| Title: | Tree Taper Curves and Sorting Based on 'TapeR' |
|---|---|
| Description: | Providing new german-wide 'TapeR' Models and functions for their evaluation. Included are the most common tree species in Germany (Norway spruce, Scots pine, European larch, Douglas fir, Silver fir as well as European beech, Common/Sessile oak and Red oak). Many other species are mapped to them so that 36 tree species / groups can be processed. Single trees are defined by species code, one or multiple diameters in arbitrary measuring height and tree height. The functions then provide information on diameters along the stem, bark thickness, height of diameters, volume of the total or parts of the trunk and total and component above-ground biomass. It is also possible to calculate assortments from the taper curves. For diameter and volume estimation, uncertainty information is given. |
| Authors: | Christian Vonderach [aut, cre], Edgar Kublin [aut], Gerald Kändler [aut] |
| Maintainer: | Christian Vonderach <[email protected]> |
| License: | BSD_2_clause + file LICENSE |
| Version: | 0.13.0 |
| Built: | 2024-11-26 11:26:20 UTC |
| Source: | https://gitlab.com/vochr/tapes |
estimate minimum cutting diameter based on diameter in breast height based on the functions implemented in BDAT
Az(sp, dbh)Az(sp, dbh)
sp |
Bdat species code [1;36], integer |
dbh |
vector of diameter in breast height, numeric |
the implemented BDAT function and parameters are used. Not all BDAT-species possess their own parameters, hence most of them are matched to one of the main tree species, especially in deciduous tree species (only parameters for beech and oak are available).
vector of minimum cutting diameter [cm].
sp <- 1 dbh <- 30 Az(sp, dbh)sp <- 1 dbh <- 30 Az(sp, dbh)
function for mapping the 36 tree species to several internal functions
BaMap(Ba = NULL, type = NULL)BaMap(Ba = NULL, type = NULL)
Ba |
BDAT tree number |
type |
a number referring to the type to be mapped |
c.f. BDAT source code, line 7622, data block Ban(36, 7) type 1: Schaftform // taper form type 2: Rinde // bark type 3: Durchschnittliche Aufarbeitungsgrenze (nach EST) //average cutting diameter type 4: Höhe unverwertbares Derbholz // percentage non-merchantable coarse wood type 5: durchschnittlicher Astdurchmesser in der Krone // average branch diameter inside crown type 6: BWI-Biomasse-Funktionen // NFI-biomass functions according to Riedel & Kändler (2017) type 7: kompartimentweise Biomassefunktionen // component biomass functions according to Vonderach et al (2018) type 8: Zuordnung zu volfao // Mapping to volume according to FAO (FIX: mapping still temporary) Not included: volume tables according to Grundner and Schwappach as well as volume tables according to Krenn for small trees below 10cm dbh
value(s), either a scalar, vector or matrix, with respect to tree species mapping to functions
BaMap(1,1) # which taper form for Norway spruce BaMap(15,1) # which taper form for European Beech BaMap(15,2) # which bark equation for European Beech BaMap(,1) # return all taper form mappings BaMap(1,) # return all mappings for Norway spruce BaMap() # return all mappings BaMap(, 6) # biomass mapping BaMap(, 7) # component biomass functions BaMap(, 8) # mapping for Vol_FAOBaMap(1,1) # which taper form for Norway spruce BaMap(15,1) # which taper form for European Beech BaMap(15,2) # which bark equation for European Beech BaMap(,1) # return all taper form mappings BaMap(1,) # return all mappings for Norway spruce BaMap() # return all mappings BaMap(, 6) # biomass mapping BaMap(, 7) # component biomass functions BaMap(, 8) # mapping for Vol_FAO
Function returns double bark thickness according to Altherr et al. 1974/75/76/78/79
bark(Ba, Dm, relH)bark(Ba, Dm, relH)
Ba |
tree species according to BDAT, cf. |
Dm |
diameter for which double bark thickness is requested |
relH |
relative height of Dm inside stem |
Function re-implemented according to Subroutine RINDE(Hhrel,Kw,Ri,Hsga,Zo), BDAT-fortran Code line 5691ff. No Functions for (historic) Heilbronner Sortierung implemented.
NB: to avoid negative double bark thickness, such values are constraint to
zero. Additionally, diameter after bark reduction might not be smaller than
zero, hence double bark thickness is reduce to Dm.
double bark thickness [cm]
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1974). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Kiefer, Buche, Hainbuche, Esche und Roterle. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1975). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Europäische Lärche, Japanische Lärche, Schwarzkiefer, Stieleiche, Traubeneiche, Roteiche, Bergahorn und Linde. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1976). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Weymouthskiefer, Robinie, Bergulme, Birke, Marilandica-Pappel und Robusta-Pappel. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1978). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Fichte, Tanne, Douglasie und Sitka-Fichte. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1979). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Neupotz-Pappel, Regenerata-Pappel, Kirsche, Spitzahorn, Feldahorn, Aspe, Weide, Flatterulme, Tulpenbaum u. Elsbeere. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
bark(1, 30, .1) bark(11, 4, .1) # zero instead of -0.2497bark(1, 30, .1) bark(11, 4, .1) # zero instead of -0.2497
Prediction of total above-ground biomass for trees defined via species, dbh, d03 and height
biomass(spp, d13, d03, h)biomass(spp, d13, d03, h)
spp |
vector of species code for biomass function [1;18] |
d13 |
vector of diameter in breast height in centimeter |
d03 |
vector of diameter in 30% of tree height in centimeter |
h |
vector of height of trees |
code taken from BDAT (Koeff.f).
a vector of total above-ground biomass
Riedel, T. and G. Kaendler (2017). "Nationale Treibhausgasberichterstattung: Neue Funktionen zur Schätzung der oberirdischen Biomasse am Einzelbaum." Forstarchiv 88(2): 31-38.
calculate variance-covariance matrix for volume segments from the estimated diameter and uncertainty information from TapeR-model
calcVCOVsekVol(estD, kovD, estL)calcVCOVsekVol(estD, kovD, estL)
estD |
vector of estimated diameter, numeric |
kovD |
variance-covariance-matrix of the estimated diameter, numeric |
estL |
vector of segment length, numeric |
Calculations according to rules for products and sums of variances
variance-covariance matrix of the segment volume
generate and/or check validity of biomass function component names
check_Comp(component = NULL)check_Comp(component = NULL)
component |
vector of biomass component names, see details |
If component is NULL, by default, component name
for total aboveground biomass is returned. If is all,
then all available component names are returned.
stw: stump wood
stb: stump bark
sw: solid wood with diameter above 7cm over bark
sb: bark of component sw
fwb: fine wood incl. bark
ndl: needles
agb: total aboveground biomass
a vector of component names
## Not run: TapeS:::checkComp() TapeS:::checkComp("AGB") TapeS:::checkComp("biomass") ## End(Not run)## Not run: TapeS:::checkComp() TapeS:::checkComp("AGB") TapeS:::checkComp("biomass") ## End(Not run)
monotonicity check for taper curve
check_monotonicity(obj, Rfn = NULL)check_monotonicity(obj, Rfn = NULL)
obj |
object of class 'tprTrees' |
Rfn |
Rfn setting for residuals error matrix, defaults to
|
Taper curves are required to decrease monotonically. To avoid the
evaluation of non-monotone taper curves, a check is done through the
constructor function and an indicator (monotone) is set for each tree
stored inside the tprTrees-class. As the data has been check on validity
before this function is applied, we can use the tpr*-functions to evaluate
the taper curve and its monotonicity.
The check is done via comparison of the expected diameters along the trunk in
1m-steps and its sorted (monotonically decreasing) version using
identical.
vector of logicals, same length as spp.
coercion functions to make NFI, segment and 'BDAT' data available as 'tprTrees' objects
nfi_as_tprtrees(nfi, mapping = NULL) seg_as_tprtrees(seg, mapping = NULL) bdat_as_tprtrees(bdat)nfi_as_tprtrees(nfi, mapping = NULL) seg_as_tprtrees(seg, mapping = NULL) bdat_as_tprtrees(bdat)
nfi |
data.frame with tree measurements as provided by german NFI |
mapping |
mapping of column names |
seg |
data.frame with measured tree segments, see details. |
bdat |
data.frame holding data to process with rBDAT |
The coerced data is automatically checked for validity by the class
constructor. For available species codes see tprSpeciesCode.
When coercing NFI (National forest inventory, BWI) data, one need
to provide the columns BaTpr (species code), Bhd (Dbh, [mm]),
D03, [mm] (diameter in 30% of tree height) and Hoehe (tree
height, [dm]). Optionally, one can provide H1 (measurement height of
Bhd, [dm]), H2 (measurement height of D03, [dm]) as well
as sHt (measurement error of tree height, i.e. standard deviation [dm]);
otherwise these are assumed to be 1.3m, 30% of tree height and 0 (zero),
respectively.
Additionally, the NFI database stores diameter as [mm] and height as [dm]; it
is *not* necessary to transform to [cm] and [m], as the function does this.
Equally, sHt [dm] is transformed to sHt [m].
Keep in mind that species codes of NFI are different from the taper models for historical reasons (c.f. BDAT). Use the NFI table ('x_Ba') to map species codes beforehand (see examples).
Sectional measurements provide more information about the trunk of
a tree and are usually stored in a different way. They exhibit an arbitrary
amount of diameter measurements which also might vary from tree to tree.
Hence, seg_as_tprtrees expects a data.frame with columns
Id, BaTpr (species code), Dm (diameter measured, [cm]),
Hm (height of Dm, [m]) and optionally Ht (height of
tree, [m]). Tree height Ht can be included to
Dm-Hm-pairs with Dm being zero
(e.g. Dm=0, Hm=25). If Ht is given, it gains priority.
coercing object of class 'datBDAT' from R-Package "rBDAT" into class 'tprTrees'
an object of class 'tprTrees', see tprTrees-class
nfi_as_tprtrees(): coercion of German NFI data
seg_as_tprtrees(): coercion of segmented data to class 'tprTrees'
bdat_as_tprtrees(): coercion of bdat data
tprTrees-class, tprTrees,
tprSpeciesCode
# NFI data usually stored as integer and units: diameter=[mm] and height=[dm] nfi <- data.frame(BaTpr=1L, Bhd=300L, D03=270L, Hoehe=250L) tpr <- nfi_as_tprtrees(nfi) tpr tpr@sHt # defaults to 0 # one can provide measurement heights explicitly nfi <- data.frame(spp=1, Bhd=300, H1=12, D03=270, H=250) nfi_as_tprtrees(nfi, mapping=c(spp="BaTpr", H="Hoehe")) # measurement error in height nfi <- data.frame(BaTpr=1L, Bhd=300L, D03=270L, Hoehe=250L, sHt=15) tpr <- nfi_as_tprtrees(nfi) tpr@sHt ## coercing sectional measurements data(DxHx.df, package = "TapeR") DxHx.df$BaTpr <- 1 # Norway spruce segtprtrees <- seg_as_tprtrees(DxHx.df, mapping=c(Dx="Dm", Hx="Hm")) ## extract tree height from Dm-Hm measurements if not given explicitly DxHx.df$Ht <- NULL # remove height, as already included with Dm=0 segtprtrees <- seg_as_tprtrees(DxHx.df, mapping=c(Dx="Dm", Hx="Hm")) segtprtrees if(require(rBDAT)){ bdt <- buildTree(list(spp=1, D1=30, D2=28, H2=7, H=25)) bdat_as_tprtrees(bdt) }# NFI data usually stored as integer and units: diameter=[mm] and height=[dm] nfi <- data.frame(BaTpr=1L, Bhd=300L, D03=270L, Hoehe=250L) tpr <- nfi_as_tprtrees(nfi) tpr tpr@sHt # defaults to 0 # one can provide measurement heights explicitly nfi <- data.frame(spp=1, Bhd=300, H1=12, D03=270, H=250) nfi_as_tprtrees(nfi, mapping=c(spp="BaTpr", H="Hoehe")) # measurement error in height nfi <- data.frame(BaTpr=1L, Bhd=300L, D03=270L, Hoehe=250L, sHt=15) tpr <- nfi_as_tprtrees(nfi) tpr@sHt ## coercing sectional measurements data(DxHx.df, package = "TapeR") DxHx.df$BaTpr <- 1 # Norway spruce segtprtrees <- seg_as_tprtrees(DxHx.df, mapping=c(Dx="Dm", Hx="Hm")) ## extract tree height from Dm-Hm measurements if not given explicitly DxHx.df$Ht <- NULL # remove height, as already included with Dm=0 segtprtrees <- seg_as_tprtrees(DxHx.df, mapping=c(Dx="Dm", Hx="Hm")) segtprtrees if(require(rBDAT)){ bdt <- buildTree(list(spp=1, D1=30, D2=28, H2=7, H=25)) bdat_as_tprtrees(bdt) }
Function extracts pre-defined diameters, e.g. dbh (in 1.3m) or
D03 (in 30% of tree height) for a tprTrees-object
Dbh(obj) Bhd(obj) D13(obj) D7(obj) D03(obj) D005(obj)Dbh(obj) Bhd(obj) D13(obj) D7(obj) D03(obj) D005(obj)
obj |
a object of class 'tprTrees' |
a wrapper around tprDiameter to calculate specifically
defined diameters like diameter in breast height (dbh), diameter in 7m above
ground or in 5% and 30% of tree height.
diameter(s) in predefined heights
Dbh(): wrapper to calculate diameter in breast height
Bhd(): German alias for function Dbh
D13(): Height specific alias for function Dbh
D7(): Function to calculate diameter over bark in 7m above ground
D03(): Function to calculate diameter over bark in 30% of tree height
D005(): Function to calculate diameter over bark in 5% of tree height
t <- tprTrees() Dbh(t) # diameter in breast height (i.e. 1.3m) Bhd(t) # same, german named function name D13(t) # same, height related function name D005(t) # diameter in 5% of tree height D7(t) # diameter in height of 7m D03(t) # diameter in 30% of tree heightt <- tprTrees() Dbh(t) # diameter in breast height (i.e. 1.3m) Bhd(t) # same, german named function name D13(t) # same, height related function name D005(t) # diameter in 5% of tree height D7(t) # diameter in height of 7m D03(t) # diameter in 30% of tree height
Functional equivalent to E_HDx_HmDm_HT.f,
finding the height of a given diameter *without* bark, i.e. double bark
thickness needs to be added on top of given diameter to find appropriate
height.
E_HDxoR_HmDm_Ht.f(DxoR, Hm, Dm, mHt, sHt = 0, par.lme, Rfn = NULL, ...) HxoR_root.f(Hx, DxoR, Hm, Dm, mHt, sHt, par.lme, Rfn, ...)E_HDxoR_HmDm_Ht.f(DxoR, Hm, Dm, mHt, sHt = 0, par.lme, Rfn = NULL, ...) HxoR_root.f(Hx, DxoR, Hm, Dm, mHt, sHt, par.lme, Rfn, ...)
DxoR |
Scalar. Diameter under bark for which to return height. |
Hm |
Numeric vector of stem heights (m) along which diameter
measurements were taken for calibration. Can be of length 1. Must be of same
length as |
Dm |
Numeric vector of diameter measurements (cm) taken for calibration.
Can be of length 1. Must be of same length as |
mHt |
Scalar. Tree height (m). |
sHt |
Scalar. Standard deviation of stem height. Can be 0 if height was measured without error. |
par.lme |
List of taper model parameters obtained by
|
Rfn |
setting for residuals error matrix, defaults to |
... |
not currently used |
Hx |
height at which taper curve is evaluated |
finds height of given diameter via uniroot.
A scalar. Estimated height (m) given a diameter without bark.
HxoR_root.f(): function to be searched
tmp <- tprTrees() spp <- spp(tmp) Hm <- Hm(tmp) Dm <- Dm(tmp) H <- Ht(tmp) SKP <- TapeS:::SKPar sppSK <- BaMap(spp, 1) # tree species for taper curve ## diameter in 5m height TapeR::E_DHx_HmDm_HT.f(c(5, 10), Hm, Dm, mHt=H, sHt = 0, par.lme = SKP[[sppSK]])$DHx (D5m <- TapeR::E_DHx_HmDm_HT.f(c(5, 10), Hm, Dm, mHt=H, sHt = 0, par.lme = SKP[[sppSK]])$DHx) ## bark thickness of diameter in 5m height (RiD5m <- bark(c(1,1), Dm = D5m, relH = c(5, 10)/H)) ## find height of diameter without bark, which should be 5m d5mub <- D5m - RiD5m E_HDxoR_HmDm_Ht.f(DxoR = d5mub, Hm = Hm, Dm = Dm, mHt = H, sHt = 0, par.lme = SKP[[sppSK]])tmp <- tprTrees() spp <- spp(tmp) Hm <- Hm(tmp) Dm <- Dm(tmp) H <- Ht(tmp) SKP <- TapeS:::SKPar sppSK <- BaMap(spp, 1) # tree species for taper curve ## diameter in 5m height TapeR::E_DHx_HmDm_HT.f(c(5, 10), Hm, Dm, mHt=H, sHt = 0, par.lme = SKP[[sppSK]])$DHx (D5m <- TapeR::E_DHx_HmDm_HT.f(c(5, 10), Hm, Dm, mHt=H, sHt = 0, par.lme = SKP[[sppSK]])$DHx) ## bark thickness of diameter in 5m height (RiD5m <- bark(c(1,1), Dm = D5m, relH = c(5, 10)/H)) ## find height of diameter without bark, which should be 5m d5mub <- D5m - RiD5m E_HDxoR_HmDm_Ht.f(DxoR = d5mub, Hm = Hm, Dm = Dm, mHt = H, sHt = 0, par.lme = SKP[[sppSK]])
Function calculates expected tree height given diameter in breast height and species code
estHeight(d13, sp, qtl = NULL)estHeight(d13, sp, qtl = NULL)
d13 |
numeric vector of diameter in breast height [cm] |
sp |
TapeS species code, see also |
qtl |
desired quantile of height, either NULL (LS-regression) or one of 17, 50, 83 (quantile regression) |
Function evaluates the expected value of a Pettersen-Height Function
based on diameter in breast height and tree species code.
The Pettersen-Function () was fitted
on NFI 3 (BWI 3) data, using the main stand only.
d13 and sp should be of equal length or one of it can
be > 1 if the other is of length 1. Then, the shorter object will be extended
to match the length of the longer object. See examples.
The quantile option return tree height at quantiles 17, 50 or 83. If
qtl is NULL, the result of a nonlinear least-squares regression is
provided.
a vector of same length as d13 or sp, with tree
height in [m].
sp <- 1 d13 <- 30 estHeight(d13, sp) sp <- 1 d13 <- seq(15, 50, 5) estHeight(d13, sp) sp <- 1:36 d13 <- 30 estHeight(d13, sp)sp <- 1 d13 <- 30 estHeight(d13, sp) sp <- 1 d13 <- seq(15, 50, 5) estHeight(d13, sp) sp <- 1:36 d13 <- 30 estHeight(d13, sp)
function extracts the percentage of unusable coarse wood according to species (beech, oak), diameter class and cutting diameter
fnUnvd(ba = NULL, dm = NULL, cd = NULL)fnUnvd(ba = NULL, dm = NULL, cd = NULL)
ba |
tree species index; see details |
dm |
diameter class; see details |
cd |
cutting diameter; see details |
Function extracts the percentage of unusable coarse wood according to three parameters: (i) tree species, which is 1 for using beech models and 2 for using the oak model; (ii) the 2cm-diameter class (from 8 and 60cm) and (iii) the cutting diameter ranging from 8 to 40cm.
Kublin and Scharnagl (1988): Verfahrens- und Programmbeschreibung zum BWI-Unterprogramm BDAT. FVA-BW 1988. ISSN: 0178-3165.
evaluates tariff functions to estimate taper form, i.e. quotient of d03 by d005
FormTariff(spp, Dbh, Ht, inv)FormTariff(spp, Dbh, Ht, inv)
spp |
species code of |
Dbh |
diameter of considered tree at 1.3m above ground [cm] |
Ht |
tree height of considered tree [m] |
inv |
indicator for inventory (0=TapeS taper curve models, 1=NFI1, 2=NsoG, 3=IS08, 4=NFI3, 5=BDAT) |
quotient of d03 / d005 [unitless]
c.f. rBDAT::getForm respectively BDAT source code FormTarif.f
## dont't run spp <- 15 Dbh <- 30 Ht <- 27 FormTariff(spp, Dbh, Ht, 0) FormTariff(spp=c(1:2), Dbh=c(30, 30), Ht=c(27, 24), inv=0) if(require("rBDAT")){ FormTariff(spp, Dbh, Ht, 0) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=0) # different taper curves! FormTariff(spp, Dbh, Ht, 1) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=1) # identical FormTariff(spp, Dbh, Ht, 2) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=2) # identical FormTariff(spp, Dbh, Ht, 3) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=3) # identical FormTariff(spp, Dbh, Ht, 4) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=4) # identical }## dont't run spp <- 15 Dbh <- 30 Ht <- 27 FormTariff(spp, Dbh, Ht, 0) FormTariff(spp=c(1:2), Dbh=c(30, 30), Ht=c(27, 24), inv=0) if(require("rBDAT")){ FormTariff(spp, Dbh, Ht, 0) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=0) # different taper curves! FormTariff(spp, Dbh, Ht, 1) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=1) # identical FormTariff(spp, Dbh, Ht, 2) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=2) # identical FormTariff(spp, Dbh, Ht, 3) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=3) # identical FormTariff(spp, Dbh, Ht, 4) rBDAT::getForm(list(spp=spp, D1=Dbh, H1=1.3, H=Ht), inv=4) # identical }
Function to provide model coefficients for Pettersen-height model
HtCoef(sp = NULL, qtl = NULL)HtCoef(sp = NULL, qtl = NULL)
sp |
BDAT species code, could be NULL then all coefficients are returned |
qtl |
quantile, either NULL or 17, 50, 83 |
a data.frame with species code and coefficients
Prediction diameter (no variances) for given tree and TapeR-object using BSpline Matrix all in C++
lmeSKEBLUP(xm, ym, xp, par, RV)lmeSKEBLUP(xm, ym, xp, par, RV)
xm |
relative height of measured diameter |
ym |
measured diameter for calibration |
xp |
relative height for which diameter prediction is required |
par |
a TapeR-object (including padded knots vector) |
RV |
numeric vector holding assumed residual variance for each observation |
code implementation in C++ following the code base of TapeR. Bspline matrix code taken from R-package splines to avoid the need of calling R from C.
a list holding several elements, perspectively only the estimated diameter
evaluation of the component biomass functions fit by nonlinear seemingly unrelated regression (NSUR) to estimate absolute or relative component mass
nsur(spp, dbh, ht, sth, d03, kl)nsur(spp, dbh, ht, sth, d03, kl)
spp |
vector of species code for biomass component function of interval
[1;8]; see |
dbh |
vector of diameter in breast height; in centimeter |
ht |
vector of tree heights, in meter |
sth |
vector of stump heights, in meter |
d03 |
vector if diameter in 30% of tree height, in centimeter |
kl |
vector of crown length, i.e. tree height minus height of crown base, in meter |
function to calculate component biomass; functions fitted using
same methodology as in Vonderach et al. (2018) with slightly updated
parameters as in Vonderach and Kändler (2021); species mapping as in
TapeS::BaMap(, type=7);
a numeric matrix holding component biomass
Vonderach, C., G. Kändler and C. F. Dormann (2018). "Consistent set of additive biomass functions for eight tree species in Germany fit by nonlinear seemingly unrelated regression." Annals of Forest Science 75(2): 49. doi:10.1007/s13595-018-0728-4
Vonderach, C. and G. Kändler (2021). Neuentwicklung von Schaftkurven- und Biomassemodellen für die Bundeswaldinventur auf Basis des TapeR-Pakets - Abschlussbericht zum Projekt BWI-TapeR. Freiburg: 150p.
nsur(spp = c(1, 6), dbh = c(30, 30), ht = c(25, 27), sth = c(0.25, 0.27), d03 = c(27, 27), kl = .7*c(25, 27))nsur(spp = c(1, 6), dbh = c(30, 30), ht = c(25, 27), sth = c(0.25, 0.27), d03 = c(27, 27), kl = .7*c(25, 27))
evaluation of the component biomass functions fit by nonlinear seemingly unrelated regression (NSUR) to estimate absolute or relative component mass
nsur2(spp, dbh, ht)nsur2(spp, dbh, ht)
spp |
vector of species code for biomass component function of interval
[1;8]; see |
dbh |
vector of diameter in breast height; in centimeter |
ht |
vector of tree heights, in meter |
simple function from Vonderach et al. (2018) to calculate component
biomass; species mapping as in TapeS::BaMap(, type=7)
a numeric matrix holding component biomass
Vonderach, C., G. Kändler and C. F. Dormann (2018). "Consistent set of additive biomass functions for eight tree species in Germany fit by nonlinear seemingly unrelated regression." Annals of Forest Science 75(2): 49. doi:10.1007/s13595-018-0728-4
nsur2(spp = c(1, 6), dbh = c(30, 30), ht = c(25, 27))nsur2(spp = c(1, 6), dbh = c(30, 30), ht = c(25, 27))
estimate variance components for component biomass functions
NSURvar( data, estBM = NULL, comp = NULL, interval = "confidence", level = 0.95, adjVarPar = TRUE, as.list = TRUE )NSURvar( data, estBM = NULL, comp = NULL, interval = "confidence", level = 0.95, adjVarPar = TRUE, as.list = TRUE )
data |
data / predictors given for prediction by |
estBM |
estimated biomass components for which variance information is
required, given as data.frame, possibly use |
comp |
which components are required, see |
interval |
either |
level |
Tolerance / confidence level, defaults 0.95 |
adjVarPar |
should the variance information be taken from stable models? defaults to TRUE |
as.list |
Should the return value be a list or |
Estimates confidence and prediction intervals according to the methods presented in Parresol (2001).
In case, adjVarPar = TRUE, the models with instable variance estimates
like Silver fir, Scots pine, Maple and Ash are, firstly, fitted by Norway
spruce and European beech, respectively, and, secondly, adjusted to the
expected value of the species specific model by substracting the difference
to the first model. With that, more stable and imho more realistic confidence
and prediction intervals are given. True, this assumes comparability of the
variances between species.
a data.frame with information on lower and upper bound of required interval as well as the (given) estimate and the respective mean squared error
d1 <- seq(42, 56, 2) h <- estHeight(d1, 1) data <- data.frame(spp = 1:8, # from BaMap(1, 7) dbh = d1, ht = h, sth = 0.01*h, D03 = 0.8 * d1, kl = 0.7 * h) estBM <- as.data.frame( nsur(spp = data$spp, dbh = data$dbh, ht = data$ht, sth = data$sth, d03 = data$D03, kl = data$kl) ) estBM$agb <- rowSums(estBM[, -which(colnames(estBM)=="id")]) comp = c("sw", "agb") interval = "confidence" level = 0.95 adjVarPar = TRUE e1 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = FALSE) ## Not run: par(mfrow=c(1, 2)) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0.5*min(e1$agb_ECBM), 1.2*max(e1$agb_ECBM)), las=1, ylab="estimated AGB", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) legend("bottomright", legend=c("Fi", "Ta", "Kie", "Dgl", "Bu", "Ei", "BAh", "Es"), pch=1:8) ## prediction intervals e1 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = FALSE) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0, 2*max(e1$agb_ECBM)), las=1, ylab="estimated AGB", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) legend("topleft", legend=c("Fi", "Ta", "Kie", "Dgl", "Bu", "Ei", "BAh", "Es"), pch=1:8) ## one species, large diameter range spp <- 1 # spruce spp <- 5 # beech spp <- 2 # silver fir spp <- 8 # ash d1 <- seq(7, 80, 2) h <- estHeight(d1, spp) data <- data.frame(spp = spp, dbh = d1, ht = h, sth = 0.01*h, D03 = 0.8 * d1, kl = 0.7 * h) estBM <- as.data.frame( nsur(spp = data$spp, dbh = data$dbh, ht = data$ht, sth = data$sth, d03 = data$D03, kl = data$kl) ) estBM$agb <- rowSums(estBM[, -which(colnames(estBM)=="id")]) comp = c("sw", "agb") interval = "confidence" level = 0.95 adjVarPar = TRUE e1 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = FALSE) par(mfrow=c(1, 2)) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0.5*min(e1$agb_ECBM), 1.2*max(e1$agb_ECBM)), las=1, ylab="estimated AGB", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) ## prediction intervals e1 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = FALSE) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0, 2*max(e1$agb_ECBM)), las=1, ylab="estimated biomass", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) ## End(Not run)d1 <- seq(42, 56, 2) h <- estHeight(d1, 1) data <- data.frame(spp = 1:8, # from BaMap(1, 7) dbh = d1, ht = h, sth = 0.01*h, D03 = 0.8 * d1, kl = 0.7 * h) estBM <- as.data.frame( nsur(spp = data$spp, dbh = data$dbh, ht = data$ht, sth = data$sth, d03 = data$D03, kl = data$kl) ) estBM$agb <- rowSums(estBM[, -which(colnames(estBM)=="id")]) comp = c("sw", "agb") interval = "confidence" level = 0.95 adjVarPar = TRUE e1 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = FALSE) ## Not run: par(mfrow=c(1, 2)) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0.5*min(e1$agb_ECBM), 1.2*max(e1$agb_ECBM)), las=1, ylab="estimated AGB", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) legend("bottomright", legend=c("Fi", "Ta", "Kie", "Dgl", "Bu", "Ei", "BAh", "Es"), pch=1:8) ## prediction intervals e1 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = FALSE) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0, 2*max(e1$agb_ECBM)), las=1, ylab="estimated AGB", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) legend("topleft", legend=c("Fi", "Ta", "Kie", "Dgl", "Bu", "Ei", "BAh", "Es"), pch=1:8) ## one species, large diameter range spp <- 1 # spruce spp <- 5 # beech spp <- 2 # silver fir spp <- 8 # ash d1 <- seq(7, 80, 2) h <- estHeight(d1, spp) data <- data.frame(spp = spp, dbh = d1, ht = h, sth = 0.01*h, D03 = 0.8 * d1, kl = 0.7 * h) estBM <- as.data.frame( nsur(spp = data$spp, dbh = data$dbh, ht = data$ht, sth = data$sth, d03 = data$D03, kl = data$kl) ) estBM$agb <- rowSums(estBM[, -which(colnames(estBM)=="id")]) comp = c("sw", "agb") interval = "confidence" level = 0.95 adjVarPar = TRUE e1 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="confidence", level=0.95, adjVarPar = FALSE) par(mfrow=c(1, 2)) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0.5*min(e1$agb_ECBM), 1.2*max(e1$agb_ECBM)), las=1, ylab="estimated AGB", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) ## prediction intervals e1 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = TRUE) e2 <- TapeS:::NSURvar(data, estBM, comp, interval="prediction", level=0.95, adjVarPar = FALSE) plot(x = data$dbh, y = e1$agb_ECBM, main="adjusted Var-Parameter", pch=data$spp, ylim=c(0, 2*max(e1$agb_ECBM)), las=1, ylab="estimated biomass", xlab = "DBH [cm]") invisible(sapply(1:nrow(e1), function(a){ # a <- 1 # lines(x = rep(data$dbh[a], 2), y = c(e2$agb_lwr[a], e2$agb_upr[a]), # col="blue", lwd=2) rect(xleft = data$dbh[a] - 0.1, xright = data$dbh[a] + 0.1, ybottom = e2$agb_lwr[a], ytop = e2$agb_upr[a], border = "blue") lines(x = rep(data$dbh[a], 2), y = c(e1$agb_lwr[a], e1$agb_upr[a]), col="red", lwd=2) })) ## End(Not run)
function to call new() on class parSort
parSort( n = 1, stH = 0, Lxh = 0, Hkz = 0L, Skz = 0L, Hsh = 0, Zsh = 0, Lsh = 0, Zab = 14, Lab = 0, Az = 0, LIh = 0, trL = 0, fixN = 0L, fixL = 0, fixZ = 0, fixA = 0, fixR = 0, ... )parSort( n = 1, stH = 0, Lxh = 0, Hkz = 0L, Skz = 0L, Hsh = 0, Zsh = 0, Lsh = 0, Zab = 14, Lab = 0, Az = 0, LIh = 0, trL = 0, fixN = 0L, fixL = 0, fixZ = 0, fixA = 0, fixR = 0, ... )
n |
the number of parameter sets to generate, defaults to 1 |
stH |
stump height |
Lxh |
length of unusable wood at stem foot, see details |
Hkz |
height indicator, see details |
Skz |
stem indicator, see details |
Hsh |
height of stem wood, see details |
Zsh |
cutting diameter of stem wood, see details |
Lsh |
length of stem wood, see details |
Zab |
cutting diameter of upper trunk, see details |
Lab |
length of upper trunk, see details |
Az |
minimal cutting diameter, defaults to 7cm, see details |
LIh |
length of industrial wood, see details |
trL |
maximum transport length |
fixN |
number of fixed length assortments, see details |
fixL |
length of fixed length assortments, see details |
fixZ |
cutting diameter of fixed length assortments, see details |
fixA |
absolute add-on for good measure of fixed length assortments, given in cm; see details |
fixR |
relative add-on for good measure of fixed length assortments, given in percentage, i.e. 1% = 1; see details |
... |
currently unused |
if n is not given (or one) and any of the other parameter is given with length greater than one, n is reset to the maximum length of all parameters; care should be taken when using n and individual parameter setting for several trees.
an object of class parSort, i.e. a list, each element of
length n or maximum of length of defined parameters
This class represents one or multiple parameter sets holding the necessary information to specify the assortment process.
using indices i and j to subset
## S4 method for signature 'parSort,ANY,ANY,ANY' x[i, j, ..., drop = FALSE]## S4 method for signature 'parSort,ANY,ANY,ANY' x[i, j, ..., drop = FALSE]
x |
object from which to extract |
i |
index i |
j |
index j |
... |
not currently used |
drop |
drop dimensions, defaults to FALSE |
The assortment process is defined by several parameters. These follow the specification of its ancestor BDAT , but are extended to allow for fix length assortments at the tree top (industrial wood / pulp wood) and relaxes transport length and stump height.
stH: stump height, defaults to 0, i.e. 1% of tree height
Lxh: length of unusable wood at stem foot [m], defaults to 0 (X-Holz)
Hkz: indicator for tree top, 0 - normal, 1 - Wipfelbruch, 2 - Gipfelbruch
0 => H=H (default)
1 => H=H+2
2 => DBH < 30 => H=DBH; dbh > 30 => H = 30 + (DBH-30) * 0.3
Skz: indicator for stem type, defaults to 0
0 => conifer trees => no assortment restriction; deciduous trees => no assortments
1 => monopodial deciduous trees => Hsh = 0.7*H
2 => branching between dbh and 7m => Hsh = 5m
3 => crown base < 3m => Hsh=0.1
4 => dead or broken stem => Az = H*0.7
5 => dead tree => non-usable wood
Hsh: usable stem height, defaults to 0, i.e. 0.7*H
Zsh: minimum cutting diameter under bark for stem wood [cm], defaults
to 0, using parameter Az if estimated length < maximum length
(i.e. 20m)
Lsh: length of stem wood, defaults to 0, i.e. length unrestricted
Zab: minimum cutting diameter under bark for top segment [cm], defaults to 0, i.e. 14cm under bark
Lab: length of top segment, defaults to 0, i.e. length unrestricted
Az: minimum cutting diameter over bark [cm], defaults to 0, using an exponential function given DBH to estimate Az
LIh: length of industrial wood [m], defaults to 0, i.e. length unrestricted
trL: maximum transport length of assortments, defaults to 0, i.e. 19m
fixN: number of fixed length assortments at stem foot, defaults to 0 (no fixed length assortments, irrespective of other fix* parameters)
fixZ: mininum diameter under bark for fixed length assortment at stem foot, defaults to 0
fixL: length of fixed length assortment at stem foot, defaults to 0
fixA: fixed length assortement add-on in [cm], defaults to 0
fixR: fixed length assortement add-on in [%], defaults to 0
a part of the original object
x[i: subsetting for class 'parSort'
stHstump height
Lxhlength of unusable wood at stem foot, see details
Hkzheight indicator, see details
Skzstem indicator, see details
Hshheight of stem wood, see details
Zshcutting diameter of stem wood, see details
Lshlength of stem wood, see details
Zabcutting diameter of upper trunk, see details
Lablength of upper trunk, see details
Azminimal cutting diameter, defaults to 7cm, see details
LIhlength of industrial wood, see details
trLmaximum transport length
fixNnumber of fixed length assortments, see details
fixLlength of fixed length assortments, see details
fixZcutting diameter of fixed length assortments, see details
fixAabsolute add-on for good measure of fixed length assortments, see details
fixRrelative add-on for good measure of fixed length assortments, see details
parSort() parSort(Lxh=1) parSort(n=2)parSort() parSort(Lxh=1) parSort(n=2)
height estimation based on diameter in breast height and species using a Petterson-function
petterson(sp, d13)petterson(sp, d13)
sp |
vector of species code for biomass function from interval [1;18];
see |
d13 |
vector of diameter in breast height; in centimeter |
a scalar: tree height
tprTrees
creating a plot of the taper curve of a tree, over or under bark
## S3 method for class 'tprTrees' plot( x, bark = NULL, col.bark = NULL, obs = FALSE, assort = NULL, legend = FALSE, ... )## S3 method for class 'tprTrees' plot( x, bark = NULL, col.bark = NULL, obs = FALSE, assort = NULL, legend = FALSE, ... )
x |
an object of class 'tprTrees' |
bark |
either NULL or logical; if TRUE taper curve over bark is plotted, if FALSE taper curve under bark is plotted; if NULL, both are plotted |
col.bark |
color to be used for plot of bark, if plot of taper curve over and under bark is requested |
obs |
should observations (measured/observed diameters) be added to the plot? |
assort |
assortments produced by |
legend |
logical, if legend should be added |
... |
further arguments for |
plots the taper curve of a tree. Either over bark or under bark, or
both. Elements design can partly be chosen. If assortments are given, these
are added to the plot. Doing that, the assortment bottom and top position is
indicated by a vertical line and mid-diameter is shown as a point with
vertical dashed line. N.B. the mid-diameter shown is under bark and rounded
downwards for 0.5 cm if mid-diameter < 20 and for 0.75 cm if bigger. Volume
is calculated using this diameter. Reason for that behaviour is that
assortment information with regard to diameter and volume reflects the legal
rules for roundwood assortments (german RVR).
Additionally, assortment names are indicated.
One can provide assortment names in a column of assort named
'assortname', which will be used if available, otherwise the 'Sort'-column
will be used. See Examples.
No return value, called for side effects
## plotting the taper curve of a tree oldpar <- par() par(mfrow = c(1, 1)) tree <- tprTrees(spp=1L, Dm=40, Hm=1.3, H=35) plot(tree, type = "l", las = 1, legend = TRUE) plot(tree, bark = TRUE, las = 1) plot(tree, bark = FALSE, las = 1, obs=TRUE) # obs incl. bark!!! tree <- tprTrees(spp=c(1, 1), Dm = c(40, 35), Hm=c(1.3, 1.3), H = c(35, 30)) plot(tree, bark = FALSE, las = 1, legend = TRUE) # both trees are plotted plot(tree, bark = TRUE, las = 1, legend = TRUE, obs=TRUE) tree <- tprTrees(spp=1L, Dm=c(40, 32), Hm=c(1.3, 10.5), H=35) plot(tree, type = "l", las = 1, legend = TRUE, obs=TRUE) ## if monotonicity is not forced: tree <- tprTrees(spp=3L, Dm=8, Hm=1.3, H=10) plot(tree, type = "l", las = 1, obs=TRUE, mono=FALSE) plot(tree, type = "l", las = 1, obs=TRUE, mono=TRUE) # default tree <- tprTrees(spp=c(1, 8), Dm = c(40, 40), Hm=c(1.3, 1.3), H = c(35, 35)) plot(tree, bark = NULL, las = 1, col.bark = "blue", legend = TRUE) plot(tree, bark = NULL, las = 1, col.bark = "blue", legend = TRUE, obs = TRUE) plot(tree[1, ], main = tprSpeciesCode(spp(tree[1, ]), out = "long")) plot(tree[2, ], main = tprSpeciesCode(spp(tree[2, ]), out = "scientific")) par(mfrow = c(1, 2)) plot(tree, bark = TRUE, las = 1) ## now add assortments into taper curve par(mfrow = c(1, 1)) pars <- parSort(n=length(tree), Lxh=1, fixN=2, fixL=4, fixA=10) ass <- tprAssortment(tree, pars=pars) plot(tree, assort = ass) plot(tree, bark = FALSE, assort = ass) plot(tree, bark = FALSE, assort = ass, legend = TRUE) plot(tree[1, ], assort = ass[ass$tree == 1, ], main = "first tree in subset") plot(tree[2, ], assort = ass[ass$tree == 2, ], main = "second tree in subset") ## adding own assortment labels using column 'assortname' ass$assortname <- ifelse(grepl("fix", ass$sort), paste0("Fix:", ass$length), ass$sort) plot(tree, assort = ass) par(oldpar)## plotting the taper curve of a tree oldpar <- par() par(mfrow = c(1, 1)) tree <- tprTrees(spp=1L, Dm=40, Hm=1.3, H=35) plot(tree, type = "l", las = 1, legend = TRUE) plot(tree, bark = TRUE, las = 1) plot(tree, bark = FALSE, las = 1, obs=TRUE) # obs incl. bark!!! tree <- tprTrees(spp=c(1, 1), Dm = c(40, 35), Hm=c(1.3, 1.3), H = c(35, 30)) plot(tree, bark = FALSE, las = 1, legend = TRUE) # both trees are plotted plot(tree, bark = TRUE, las = 1, legend = TRUE, obs=TRUE) tree <- tprTrees(spp=1L, Dm=c(40, 32), Hm=c(1.3, 10.5), H=35) plot(tree, type = "l", las = 1, legend = TRUE, obs=TRUE) ## if monotonicity is not forced: tree <- tprTrees(spp=3L, Dm=8, Hm=1.3, H=10) plot(tree, type = "l", las = 1, obs=TRUE, mono=FALSE) plot(tree, type = "l", las = 1, obs=TRUE, mono=TRUE) # default tree <- tprTrees(spp=c(1, 8), Dm = c(40, 40), Hm=c(1.3, 1.3), H = c(35, 35)) plot(tree, bark = NULL, las = 1, col.bark = "blue", legend = TRUE) plot(tree, bark = NULL, las = 1, col.bark = "blue", legend = TRUE, obs = TRUE) plot(tree[1, ], main = tprSpeciesCode(spp(tree[1, ]), out = "long")) plot(tree[2, ], main = tprSpeciesCode(spp(tree[2, ]), out = "scientific")) par(mfrow = c(1, 2)) plot(tree, bark = TRUE, las = 1) ## now add assortments into taper curve par(mfrow = c(1, 1)) pars <- parSort(n=length(tree), Lxh=1, fixN=2, fixL=4, fixA=10) ass <- tprAssortment(tree, pars=pars) plot(tree, assort = ass) plot(tree, bark = FALSE, assort = ass) plot(tree, bark = FALSE, assort = ass, legend = TRUE) plot(tree[1, ], assort = ass[ass$tree == 1, ], main = "first tree in subset") plot(tree[2, ], assort = ass[ass$tree == 2, ], main = "second tree in subset") ## adding own assortment labels using column 'assortname' ass$assortname <- ifelse(grepl("fix", ass$sort), paste0("Fix:", ass$length), ass$sort) plot(tree, assort = ass) par(oldpar)
extract parameter of bark functions according to Altherr et al. 1974 - 1979
RiPar(ba = NULL, fn = NULL, par = NULL)RiPar(ba = NULL, fn = NULL, par = NULL)
ba |
tree species code; returned by |
fn |
function number; see details |
par |
parameter; see details |
Function extracts the parameter according to tree species, function type and parameter number. There are three parameters in each of four functions. The first one refers to butt log (dt. Erdstamm), the second to middle log (dt. Mittelstammstück), the third to the top log (dt. Gipfelstammstück) and the fourth to the complete stem (dt. Gesamtstamm).
Function to set and get options on how the TapeS-package works.
setTapeSoptions(Rfn = list(fn = "sig2"), mono = TRUE) getTapeSoptions(name = NULL)setTapeSoptions(Rfn = list(fn = "sig2"), mono = TRUE) getTapeSoptions(name = NULL)
Rfn |
setting for residuals error matrix, defaults to |
mono |
logical, defaults to true. If calibrated taper curve is non-monotonic at stem base, an support diameter is added. |
name |
name of options to be returned |
So far, only two options are implemented: TapeS_Rfn and
TapeS_mono. Teh first defaults to "sig2" (i.e. 'sigma squared') and
the second to "TRUE".
The TapeR-taper curves can be evaluated in basically two ways: (i) either
as defined in the TapeR-package, i.e. the diameters and volumes are
estimated using the estimated error structure and find an optimal taper
curve given the measured diameters or (ii) by interpolating the measured
diameters, i.e. forcing the estimated taper curve through those
measurements by setting the residual error structure to zero. See Kublin et
al. (2013), p.987 bottom left. Technically, forcing the taper curve through
the measurements is achieved by setting the residual error matrix R to
zero, that is Rfn = list(fn="zero"). Defaults to Rfn =
list(fn="sig2"). Besides, one can defined other functions about
assumptions about the errors at the measurement positions, see
resVar for options.
NB: Caution is required in applying Rfn=list(fn="zero"), since
forcing the taper curve through too many points might lead to singularities
or implausible results!
The option 'mono=TRUE' assures that no taper curve is generated which shows lower diameter in lower heights, possibly adding a support diameter at 1% of tree height.
by defaults, sets options()$TapeS_Rfn to "sig2"
Kublin, E., Breidenbach, J., Kaendler, G. (2013) A flexible stem taper and volume prediction method based on mixed-effects B-spline regression, Eur J For Res, 132:983-997.
## reset option TapeS_Rfn to "sig2", i.e. model based errors by setTapeSoptions(Rfn = list(fn="sig2")) ## or to force the taper curve through the measurements, set options("TapeS_Rfn" = list(fn="zero")) ## see the actual state of options by options()[grep("^TapeS_", names(options()))] ## or easier getTapeSoptions()## reset option TapeS_Rfn to "sig2", i.e. model based errors by setTapeSoptions(Rfn = list(fn="sig2")) ## or to force the taper curve through the measurements, set options("TapeS_Rfn" = list(fn="zero")) ## see the actual state of options by options()[grep("^TapeS_", names(options()))] ## or easier getTapeSoptions()
tprTrees
Function to simulate an object of class tprTrees
simTrees(par = NULL)simTrees(par = NULL)
par |
list of lists, one for each species |
Function simulates trees based on given distributions and petterson height function. Dbh can be simulated using normal ('norm'), weibull or gamma distribution. Others might be added.
The par-list of each species needs the following named entries:
spp - species code, n - number of trees, ddist -
distribution of dbh, dpar - list of parameter of the distribution,
i.e. mu and sd for normal distribution and shape and
scale for weibull and gamma distribution. The latter both might use
lag to offset the estimated diameter by this amount.
an object of class tprTrees
petterson for the implemented height function and
dnorm, dweibull and dgamma for the
diameter distributions.
par <- list(list(spp=1, n=10, ddist="norm", dpar=list(mu=30, sd=4)), list(spp=3, n=5, ddist="norm", dpar=list(mu=40, sd=2))) simTrees(par)par <- list(list(spp=1, n=10, ddist="norm", dpar=list(mu=30, sd=4)), list(spp=3, n=5, ddist="norm", dpar=list(mu=40, sd=2))) simTrees(par)
get and set slot values
spp(obj) ## S4 method for signature 'tprTrees' spp(obj) spp(obj) <- value ## S4 replacement method for signature 'tprTrees' spp(obj) <- value Dm(obj) ## S4 method for signature 'tprTrees' Dm(obj) Dm(obj) <- value ## S4 replacement method for signature 'tprTrees' Dm(obj) <- value Hm(obj) ## S4 method for signature 'tprTrees' Hm(obj) Hm(obj) <- value ## S4 replacement method for signature 'tprTrees' Hm(obj) <- value Ht(obj) ## S4 method for signature 'tprTrees' Ht(obj) Ht(obj) <- value ## S4 replacement method for signature 'tprTrees' Ht(obj) <- value sHt(obj) ## S4 method for signature 'tprTrees' sHt(obj) sHt(obj) <- value ## S4 replacement method for signature 'tprTrees' sHt(obj) <- value mono(obj) ## S4 method for signature 'tprTrees' mono(obj)spp(obj) ## S4 method for signature 'tprTrees' spp(obj) spp(obj) <- value ## S4 replacement method for signature 'tprTrees' spp(obj) <- value Dm(obj) ## S4 method for signature 'tprTrees' Dm(obj) Dm(obj) <- value ## S4 replacement method for signature 'tprTrees' Dm(obj) <- value Hm(obj) ## S4 method for signature 'tprTrees' Hm(obj) Hm(obj) <- value ## S4 replacement method for signature 'tprTrees' Hm(obj) <- value Ht(obj) ## S4 method for signature 'tprTrees' Ht(obj) Ht(obj) <- value ## S4 replacement method for signature 'tprTrees' Ht(obj) <- value sHt(obj) ## S4 method for signature 'tprTrees' sHt(obj) sHt(obj) <- value ## S4 replacement method for signature 'tprTrees' sHt(obj) <- value mono(obj) ## S4 method for signature 'tprTrees' mono(obj)
obj |
object of class 'tprtrees' |
value |
depending on slot, see details |
Getting and setting the values of the different slots of
'tprTrees'-objects. For slot mono no setting function has been defined,
as this slot is computed by check_monotonicity and should not
be reset by users.
Setting of spp requires mode integer. For convenience,
value is coerced by as.integer.
Setting spp and H, a vector of length equal
length(spp(obj)) is required.
For setting slots Dm and Hm value must be a
list of vectors of length equal length(spp(obj)) and the length of
each vector must correspond to the length of the vectors in Hm and
Dm.
the accessor functions return the value of the specified slot and the setting functions update the object
spp(): getting slot 'spp' of obj
spp(tprTrees): method for class 'tprTrees'
spp(obj) <- value: setting 'spp' slot of object
spp(tprTrees) <- value: method for class 'tprTrees'
Dm(): getting slot 'Dm' of obj
Dm(tprTrees): method for class 'tprTrees'
Dm(obj) <- value: setting 'Dm' slot of object
Dm(tprTrees) <- value: method for class 'tprTrees'
Hm(): getting slot 'Hm' of obj
Hm(tprTrees): method for class 'tprTrees'
Hm(obj) <- value: setting 'Hm' slot of object
Hm(tprTrees) <- value: method for class 'tprTrees'
Ht(): getting slot 'Ht' of obj
Ht(tprTrees): method for class 'tprTrees'
Ht(obj) <- value: setting 'Ht' slot of object
Ht(tprTrees) <- value: method for class 'tprTrees'
sHt(): getting slot 'sHt' of obj
sHt(tprTrees): method for class 'tprTrees'
sHt(obj) <- value: setting 'sHt' slot of object
sHt(tprTrees) <- value: method for class 'tprTrees'
mono(): getting slot 'monotone' of obj
mono(tprTrees): method for class 'tprTrees'
Function calculates assortments for given tree according to assortment specification
tprAssortment(obj, pars = NULL, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprAssortment(obj, pars = NULL, mono = TRUE, Rfn = NULL)tprAssortment(obj, pars = NULL, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprAssortment(obj, pars = NULL, mono = TRUE, Rfn = NULL)
obj |
an object of class 'tprTrees' |
pars |
parameters to specify assortments, see |
mono |
logical, defaults to true. If calibrated taper curve is non-monotonic at stem base, a support diameter is added. |
Rfn |
Rfn setting for residuals error matrix, defaults to
|
a data.frame with columns tree: tree identifier,
sort: assortment name, height: beginning of assortment along
trunk, length: length of assortment, mdm: mid-diameter of
assortment, zdm: top-diameter of assortment and vol: volume.
tprAssortment(tprTrees): method for class 'tprTrees'
## conifer wood obj <- tprTrees(spp=c(1, 8), Dm=list(30, 40), Hm=list(1.3, 1.3), Ht=c(30, 40)) tprAssortment(obj) pars <- parSort(stH=0.2, Lxh=c(1, 1.5), fixN=2, fixL=4) (ass <- tprAssortment(obj, pars)) plot(obj, assort = ass) ## deciduous wood obj <- tprTrees(spp=c(15), Dm=list(40), Hm=list(1.3), Ht=c(40)) tprAssortment(obj) pars <- parSort(n=length(obj), Lxh=c(1), Hsh=10, Az=10) ass <- tprAssortment(obj, pars) plot(obj, assort=ass)## conifer wood obj <- tprTrees(spp=c(1, 8), Dm=list(30, 40), Hm=list(1.3, 1.3), Ht=c(30, 40)) tprAssortment(obj) pars <- parSort(stH=0.2, Lxh=c(1, 1.5), fixN=2, fixL=4) (ass <- tprAssortment(obj, pars)) plot(obj, assort = ass) ## deciduous wood obj <- tprTrees(spp=c(15), Dm=list(40), Hm=list(1.3), Ht=c(40)) tprAssortment(obj) pars <- parSort(n=length(obj), Lxh=c(1), Hsh=10, Az=10) ass <- tprAssortment(obj, pars) plot(obj, assort=ass)
Hx
Funktion evaluates the double bark thickness models developed by Altherr et al (1974-79).
tprBark(obj, Hx, cp = TRUE, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprBark(obj, Hx, cp = TRUE, mono = TRUE, Rfn = NULL)tprBark(obj, Hx, cp = TRUE, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprBark(obj, Hx, cp = TRUE, mono = TRUE, Rfn = NULL)
obj |
object of class 'tprTrees' |
Hx |
height for which double bark thickness is required |
cp |
cartesian product, i.e. apply all |
mono |
logical, defaults to true. If calibrated taper curve is non-monotonic at stem base, an support diameter is added. |
Rfn |
Rfn setting for residuals error matrix, defaults to
|
double bark thickness [cm]
tprBark(tprTrees): method for class 'tprTrees'
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1974). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Kiefer, Buche, Hainbuche, Esche und Roterle. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1975). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Europäische Lärche, Japanische Lärche, Schwarzkiefer, Stieleiche, Traubeneiche, Roteiche, Bergahorn und Linde. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1976). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Weymouthskiefer, Robinie, Bergulme, Birke, Marilandica-Pappel und Robusta-Pappel. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1978). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Fichte, Tanne, Douglasie und Sitka-Fichte. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
Altherr, E., P. Unfried, J. Hradetzky and V. Hradetzky (1979). Statistische Rindenbeziehungen als Hilfsmittel zur Ausformung und Aufmessung unentrindeten Stammholzes. Neupotz-Pappel, Regenerata-Pappel, Kirsche, Spitzahorn, Feldahorn, Aspe, Weide, Flatterulme, Tulpenbaum u. Elsbeere. Freiburg i. Br., Forstl. Versuchs- u. Forschungsanst. Baden-Württenberg.
## calculating bark thickness depends on diameter estimation and hence on the ## assumed residual variance at calibration. ## can be Rfn=list(fn="sig2") (default), i.e. EBLUP estimation from taper curve ## or e.g. Rfn=list(fn="zero"), i.e. force taper curve through the given measurements options("TapeS_Rfn") # "sig2", default in TapeS tmp <- tprTrees() Dm(tmp); Hm(tmp) # Dbh = D(Hx=1.3) = 30cm (measured) Dbh(tmp) # estimated via EBLUP from taper curve tprBark(tmp, Hx = c(1.3, 5)) # bark thickness corresponds to Dbh(tmp) (d <- tprDiameter(tmp, Hx = c(1.3, 5), bark=TRUE)) ## predicted bark(1, d[1], 1.3/30) # the same! bark(1, d[2], 5/30) # the same! ## if using option TapeS_Rfn = list(fn="zero"), force taper curve through measurements setTapeSoptions(Rfn = list(fn="zero")) options()$TapeS_Rfn tprBark(tmp, Hx = c(1.3, 5)) bark(1, 30, 1.3/30) # the same but different to above bark(1, d[1], 1.3/30) # cf. above bark(1, 28, 5/30) # the same but different to above bark(1, d[2], 1.3/30) # cf. above## calculating bark thickness depends on diameter estimation and hence on the ## assumed residual variance at calibration. ## can be Rfn=list(fn="sig2") (default), i.e. EBLUP estimation from taper curve ## or e.g. Rfn=list(fn="zero"), i.e. force taper curve through the given measurements options("TapeS_Rfn") # "sig2", default in TapeS tmp <- tprTrees() Dm(tmp); Hm(tmp) # Dbh = D(Hx=1.3) = 30cm (measured) Dbh(tmp) # estimated via EBLUP from taper curve tprBark(tmp, Hx = c(1.3, 5)) # bark thickness corresponds to Dbh(tmp) (d <- tprDiameter(tmp, Hx = c(1.3, 5), bark=TRUE)) ## predicted bark(1, d[1], 1.3/30) # the same! bark(1, d[2], 5/30) # the same! ## if using option TapeS_Rfn = list(fn="zero"), force taper curve through measurements setTapeSoptions(Rfn = list(fn="zero")) options()$TapeS_Rfn tprBark(tmp, Hx = c(1.3, 5)) bark(1, 30, 1.3/30) # the same but different to above bark(1, d[1], 1.3/30) # cf. above bark(1, 28, 5/30) # the same but different to above bark(1, d[2], 1.3/30) # cf. above
calculate total above ground and optionally component biomass for given trees
tprBiomass( obj, component = NULL, useNFI = TRUE, interval = "none", mono = TRUE, Rfn = NULL ) ## S4 method for signature 'tprTrees' tprBiomass( obj, component = NULL, useNFI = TRUE, interval = "none", mono = TRUE, Rfn = NULL )tprBiomass( obj, component = NULL, useNFI = TRUE, interval = "none", mono = TRUE, Rfn = NULL ) ## S4 method for signature 'tprTrees' tprBiomass( obj, component = NULL, useNFI = TRUE, interval = "none", mono = TRUE, Rfn = NULL )
obj |
object of class 'tprTrees' |
component |
component for which biomass should be returned. If NULL, total aboveground biomass is returned, if 'all', all components are returned. See details. |
useNFI |
if |
interval |
character to indicate whether and which type of interval is
required; one of |
mono |
logical, defaults to true. If calibrated taper curve is non-monotonic at stem base, a support diameter is added. |
Rfn |
Rfn setting for residuals error matrix, defaults to
|
The available components are agb (= total aboveground biomass), stw (=stump wood), stb (=stump bark), sw (=solid wood with diameter above 7cm over bark), sb (=bark of component sw), fwb (=fine wood incl. bark) and ndl (=needles), if applicable. The needles-component is set to zero for deciduous tree species, no mass for leaves is available. One can request 'all' components to receive all components.
a vector in case agb or only one component is requested, otherwise a matrix with one row per tree
tprBiomass(tprTrees): method for class 'tprTrees'
Kändler, G. and B. Bösch (2012). Methodenentwicklung für die 3. Bundeswaldinventur: Modul 3 Überprüfung und Neukonzeption einer Biomassefunktion - Abschlussbericht. Im Auftrag des Bundesministeriums für Ernährung, Landwirtschaft und Verbraucherschutz in Zusammenarbeit mit dem Institut für Waldökologie und Waldinventur des Johann Heinrich von Thünen-Instituts, FVA-BW: 71.
Kaendler (2021): Biometrische Modelle für die Ermittlung des Holzvorrats, seiner Sortimentsstruktur und der oberirdischen Biomasse im Rahmen der Bundeswaldinventur. Allg. Forst- u. J.-Ztg., 191. Jg., 5/6 83
Vonderach, C., G. Kändler and C. Dormann (2018): Consistent set of additive biomass equations for eight tree species in Germany fitted by nonlinear seemingly unrelated regression. Annals of Forest Science (2018) 75:49 doi: 10.1007/s13595-018-0728-4
obj <- tprTrees(spp=c(1, 15), Dm=list(c(30, 28), c(30, 28)), Hm=list(c(1, 3), c(1, 3)), Ht = rep(30, 2)) (tmp <- tprBiomass(obj, component="all")) tprBiomass(obj, component=NULL) # aboveground biomass component <- c("agb", "sw", "sb", "ndl") tprBiomass(obj, component=component) component <- c("sw", "sb", "ndl") tprBiomass(obj, component="all") # use NSUR-functions from Vonderach et al. 2018 # obs: currently sth=1% of tree height # and kl=70% of tree height tprBiomass(obj, component="all", useNFI = FALSE) ## getting confidence and prediction intervals useNFI <- FALSE interval <- "confidence" component <- c("sw", "agb") mono <- TRUE Rfn <- NULL tprBiomass(obj, component, useNFI, interval) tprBiomass(obj, component, useNFI, interval="none") tprBiomass(obj, component, useNFI=TRUE, interval) tprBiomass(obj, component, useNFI=TRUE, interval="none")obj <- tprTrees(spp=c(1, 15), Dm=list(c(30, 28), c(30, 28)), Hm=list(c(1, 3), c(1, 3)), Ht = rep(30, 2)) (tmp <- tprBiomass(obj, component="all")) tprBiomass(obj, component=NULL) # aboveground biomass component <- c("agb", "sw", "sb", "ndl") tprBiomass(obj, component=component) component <- c("sw", "sb", "ndl") tprBiomass(obj, component="all") # use NSUR-functions from Vonderach et al. 2018 # obs: currently sth=1% of tree height # and kl=70% of tree height tprBiomass(obj, component="all", useNFI = FALSE) ## getting confidence and prediction intervals useNFI <- FALSE interval <- "confidence" component <- c("sw", "agb") mono <- TRUE Rfn <- NULL tprBiomass(obj, component, useNFI, interval) tprBiomass(obj, component, useNFI, interval="none") tprBiomass(obj, component, useNFI=TRUE, interval) tprBiomass(obj, component, useNFI=TRUE, interval="none")
Function evaluates TapeR taper curve models for given trees according to species, required height and optionally substracts double bark thickness.
tprDiameter( obj, Hx, bark = TRUE, interval = "none", cp = TRUE, mono = TRUE, Rfn = NULL ) ## S4 method for signature 'tprTrees' tprDiameter( obj, Hx, bark = TRUE, interval = "none", cp = TRUE, mono = TRUE, Rfn = NULL )tprDiameter( obj, Hx, bark = TRUE, interval = "none", cp = TRUE, mono = TRUE, Rfn = NULL ) ## S4 method for signature 'tprTrees' tprDiameter( obj, Hx, bark = TRUE, interval = "none", cp = TRUE, mono = TRUE, Rfn = NULL )
obj |
object of class 'tprTrees' |
Hx |
vector of heights for which diameter w/ or w/0 bark are required |
bark |
should diameter over or under bark be returned? |
interval |
indicator about whether 'confidence' or 'prediction' intervals are required (defaults to 'none'), optionally function returns the mean squared error of the mean and predictions ('MSE'). |
cp |
cartesian product, i.e. apply all |
mono |
logical, defaults to true. If calibrated taper curve is non-monotonic at stem base, a support diameter is added. |
Rfn |
Rfn setting for residuals error matrix, defaults to
|
Function evaluates taper curves at required height Hx. By
default (cp==TRUE), the taper curve is evaluated at Hx for each
tree. If cp==FALSE, each tree is evaluated at exactly one Hx (recycled
if necessary). This feature is intended for situations where diameter in
relative heights are required. Then, the recycling of one height Hx (e.g.
1.3m) is not possible, since relative heights depend on absolute tree height,
which might be different for each tree. Hence a call like
tprDiameter(obj, Hx=0.3*Ht(obj), cp=FALSE) is necessary.
a matrix or data.frame depending on value of interval. If
'none' (the default), a matrix of size [length(obj@Ht), length(Hx)] is
returned, otherwise a data.frame of size [length(obj@Ht) * length(Hx), 5].
The five columns hold a tree identifier, Hx, lower confidence/prediction
interval, the estimated diameter and the upper confidence/prediction
interval. In case 'interval=MSE' the returned columns contain a tree
identifier, Hx, the estimated diameter and mean squared error (MSE) of the
mean and of the prediction. Estimates and intervals include bark or not,
depending on bark.
tprDiameter(tprTrees): method for class 'tprTrees'
tprDiameterCpp for a faster implementation if no
confidence or prediction information are required and tprBark
for the applied bark reduction.
## prediction for new tree using implemented 'TapeR' taper curve model obj <- tprTrees(spp=c(1, 3), Hm=list(c(1.3, 5), c(1.3, 5)), Dm=list(c(27, 25), c(27, 25)), Ht=c(27, 27)) hx <- c(1.3, 5, 7) ## by default, Hx applied on each tree, i.e. result is a 2x3 matrix tprDiameter(obj, Hx = hx) ## if cp=FALSE, each tree only 'sees' one Hx, i.e. results is a vector ## (obs: length of Hx must be identical to length of obj) tprDiameter(obj, Hx = c(1.3, 5), cp=FALSE) tprDiameter(obj, Hx = hx, bark = FALSE) tprDiameter(obj, Hx = hx, interval = "confidence") tprDiameter(obj, Hx = hx, bark = FALSE, interval = "prediction") tprDiameter(obj, Hx = hx, interval = "MSE") tprDiameter(obj, Hx = hx, bark=FALSE, interval = "MSE") ## here same behaviour, if cp=FALSE tprDiameter(obj, Hx = c(1.3, 5), bark = FALSE, interval = "prediction", cp=FALSE) ## using Cpp-implementation ## faster, but no intervals available tprDiameterCpp(obj, Hx = hx) tprDiameterCpp(obj, Hx = c(1.3, 5), cp=FALSE) ## prediction for objects of class 'datBDAT': if(require(rBDAT)){ tree <- rBDAT::buildTree(list(spp=1, D1=20:30, H1=1.3, H2=50, H=20:30)) tree <- bdat_as_tprtrees(tree) tprDiameter(tree, Hx = 1.3) }## prediction for new tree using implemented 'TapeR' taper curve model obj <- tprTrees(spp=c(1, 3), Hm=list(c(1.3, 5), c(1.3, 5)), Dm=list(c(27, 25), c(27, 25)), Ht=c(27, 27)) hx <- c(1.3, 5, 7) ## by default, Hx applied on each tree, i.e. result is a 2x3 matrix tprDiameter(obj, Hx = hx) ## if cp=FALSE, each tree only 'sees' one Hx, i.e. results is a vector ## (obs: length of Hx must be identical to length of obj) tprDiameter(obj, Hx = c(1.3, 5), cp=FALSE) tprDiameter(obj, Hx = hx, bark = FALSE) tprDiameter(obj, Hx = hx, interval = "confidence") tprDiameter(obj, Hx = hx, bark = FALSE, interval = "prediction") tprDiameter(obj, Hx = hx, interval = "MSE") tprDiameter(obj, Hx = hx, bark=FALSE, interval = "MSE") ## here same behaviour, if cp=FALSE tprDiameter(obj, Hx = c(1.3, 5), bark = FALSE, interval = "prediction", cp=FALSE) ## using Cpp-implementation ## faster, but no intervals available tprDiameterCpp(obj, Hx = hx) tprDiameterCpp(obj, Hx = c(1.3, 5), cp=FALSE) ## prediction for objects of class 'datBDAT': if(require(rBDAT)){ tree <- rBDAT::buildTree(list(spp=1, D1=20:30, H1=1.3, H2=50, H=20:30)) tree <- bdat_as_tprtrees(tree) tprDiameter(tree, Hx = 1.3) }
This function uses Rcpp and C-code to implement the diameter estimation of package TapeR to allow for faster estimation if no interval information is required.
tprDiameterCpp(obj, Hx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprDiameterCpp(obj, Hx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL)tprDiameterCpp(obj, Hx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprDiameterCpp(obj, Hx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL)
obj |
object of class 'tprTrees' |
Hx |
vector of heights for which diameter are required |
bark |
should diameter over or under bark be returned? |
cp |
cartesian product, i.e. apply all |
mono |
logical to decide whether a supporting diameter should be added in case the taper curve is regarded as non-monotonic. Defaults to TRUE. |
Rfn |
setting for residuals error matrix, defaults to |
Function evaluates taper curves at required height Hx. By
default (cp==TRUE), the taper curve is evaluated at Hx for each
tree. If cp==FALSE, each tree is evaluated at exactly one Hx (recycled
if necessary). This feature is intended for situations where diameter in
relative heights are required. Then, the recycling of one height Hx (e.g.
1.3m) is not possible, since relative heights depend on absolute tree height,
which might be different for each tree. Hence a call like
tprDiameter(obj, Hx=0.3*Ht(obj), cp=FALSE) is necessary.
a vector, in case only one diameter (i.e. Hx) is required per tree
(cp=FALSE) or a matrix of size
length(trees) x length(Hx) (cp=TRUE).
tprDiameterCpp(tprTrees): method for class 'tprTrees'
tprDiameter if confidence or prediction intervals
are required.
obj <- tprTrees(spp=c(1 , 3), Hm=list(c(1.3, 5), c(1.3, 5)), Dm=list(c(27, 25), c(27, 25)), Ht=c(27, 27)) Hx <- seq(0, 1, 0.1) tprDiameterCpp(obj, Hx = Hx) tprDiameterCpp(obj, Hx = Hx, bark=FALSE) tprDiameterCpp(obj, Hx = c(1, 2), bark=FALSE, cp=FALSE) require(rbenchmark) benchmark(tprDiameter(obj, Hx, bark = TRUE), tprDiameterCpp(obj, Hx, bark = TRUE), replications = 10000)[,1:4]obj <- tprTrees(spp=c(1 , 3), Hm=list(c(1.3, 5), c(1.3, 5)), Dm=list(c(27, 25), c(27, 25)), Ht=c(27, 27)) Hx <- seq(0, 1, 0.1) tprDiameterCpp(obj, Hx = Hx) tprDiameterCpp(obj, Hx = Hx, bark=FALSE) tprDiameterCpp(obj, Hx = c(1, 2), bark=FALSE, cp=FALSE) require(rbenchmark) benchmark(tprDiameter(obj, Hx, bark = TRUE), tprDiameterCpp(obj, Hx, bark = TRUE), replications = 10000)[,1:4]
Function to extract the height of given diameter w/ or w/o bark from taper curve
tprHeight(obj, Dx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprHeight(obj, Dx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL)tprHeight(obj, Dx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL) ## S4 method for signature 'tprTrees' tprHeight(obj, Dx, bark = TRUE, cp = TRUE, mono = TRUE, Rfn = NULL)
obj |
object of class 'tprTrees' |
Dx |
diameter for which height is required |
bark |
should given diameter be considered over or under bark? |
cp |
cartesian product, i.e. apply all |
mono |
logical, defaults to true. If calibrated taper curve is non-monotonic at stem base, a support diameter is added. |
Rfn |
Rfn setting for residuals error matrix, defaults to
|
estimated height of given diameter
tprHeight(tprTrees): method for class 'tprTrees'
obj <- tprTrees(spp=c(1, 3, 8, 15), Dm=list(c(30, 28), c(30, 28), c(30, 28), c(30, 28)), Hm=list(c(1.3, 5), c(1.3, 5), c(1.3, 5), c(1.3, 5)), Ht = rep(30, 4)) tprHeight(obj, Dx = c(30, 7), bark=TRUE) tprHeight(obj, Dx = c(30, 7), bark=FALSE) ## no cartesion product between obj and Dx, i.e. cp=FALSE ## Dx is recycled if necessary tprHeight(obj, Dx = c(30, 7), bark=FALSE, cp=FALSE)obj <- tprTrees(spp=c(1, 3, 8, 15), Dm=list(c(30, 28), c(30, 28), c(30, 28), c(30, 28)), Hm=list(c(1.3, 5), c(1.3, 5), c(1.3, 5), c(1.3, 5)), Ht = rep(30, 4)) tprHeight(obj, Dx = c(30, 7), bark=TRUE) tprHeight(obj, Dx = c(30, 7), bark=FALSE) ## no cartesion product between obj and Dx, i.e. cp=FALSE ## Dx is recycled if necessary tprHeight(obj, Dx = c(30, 7), bark=FALSE, cp=FALSE)
Function to get BDAT species code, or transform it to a german or english name, possibly an abbreviated version or even a scientific name
tprSpeciesCode(inSp = NULL, outSp = NULL)tprSpeciesCode(inSp = NULL, outSp = NULL)
inSp |
species information given, either numeric or character |
outSp |
character vector of names, for which information should be returned |
The function matches inSp to outSp. Depending on inSp, being either a numeric vector of values between 1 and 36 or a character vector of species names. Possible names are those which could be return values. One can get all names and the respective species code by calling the function with inSP=NULL and outSP=NULL (the default).
English species names and codes are taken from https://www.forestry.gov.uk/pdf/PF2011_Tree_Species.pdf/$FILE/PF2011_Tree_Species.pdf while slightly adjusting the codes to be unique compared to the german codes (e.g. European larch is now ELA instead of EL).
Any given species code outside the interval [1, 36] is given the code 1 (i.e. Norway spruce), while throwing a warning. If any inSp - name is invalid, i.e. not in species list, this throws an error.
All elements of outSp, which are not colnames of the default returned data.frame, are silently dropped.
vector or data.frame, depending on length of 'outSp'.
tprSpeciesCode(inSp=NULL, outSp=NULL) ## the default tprSpeciesCode() ## the same tprSpeciesCode(outSp = "scientific") tprSpeciesCode(inSp = c(1, 2)) ## giving codes tprSpeciesCode(inSp = c(1, 2, -1, 37)) ## values outside [1, 36] are given code 1 tprSpeciesCode(inSp = c(1, 2), outSp = c("scientific")) ## output a vector tprSpeciesCode(inSp = c("Bu", "Fi")) ## asking for codes of abbreviated german names tprSpeciesCode(inSp = c("Bu", "Fi", "Bu")) ## order is preserved tprSpeciesCode(inSp = c("Buche", "Fichte")) ## asking for codes of german names tprSpeciesCode(inSp = c("BE", "NS")) ## ... abbreviated english names tprSpeciesCode(inSp = c("beech", "Norway spruce")) ## ... english names tprSpeciesCode(inSp = c("Fagus sylvatica", "Picea abies")) ### ... scientific names ## not run ## tprSpeciesCode(inSp = c("Fagus sylvatica", "Picea")) ## error, 2nd name wrong ## end not runtprSpeciesCode(inSp=NULL, outSp=NULL) ## the default tprSpeciesCode() ## the same tprSpeciesCode(outSp = "scientific") tprSpeciesCode(inSp = c(1, 2)) ## giving codes tprSpeciesCode(inSp = c(1, 2, -1, 37)) ## values outside [1, 36] are given code 1 tprSpeciesCode(inSp = c(1, 2), outSp = c("scientific")) ## output a vector tprSpeciesCode(inSp = c("Bu", "Fi")) ## asking for codes of abbreviated german names tprSpeciesCode(inSp = c("Bu", "Fi", "Bu")) ## order is preserved tprSpeciesCode(inSp = c("Buche", "Fichte")) ## asking for codes of german names tprSpeciesCode(inSp = c("BE", "NS")) ## ... abbreviated english names tprSpeciesCode(inSp = c("beech", "Norway spruce")) ## ... english names tprSpeciesCode(inSp = c("Fagus sylvatica", "Picea abies")) ### ... scientific names ## not run ## tprSpeciesCode(inSp = c("Fagus sylvatica", "Picea")) ## error, 2nd name wrong ## end not run
constructor for class tprTrees
tprTrees( spp = 1L, Dm = list(c(30, 28)), Hm = list(c(1.3, 5)), Ht = 30, sHt = rep(0, length(Ht)), inv = NULL, Rfn = NULL, ... )tprTrees( spp = 1L, Dm = list(c(30, 28)), Hm = list(c(1.3, 5)), Ht = 30, sHt = rep(0, length(Ht)), inv = NULL, Rfn = NULL, ... )
spp |
species code, see |
Dm |
measurements of diameter along trunk |
Hm |
height of measurements along trunk |
Ht |
tree height |
sHt |
standard deviation of stem height |
inv |
indicator (0-5) for inventory to assess taper form; numeric scalar
see |
Rfn |
function to populate residual variance matrix R |
... |
arguments to be passed to |
constructor for a tprTrees object, includes a check on monotonicity of the taper curve.
object of class tprTrees.
# just define a tree tpr <- tprTrees(spp=1, Dm=30, Hm=1.3, Ht=27) plot(tpr) # define 2 trees with only dbh tpr <- tprTrees(spp=c(1,3), Dm=c(30, 35), Hm=c(1.3, 1.3), Ht=c(27, 30)) plot(tpr) # define 2 trees with several measurement tpr <- tprTrees(spp=c(1,3), Dm=list(c(30, 28), c(35, 33, 31)), Hm=list(c(1.3, 8), c(1.3, 5, 8)), Ht=c(27, 30)) plot(tpr) # define 2 trees with only dbh and inventory indicator (form) tpr <- tprTrees(spp=c(1,3), Dm=c(30, 35), Hm=c(1.3, 1.3), Ht=c(27, 30), inv=4) plot(tpr)# just define a tree tpr <- tprTrees(spp=1, Dm=30, Hm=1.3, Ht=27) plot(tpr) # define 2 trees with only dbh tpr <- tprTrees(spp=c(1,3), Dm=c(30, 35), Hm=c(1.3, 1.3), Ht=c(27, 30)) plot(tpr) # define 2 trees with several measurement tpr <- tprTrees(spp=c(1,3), Dm=list(c(30, 28), c(35, 33, 31)), Hm=list(c(1.3, 8), c(1.3, 5, 8)), Ht=c(27, 30)) plot(tpr) # define 2 trees with only dbh and inventory indicator (form) tpr <- tprTrees(spp=c(1,3), Dm=c(30, 35), Hm=c(1.3, 1.3), Ht=c(27, 30), inv=4) plot(tpr)
This class represents one or multiple trees by their biometric characteristics.
using indices i and j to subset
## S4 method for signature 'tprTrees,ANY,ANY,ANY' x[i, j, ..., drop = FALSE] ## S4 method for signature 'tprTrees' length(x) ## S4 method for signature 'tprTrees' show(object)## S4 method for signature 'tprTrees,ANY,ANY,ANY' x[i, j, ..., drop = FALSE] ## S4 method for signature 'tprTrees' length(x) ## S4 method for signature 'tprTrees' show(object)
x |
object of class 'tprTrees' |
i |
index i |
j |
index j |
... |
not currently used |
drop |
drop dimensions, defaults to FALSE |
object |
object of class 'tprTrees' |
blabla
a part of the original object
x[i: subsetting for class 'tprTrees'
length(tprTrees): length function for class 'tprTrees'
show(tprTrees): length function for class 'tprTrees'
sppspecies code of trees
Dmlist of measured diameters
Hmlist of heights of measured diameters
Httotal height of trees
sHtstandard deviation of total tree height, defaults to 0 for exact height measurements without error
monotonelogical indicator about monotonicity of taper curve
tprTrees() # initialise object by constructor (tmp <- tprTrees(spp=c(1L,3L), Dm=list(c(30, 28), c(40, 38)), Hm=list(c(1.3, 5), c(1.3, 5)), Ht=c(30, 40)))tprTrees() # initialise object by constructor (tmp <- tprTrees(spp=c(1L,3L), Dm=list(c(30, 28), c(40, 38)), Hm=list(c(1.3, 5), c(1.3, 5)), Ht=c(30, 40)))
Function calculates stem volume from taper curve for given trees, depending definition of segment and on bark indicator. It is possible to request confidence or prediction intervals.
tprVolume( obj, AB = NULL, iAB = NULL, bark = NULL, interval = "none", mono = TRUE, Rfn = NULL ) ## S4 method for signature 'tprTrees' tprVolume( obj, AB = list(A = 0, B = 7, sl = 2), iAB = c("h", "dob"), bark = TRUE, interval = "none", mono = TRUE, Rfn = NULL )tprVolume( obj, AB = NULL, iAB = NULL, bark = NULL, interval = "none", mono = TRUE, Rfn = NULL ) ## S4 method for signature 'tprTrees' tprVolume( obj, AB = list(A = 0, B = 7, sl = 2), iAB = c("h", "dob"), bark = TRUE, interval = "none", mono = TRUE, Rfn = NULL )
obj |
object of class 'tprTrees' |
AB |
list with heights or diameters A and B of section for which volume
over or under bark should be calculated. Additionally, add in |
iAB |
character indicating how to interpret given A and B values. Either "H" (the default), "Dob" (diameter over bark) or "Dub" (diameter under bark). Could be of length one or two, depending on whether A and B are both height or diameter variables or not. See examples. |
bark |
should volume be returned including ( |
interval |
character to indicate whether and which type of interval is
required; one of |
mono |
logical, defaults to true. If calibrated taper curve is non-monotonic at stem base, a support diameter is added. |
Rfn |
Rfn setting for residuals error matrix, defaults to
|
The function returns total solid wood w/ bark (i.e. from H=0 to
D=7cm) by default. Using AB, one can specify lower A and upper
B end of segments for which volume is required, w/ or w/o bark.
iAB can be a vector of length two, indicating how to interpret A and
B. Hence, one can calculate volume between a given height and a given
diameter, either over or under bark. If of length one, it is assumed that the
indicator applies to both A and B.
Defining interval 'confidence' or 'prediction' returns
lower (lwr) and upper (upr) interval bounds on confidence level
= qt(0.025, ...). NB: The volume confidence bounds only
incorporate the uncertainty of diameter estimation at a pre-fixed position
(e.g. H=1.3m). If the position is given as diameter (e.g. iAB="Dob"),
the absolute height position is calculated using the *estimated* diameter,
hence, the uncertainty of the estimated absolute height is not (yet)
included. Neither is the uncertainty of the models for bark reduction.
In contrast to the underlying R-package TapeR, which uses
E_VOL_AB_HmDm_HT.f for volume calculation, this function
calculates volume based on stem-section (default: 2m, see parameter
AB). Additionally, with that approach, bark reduction is easily
possible.
if interval='none' a vector else a matrix.
tprVolume(tprTrees): method for class 'tprTrees'
E_DHx_HmDm_HT.f for the underlying diameter
calculation.
obj <- simTrees() # default is: simulate 10 Norway spruce with mean dbh of 40 A <- 1 B <- 10 tprVolume(obj) # default is: coarse wood volume w/ bark tprVolume(obj, AB = list(A=A, B=B, sl=2), iAB = "H", bark=FALSE) tprVolume(obj, AB = list(A=A, B=B, sl=0.01), iAB = "H", bark=FALSE) tprVolume(obj, AB = list(A=A, B=B, sl=0.01), iAB = "H", bark=TRUE) ## compare against integrated taper curve volume via package TapeR ## TapeR integrates over the taper curve, while TapeS uses segments of length 'sl' SKP <- TapeS:::SKPar TapeR::E_VOL_AB_HmDm_HT.f(Hm=obj@Hm[[1]], Dm = obj@Dm[[1]], iDH = "H", mHt = obj@Ht[1], sHt = 0, A = A, B = B, par.lme=SKP[[1]])$E_VOL ## returning intervals tprVolume(obj, interval="none") tprVolume(obj, interval="confidence") tprVolume(obj, interval="prediction") tprVolume(obj, interval="prediction", bark=FALSE) tprVolume(obj, interval="prediction", AB=list(A=0.1, B=5.1, sl=0.1), iAB="H")obj <- simTrees() # default is: simulate 10 Norway spruce with mean dbh of 40 A <- 1 B <- 10 tprVolume(obj) # default is: coarse wood volume w/ bark tprVolume(obj, AB = list(A=A, B=B, sl=2), iAB = "H", bark=FALSE) tprVolume(obj, AB = list(A=A, B=B, sl=0.01), iAB = "H", bark=FALSE) tprVolume(obj, AB = list(A=A, B=B, sl=0.01), iAB = "H", bark=TRUE) ## compare against integrated taper curve volume via package TapeR ## TapeR integrates over the taper curve, while TapeS uses segments of length 'sl' SKP <- TapeS:::SKPar TapeR::E_VOL_AB_HmDm_HT.f(Hm=obj@Hm[[1]], Dm = obj@Dm[[1]], iDH = "H", mHt = obj@Ht[1], sHt = 0, A = A, B = B, par.lme=SKP[[1]])$E_VOL ## returning intervals tprVolume(obj, interval="none") tprVolume(obj, interval="confidence") tprVolume(obj, interval="prediction") tprVolume(obj, interval="prediction", bark=FALSE) tprVolume(obj, interval="prediction", AB=list(A=0.1, B=5.1, sl=0.1), iAB="H")
Wrapper to get specific type of volume from taper curve
Vfm(obj) Efm(obj, stH = 0.01) VolR(obj) VolE(obj) VolFAO(obj) Vfm_phys(obj) Efm_phys(obj, stH = 0.01)Vfm(obj) Efm(obj, stH = 0.01) VolR(obj) VolE(obj) VolFAO(obj) Vfm_phys(obj) Efm_phys(obj, stH = 0.01)
obj |
a object of class 'tprTrees' |
stH |
assumed or known relative or absolute stump height, from which volume calculation should starts, defaults to 0.01 |
wrapper functions around tprVolume, which return
specific definitions of stem volume.
Function Efm uses parameter stH to define starting
point, i.e. stump height, of volume calculation. stH can be defined
relative to total tree height (0 < stH <= 1) or in absolute measure
(unit=cm) in case stH > 1
VolE calculates as the sum of volume of default assortments
(stem wood, top log, industrial wood, X-wood, non-usuable wood according to
RVR. For dbh < 7cm a linear regression is applied.
VolFAO calculates tree volume starting from stump up to tree
top (in contrast to german definition, which uses D=7cm over bark), and
includes bark component. Stump height is defined as 1% of tree height.
Volume calculation is based on 2m-sections. For trees with dbh < 7cm,
tabulated values are used, see Riedel et al. (2017) for details (e.g. p.35,
table 5.6).
Vfm_phys is equal to Vfm, except that the taper curve
is numerically integrated, by use of section length of 0.01m. This is
relevant if biomass or nutrient export is to be calculate. Numerical
integration is quite slow.
Efm_phys is equal to Efm, except that the taper curve
is numerically integrated, by use of section length of 0.01m. This is
relevant if biomass or nutrient export is to be calculate. Numerical
integration is quite slow.
vector of volume estimates
Efm(): Efm, i.e. coarse wood excl. bark from Ht=stH*Ht to Dob=7cm
VolR(): VolR: Volume from H=0 to D=7cm over bark, measured as 2m sections
VolE(): VolE: sum of volume of default assortments according to RVR
VolFAO(): VolFAO: from stump to tree top incl. bark; if dbh < 7cm using
tabulated values
Vfm_phys(): Vfm_phys physical volume of tree incl. bark from A=0
Efm_phys(): Efm_phys physical volume of tree excl. bark from A=0.1*Ht
t <- tprTrees() # constructor of class 'tprTrees' Vfm(t) Efm(t) Efm(t, stH=0.01) # stump height = 1\% of tree height Efm(t, stH=10) # stump height=10cm VolR(t) VolE(t) VolFAO(t) Vfm_phys(t) # slower since much more evaluations of taper curve (every 1 cm) Efm_phys(t, stH=0.01) # slower since much more evaluations of taper curve (every 1 cm)t <- tprTrees() # constructor of class 'tprTrees' Vfm(t) Efm(t) Efm(t, stH=0.01) # stump height = 1\% of tree height Efm(t, stH=10) # stump height=10cm VolR(t) VolE(t) VolFAO(t) Vfm_phys(t) # slower since much more evaluations of taper curve (every 1 cm) Efm_phys(t, stH=0.01) # slower since much more evaluations of taper curve (every 1 cm)