boundaryFit: Fit multiple boundary lines.

In TRADER: Tree Ring Analysis of Disturbance Events in R

Description

Fit multiple boundary lines, write their results and choose the best one.

Usage

boundaryFit(boundaries, x, y, boundary = NULL, prefix = "bo", 
  store = TRUE, storedev = pdf, initNLS = NULL)

Arguments

boundaries	Data frame with segments (x-axis) and tops(y-axis).
x	x coordinates of all priors.
y	y coordinates of all priors.
boundary	Own boundary line function of one argument, eg. boundary=function(x) {5.0067*exp(-0.664*x)}
prefix	Prefix of saved files.
store	If to save figures.
storedev	Format for saving the graphical outputs, eg. pdf or jpeg.
initNLS	Vector for initialization of start values for nls (set a,b,c,d for nls).

Details

Boundary-line method scales the percent growth change of Nowacki & Abrams (1997) according to growth rate prior to disturbance. In their example, Black & Abrams (2003) defined moderate and major releases as those falling within 20-49.9%, and 50-100% of the boundary line, respectively. Advantage of the boundary-line is standardization, which takes into account the relationships among tree age, size, and canopy class determining radial growth rate (Black et al. 2004). On the downside, Black et al. (2009) suggest approximately 50000 ring width measurements is necessary for boundary line determination for a given species (Black et al. 2009).

Value

Return list object with

fun	Fitted function (boundary line).
rsq	R square of the fit.
bestModel	Best fitted model.

Note

Check reference.

Author(s)

Pavel Fibich <[e-mail]>, Jan Altman <[e-mail]>, Tuomas Aakala <[e-mail]>, Jiri Dolezal <[e-mail]>

References

Black, B.A. & Abrams, M.D. (2003) Use of boundary-line growth patterns as a basis for dendroecological release criteria. Ecological Applications, 13, 1733-1749.
Black, B.A., Abrams, M.D., Gagen, M., Daniels, L.D., Kipfmueller, K.F., Speer, J.H. & Anchukaitis, K.J. (2004) Development and application of boundary-line release criteria. Dendrochronologia, 22, 31-42.
Black, B.A., Abrams, M.D., Rentch, J.S. & Gould, P.J. (2009) Properties of boundary-line release criteria in North American tree species. Annals of Forest Science, 66.

See Also

boundaryGet, plotBoundary, nls

Examples

data(relData)
bo<-boundaryGet(relData1)
bofit<-boundaryFit(bo$bo,bo$x,bo$y)

plotBoundary(bo$bo,bo$x,bo$y,boundary=bofit$fun,rsq=bofit$rsq)
plotBoundary(bo$bo,bo$x,bo$y,boundary=function(x) {5.0067*exp(-0.664*x)})

Example output

[1] "--Summary of y=a+bx fit."

Call:
lm(formula = tops ~ segments, data = boundaries)

Residuals:
      1       2       3       4       5       6       7 
-2.1637  1.7473  1.3473 -0.1650 -0.1749 -0.2988 -0.2922 

Coefficients:
            Estimate Std. Error t value Pr(>|t|)  
(Intercept)   2.6631     1.0656   2.499   0.0545 .
segments     -0.7266     0.5287  -1.374   0.2277  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 1.399 on 5 degrees of freedom
Multiple R-squared:  0.2742,	Adjusted R-squared:  0.129 
F-statistic: 1.889 on 1 and 5 DF,  p-value: 0.2277

[1] "--Summary of y=a+bx+cx^2 fit."

Call:
lm(formula = tops ~ segments + I(segments^2), data = boundaries)

Residuals:
      1       2       3       4       5       6       7 
-1.2627  1.7473  0.8067 -0.8858 -0.7155 -0.2988  0.6088 

Coefficients:
              Estimate Std. Error t value Pr(>|t|)
(Intercept)     1.1766     1.5661   0.751    0.494
segments        1.7961     2.0898   0.859    0.439
I(segments^2)  -0.7208     0.5796  -1.244    0.282

Residual standard error: 1.328 on 4 degrees of freedom
Multiple R-squared:  0.4765,	Adjusted R-squared:  0.2148 
F-statistic: 1.821 on 2 and 4 DF,  p-value: 0.274

[1] "--Summary of y=ae^bx fit."

Formula: tops ~ a * exp(b * segments)

Parameters:
  Estimate Std. Error t value Pr(>|t|)
a   2.4721     1.4860   1.664    0.157
b  -0.3441     0.4231  -0.813    0.453

Residual standard error: 1.48 on 5 degrees of freedom

Number of iterations to convergence: 19 
Achieved convergence tolerance: 7.85e-06

[1] "y=c+ae^bx nls error: step factor 0.000488281 reduced below 'minFactor' of 0.000976562"
[1] "y=c+dx+ae^bx nls error: singular gradient"
[1] "y=ae^bx+ce^dx nls error: Missing value or an infinity produced when evaluating the model"
[1] "--Summary of y=a+blog(x) fit."

Formula: tops ~ a + b * log(segments)

Parameters:
  Estimate Std. Error t value Pr(>|t|)  
a   1.5195     0.6389   2.378   0.0633 .
b  -0.4241     0.7248  -0.585   0.5839  
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 1.588 on 5 degrees of freedom

Number of iterations to convergence: 1 
Achieved convergence tolerance: 8.688e-09

[1] "--Summary of y=a+bx+clog(x)+dxlog(x) fit."

Call:
lm(formula = tops ~ segments + log(segments) + segments:log(segments), 
    data = boundaries)

Residuals:
        1         2         3         4         5         6         7 
-0.007628  0.001627  0.189574 -0.418157  0.170181  0.193449 -0.129045 

Coefficients:
                       Estimate Std. Error t value Pr(>|t|)   
(Intercept)              30.061      4.285   7.016  0.00595 **
segments                -26.547      4.156  -6.388  0.00777 **
log(segments)            14.067      1.957   7.189  0.00555 **
segments:log(segments)   10.384      1.852   5.606  0.01121 * 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.313 on 3 degrees of freedom
Multiple R-squared:  0.9782,	Adjusted R-squared:  0.9564 
F-statistic: 44.86 on 3 and 3 DF,  p-value: 0.005431

[1] "The fitted boundary line summary!"
[1] "Logarithmic model y=a+bx+clog(x)+dxlog(x) was the best!"

Call:
lm(formula = tops ~ segments + log(segments) + segments:log(segments), 
    data = boundaries)

Residuals:
        1         2         3         4         5         6         7 
-0.007628  0.001627  0.189574 -0.418157  0.170181  0.193449 -0.129045 

Coefficients:
                       Estimate Std. Error t value Pr(>|t|)   
(Intercept)              30.061      4.285   7.016  0.00595 **
segments                -26.547      4.156  -6.388  0.00777 **
log(segments)            14.067      1.957   7.189  0.00555 **
segments:log(segments)   10.384      1.852   5.606  0.01121 * 
---
Signif. codes:  0 '***' 0.001 '**' 0.01 '*' 0.05 '.' 0.1 ' ' 1

Residual standard error: 0.313 on 3 degrees of freedom
Multiple R-squared:  0.9782,	Adjusted R-squared:  0.9564 
F-statistic: 44.86 on 3 and 3 DF,  p-value: 0.005431

png 
  2 
png 
  2 

TRADER documentation built on May 2, 2019, 9:02 a.m.