
- Abstract
The morphological and biochemical properties of plant canopies are strong predictors of photosynthetic capacity and nutrient cycling. Remote sensing research at the leaf and canopy scales has demonstrated the ability to characterize the biochemical status of vegetation canopies using reflectance spectroscopy, including at the leaf level and canopy level from air- and spaceborne imaging spectrometers. We developed a set of accurate and precise spectroscopic calibrations for the determination of leaf chemistry (contents of nitrogen, carbon, and fiber constituents), morphology (leaf mass per area, Marea), and isotopic composition (δ15N) of temperate and boreal tree species using spectra of dried and ground leaf material. The data set consisted of leaves from both broadleaf and needle-leaf conifer species and displayed a wide range in values, determined with standard analytical approaches: 0.7- 4.4% for nitrogen (Nmass), 42-54% for carbon (Cmass), 17-58% for fiber (acid-digestible fiber, ADF), 7-44% for lignin (acid-digestible lignin, ADL), 3-31% for cellulose, 17-265 g/m² for Marea, and -9.4% to 0.8% for δ15N. The calibrations were developed using a partial leastsquares regression (PLSR) modeling approach combined with a novel uncertainty analysis. Our PLSR models yielded model calibration (independent validation shown in parentheses) R² and the root mean square error (RMSE) values, respectively, of 0.98 (0.97) and 0.10% (0.13%) for Nmass, R² = 0.77 (0.73) and RMSE = 0.88% (0.95%) for Cmass, R² = 0.89 (0.84) and RMSE = 2.8% (3.4%) for ADF, R² = 0.77 (0.69) and RMSE = 2.4% (3.9%) for ADL, R² = 0.77 (0.72) and RMSE = 1.4% (1.9%) for leaf cellulose, R² = 0.62 (0.60) and RMSE = 0.91%(1.5%) for δ15N, and R² = 0.88 (0.87) with RMSE = 17.2 g/m² (22.8 g/m²) for Marea. This study demonstrates the potential for rapid and accurate estimation of key foliar traits of forest canopies that are important for ecological research and modeling activities, with a single calibration equation valid over a wide range of northern temperate and boreal species and leaf physiognomies. The results provide the basis to characterize important variability between and within species, and across ecological gradients using a rapid, cost-effective, easily replicated method.
- Subjects
PLANT canopies; PHOTOSYNTHESIS; NUTRIENT cycles; STANDARD deviations; REFLECTANCE spectroscopy; REMOTE sensing; TAIGA ecology
- Publication
Ecological Applications, 2014, Vol 24, Issue 7, p1651
- ISSN
1051-0761
- Publication type
Article
- DOI
10.1890/13-2110.1