Defining Mixed Species Maximum Density Technical Advisory Meeting Spokane, WA 25 June 2014

Transcription

1 L n T P A Defining Mixed Species Maximum Density Technical Advisory Meeting Spokane, WA 25 June L n _ Q M D

2 The utility of the normal stand as a reference Thinning meets numerous objectives: improved economic value Increased resistance to wildfire, drought, pests and disease Thinning is prescribed based on a maximum reference density Typically define maximum density for single species

3 Ln(TPA) Progress on single species density models Modeled SDImax for 4 species in the INW: DF, GF, PP, WL Based on IFTNC Database Developed Predictive layers Target species proportion always remains in models predicting maximum stand density Ponderosa Pine SDImax = 475 Ln(QMD) Ln(QMD)

4 P e rc e n t o f a ll s ta n d s Background The challenge of mixed species stands Mixed species stands are typical in the intermountain region Maximum density for mixed stands is difficult to define. But still necessary for management 6 0 M a n a g e d F o r e s t S ta n d s D o m in a te S p e c ie s B a s a l A re a (% )

5 Previous approaches to mixed-stand SDImax Weighted sums Lowest component Two-species conceptual model Difficult to model SDImax for all possible combinations SDImax A Diversity-productivity inhibition SDImax B

6 Proportion of the basal area (PBA) is always an important factor in single species models Shifts in Density by Factor By subtraction, the PBA of other species in stand have equal and opposite impact

7 S D Im a x (T P A ) Successional climax species have the greatest stocking potential Single species models predict greater SDImax for the climax species than others A B G R S e r ie s P S M E S e r ie s 850 stands 586 stands P P D F G F P P D F G F M o d e le d S ta n d T y p e

8 L n T P A IFTNC Mixed Species Model Remove species filters Predict maximum density based on the entire IFTNC database Used as reference to compare single species models L n _ Q M D

9 S D Im a x (T P A ) Mixed stand likely have higher carrying capacity than pure stands More likely on PSME series than on ABGR series Predicted SDImax for mixed stands is more variable than for pure stands A B G R S e r ie s P S M E S e r ie s 850 stands 586 stands P P D F G F M x d P P D F G F M x d M o d e le d S ta n d T y p e

10 S D Im a x (T P A ) Site quality effect are not greater than species effects Predicted SDImax for pine and Douglas-fir are similar on either veg series Yet there are strong species differences Variation in mixed stands is probably due to species mix rather than site variation A B G R S e r ie s P S M E S e r ie s 850 stands 586 stands 0 P P D F G F M x d P P D F G F M x d M o d e le d S ta n d T y p e

11 How do we determine SDImax for mixed stands? The all-species mixed stand model is a lumped average suggests individual species are important How to separate species in the mixed stand model? Wood specific gravity approach Multiple regression approach

12 Potential mixed-species model S D Im a x (T P A ) SDImax is related to wood density R 2 = W c o n ife r S P in e s T e ak R e d w o o d S p e c ific G ra v ity Table 2, Dean and Baldwin 1996 FEM 81:25

13 Potential mixed-species model 99 th percentile of SDI for 26 specific gravity classes R 2 = ,235 FIA plots Figure 3, Woodall et al 2005 FEM 216:367

14 Application of the specific gravity approach Relative Density RD = SDI / SDImax Where SDImax = SG mean Woodall et al 2006 FEM 226:368

15 ln T P A ln T P A 4 of 13 specific gravity classes represent individual species GF S G = PP S G = DF S G = WL S G = L n _ Q M D L n _ Q M D

16 Potential mixed-species model 99 th percentile of SDI for 13 specific gravity classes IF T N C D a ta b a s e 1 0 1,4 4 3 p lo ts S D I R 2 = M e a n S p e c ific g ra v ity (p e r c la s s )

17 Stochastic Frontier Regression Model: Ln(TPA) = α + β 1 *Ln(QMD) + β 2-i *(Factors) + e Proc QLIM in SAS Single-Species Factors Mixed-Species Factors Basal Area Rock Type Elevation Aspect ADI DD5 FFP MTCM SMRSPRPB Specific Gravity Rock Type Elevation Aspect ADI DD5 FFP MTCM SMRSPRPB

18 S D Im a x c h a n g e (% ) Specific gravity effect on SDImax 5 0 M ix e d S p e c ie s M T C M D D 5 S p e c Ific G r a v ity A D I E le v a tio n F F P R o c k _ T y p e A s p e c t S M R S P R P B

19 E s tim a te d S D Im a x (T P A ) Potential mixed-species model Variation in mixed species model and specific gravity A B G R S e r ie s P S M E S e r ie s A B G R S e r ie s P S M E S e r ie s P P D F G F M x d P P D F G F M x d M o d e le d S ta n d T y p e W D P G W D P G L F P F L F P F M o d e le d A v e ra g e S p e c ific G ra v ity

20 Addressing mixed species by including variables for each species DF BA ratio GF BA ratio PP BA ratio WL BA ratio

21 Maximum Stand Density Model Comparisons at 100% BA

22 Maximum Stand Density Model Comparisons at 25% BA

23 Conclusions Forest site carrying capacity is essential for identifying management thresholds Species-specific size-density relations are available Half of Inland Northwest managed forest stands include more than one species Mixed-species size-density relations are not available Approaches used previously are summations, approximations or 2-species mixes

24 Conclusions Mixed species SDImax models were developed using IFTNC DB Mixed-species models are more variable than single species Variation in mixed species stands is likely due to range of species and sites included Average specific gravity of species mix holds promise for identifying SDImax on individual sites Properly implemented generalized model for mixed species may replace numerous single-species models

25 Validating SDImax models 99 th percentiles as proxy for SDImax Follow trend of density in long-term data Identify stands with a range of mortality rates Comparing FVS predictions to actual stocking while changing maximum density setting

26 E s tim a te d S D Im a x (T P A ) 99 th percentiles as proxy for SDImax Species ranks remain consistent Magnitude is similar Under estimate SDImax for early seral species A B G R S e r ie s P S M E S e r ie s 9 9 th P e r c e n tile P P D F G F M x P P D F G F M x P P D F G F M x M o d e le d S ta n d T y p e

27 9 9 th p e rc e n tia l o f S D I 99 th percentiles as proxy for SDImax SSImax is highest at low BA ratios Tends to decline as one species becomes dominate Steepest decline is in WL and DF, shallowest for GF D F G F P P W L Represents mixed stand in previous slide B a s a l A re a R a tio Represents Pure stands in previous slide

28 Follow trend of density in long-term data Tree growth continues until carrying capacity is reached Stressed trees succumb to various agents at immenent mortality through the self thinning line Predicted SDImax should exceed the zone of mortality Drew and Flewelling 1977

29 Rotation for different spacing