Comparison of biomass allometric approaches for regional scale

  • Slides: 25
Download presentation
Comparison of biomass allometric approaches for regional scale carbon mapping Scott Powell – Montana

Comparison of biomass allometric approaches for regional scale carbon mapping Scott Powell – Montana State University Robert Kennedy – Boston University Janet Ohmann – USDA Forest Service Warren Cohen – USDA Forest Service Matthew Gregory – Oregon State University Heather Roberts – Oregon State University Van Kane – University of Washington Jim Lutz – University of Washington Forest. SAT: Corvallis, Oregon, September 2012

Regional Carbon Mapping • Yearly (1990 -2010) maps of aboveground live biomass • Sources

Regional Carbon Mapping • Yearly (1990 -2010) maps of aboveground live biomass • Sources of uncertainty – Spectral data • 3 permutations – Modeling parameters • 3 permutations – Biomass allometrics • 2 permutations

Biomass Allometric Equations • Enable scaling of tree-level measurements to biomass. • Variety of

Biomass Allometric Equations • Enable scaling of tree-level measurements to biomass. • Variety of approaches ranging from generic to site-specific. – Different scales, assumptions, uses, and interpretations. – Carbon accounting vs. carbon mapping

Objectives • Compare mapped predictions of aboveground biomass based on two common allometric approaches.

Objectives • Compare mapped predictions of aboveground biomass based on two common allometric approaches. • Improve understanding of the range of uncertainty introduced into carbon mapping from selection of biomass allometric approach. • Assess differences in estimated biomass based on forest structure, composition, and land ownership.

Methods Allometric approaches: 1. Jenkins Equations: Nationally generic Jenkins, J. C. , D. C.

Methods Allometric approaches: 1. Jenkins Equations: Nationally generic Jenkins, J. C. , D. C. Chojnacky, L. S. Heath, and R. A. Birdsey. 2003. National-scale biomass estimators for United States tree species. Forest Science 49(1): 12 -35. 2. Component Ratio Method (CRM): Regionally-tailored but nationally consistent Heath, L. S. , M. H. Hansen, J. E. Smith, W. B. Smith, and P. D. Miles. 2009. Investigation into calculating tree biomass and carbon in the FIADB using a biomass expansion factor approach. In: Mc. Williams, W. , Moisen, G. , Czaplewski, R. , comps. 2009. 2008 Forest Inventory and Analysis (FIA) Symposium; October 21 -23, 2008: Park City, UT. Proc. RMRS-P-56 CD. Fort Collins, CO: U. S. Department of Agriculture, Forest Service, Rocky Mountain Research Station. 1 CD.

Jenkins Equations • 10 national-level generalized biomass equations based on metaanalysis of published equations.

Jenkins Equations • 10 national-level generalized biomass equations based on metaanalysis of published equations. • Current basis for U. S. greenhouse gas inventories. • Based solely on DBH measurements, and do not include tree height measurements. Aboveground Biomass = Exp(β 0 + β 1 ln DBH)

Component Ratio Method (CRM) • Basis for current FIA biomass estimates • Nationally-consistent method

Component Ratio Method (CRM) • Basis for current FIA biomass estimates • Nationally-consistent method that relies on regional FIA volume equations and specific gravity to estimate biomass. • Volume equations incorporate tree height (or surrogate)

Previous Studies • Zhou and Hemstrom, 2009 – PNW-RP-584 – CRM biomass estimates were

Previous Studies • Zhou and Hemstrom, 2009 – PNW-RP-584 – CRM biomass estimates were 17% lower than Jenkins biomass estimates for aboveground softwood biomass in Oregon. • Domke et al. , 2012 – Forest Ecology and Management. – CRM biomass estimates were 16% lower than Jenkins biomass estimates for the 20 most common species in the U. S.

Results: Overall Difference

Results: Overall Difference

Differences by Vegetation Class

Differences by Vegetation Class

Spatial Variation: Relative Differences by Height and Age Ratio = Jenkins/CRM

Spatial Variation: Relative Differences by Height and Age Ratio = Jenkins/CRM

Spatial Variation: Absolute Differences by Height and Age Difference = Jenkins - CRM

Spatial Variation: Absolute Differences by Height and Age Difference = Jenkins - CRM

Exceptions: Forest types where Jenkins < CRM • 0. 4% of study area -

Exceptions: Forest types where Jenkins < CRM • 0. 4% of study area - (19, 026 ha) • Abies amabilis/Chamaecyparis nootkatensis (384 ha) • Populus tremuloides/Acer macrophyllum (2, 330 ha) • Alnus rubra/Tsuga heterophylla (4, 967 ha) • Arbutus menziesii (4, 818 ha) • Larix occidentalis/Pinus ponderosa (168 ha) • Pinus monticola (494 ha) • Pseudotsuga menziesii/Fraxinus latifolia (1, 944 ha) • Pinus lambertiana/Pseudotsuga menziesii (3, 920 ha)

Height Class Distribution Ratio Difference

Height Class Distribution Ratio Difference

Height Class

Height Class

Age Class Distribution Ratio Difference

Age Class Distribution Ratio Difference

Vegetation Class Distribution

Vegetation Class Distribution

Vegetation Class Comparison Ratio of Jenkins/CRM Difference Jenkins-CRM

Vegetation Class Comparison Ratio of Jenkins/CRM Difference Jenkins-CRM

Ownership Class Distribution

Ownership Class Distribution

Ownership Class Comparison Ratio of Jenkins/CRM Difference Jenkins-CRM

Ownership Class Comparison Ratio of Jenkins/CRM Difference Jenkins-CRM

Conclusions • Overall difference between methods is 18% but there is significant spatial variation

Conclusions • Overall difference between methods is 18% but there is significant spatial variation (up to 31% in young, open stands). • Jenkins biomass > CRM biomass, especially in younger, shorter, more open stands on private lands.

Conclusions • Absolute differences are smaller in these lower biomass locations, but contribution is

Conclusions • Absolute differences are smaller in these lower biomass locations, but contribution is important due to large area. Stand Height Stand Age

Conclusions • Neither approach is inherently “correct”. – Incorporation of regionally-tailored volume equations within

Conclusions • Neither approach is inherently “correct”. – Incorporation of regionally-tailored volume equations within a nationally-consistent framework is an improvement for spatially explicit purposes. • Need additional scales of validation, including Lidar-derived biomass estimates (with “local” allometric equations).

Conclusions • Implications for strict accounting purposes AND mapping applications. • Careful equation selection

Conclusions • Implications for strict accounting purposes AND mapping applications. • Careful equation selection in highly disturbed landscapes (young, short, open stands). • Temporal considerations: Jenkins would potentially over-estimate biomass (relative to CRM) in post-disturbance, regenerating stands.

Thank You. Questions? Contact me at: spowell@montana. edu (406) 994 -5017

Thank You. Questions? Contact me at: [email protected] edu (406) 994 -5017