4. Methods of Estimating Belowground Net Primary Production
63
al. (1996) analysis, N adlehoffer et al. (1998) argued
that the Gower et al. (1996) analysis was largely
inappropriate and added no new information to the
question of the applicability the TRCA method.
Smith and Resh (1999) successfully applied the
conceptual (not the statistical) TRCA approach to
a group of stands to evaluate whether increasing
allocation to BNPP as forests age would provide an
explanation for the widely observed decline in
ANPP with increasing age. They were able to reject
the increasing belowground allocation hypothesis
as a single explanation for the decline in ANPP,
suggesting that a detailed carbon budget evaluation
can provided useful information in an analysis of
BNPP and NFRP.
Nitrogen Balance
This method assumes that all available nitrogen is
taken up by plants and allocated to either aboveground litter, woody tissue, or fine roots (Aber et
al. 1985; Nadelhoffer et al. 1985; see also Chapter
16). The amount of nitrogen allocated to fine roots
is:
Nfr = Na - N w - Nal
(4.7)
where Nfr is the amount of nitrogen allocated to fine
roots, Na is available nitrogen, N w is nitrogen allocated to woody tissue, and Nal is nitrogen allocated to aboveground litter. NFRP is then calculated by:
NFRP = N fr
Neone
(4.8)
where N eone is the concentration of nitrogen in fine
roots. This method transforms the problem of estimating BNPP from a focus on changes in root
biomass to a focus on annual nitrogen availability
(see Chapter 14) and the amounts of nitrogen allocated to aboveground litter and to wood (see
Chapters 2 and 16). Available nitrogen is calculated
using the following assumption:
where N m is measured nitrogen mineralization (see
Chapter 14), Np is the amount of nitrogen received
in precipitation (see Chapter 17) and Nl is the
amount of nitrogen lost to leaching (see Chapter
16). A sample calculation is illustrated in Table 4.2.
This method has not been widely used. Aber et
al. (1985) compared the nitrogen balance method
to a biomass method and concluded that for stands
with low nitrification the two methods compared
very well. For stands with high nitrification, Aber
et al. (1985) suggested that the larger estimates of
NFRP from the nitrogen balance method were better estimates than those from the biomass method.
The limitations of this method are related to the
assumptions that it makes about nitrogen allocation
and the accuracy of estimates of annual nitrogen
availability and nitrogen concentration in fine roots
(Aber et al. 1985; Nadelhoffer and Raich 1992).
Minirhizotrons
Root observation methods are among the oldest
methods to be used to evaluate root and rhizome
dynamics (B6hm 1979). However, it has only been
since the recent availability of small video cameras
that the method has been considered for estimating
BNPP (Upchurch and Ritchie 1983; Ferguson and
Smucker 1989; Cheng et al. 1990; Hendrick and
Pregitzer 1992).
The method involves placing transparent tubes
in the soil to a depth determined by the objective
of the study (Brown and Upchurch 1987). Observations of roots intersecting the tubes are made at
intervals and the images are recorded on videotape
(Brown and Upchurch 1987). The video images
must be converted to digital images for analysis,
which requires digitizing the root outline (Hendrick
and Pregitzer 1992) or counting the number of intersections of roots with the tube (Merrill and Upchurch 1994). While the potential for automating
the digitizing step exists, to date the problem of
extracting the root images from the background of
organic debris has proven to be very difficult
(Smucker 1993). Manually digitizing the root images is very time consuming.
Root length is measured at each sample date and
the change in root length is expressed as a proportion of the root length on the first sample date (Hendrick and Pregitzer 1992, 1993). The amount of
root length added as a proportion of the initial
amount is a turnover coefficient. Production can be
expressed either in terms of root length (Hendrick
and Pregitzer 1992) or in terms of biomass (Hendrick and Pregitzer 1993). NFRP or BNPP is calculated by assuming that the turnover in root length
63
al. (1996) analysis, N adlehoffer et al. (1998) argued
that the Gower et al. (1996) analysis was largely
inappropriate and added no new information to the
question of the applicability the TRCA method.
Smith and Resh (1999) successfully applied the
conceptual (not the statistical) TRCA approach to
a group of stands to evaluate whether increasing
allocation to BNPP as forests age would provide an
explanation for the widely observed decline in
ANPP with increasing age. They were able to reject
the increasing belowground allocation hypothesis
as a single explanation for the decline in ANPP,
suggesting that a detailed carbon budget evaluation
can provided useful information in an analysis of
BNPP and NFRP.
Nitrogen Balance
This method assumes that all available nitrogen is
taken up by plants and allocated to either aboveground litter, woody tissue, or fine roots (Aber et
al. 1985; Nadelhoffer et al. 1985; see also Chapter
16). The amount of nitrogen allocated to fine roots
is:
Nfr = Na - N w - Nal
(4.7)
where Nfr is the amount of nitrogen allocated to fine
roots, Na is available nitrogen, N w is nitrogen allocated to woody tissue, and Nal is nitrogen allocated to aboveground litter. NFRP is then calculated by:
NFRP = N fr
Neone
(4.8)
where N eone is the concentration of nitrogen in fine
roots. This method transforms the problem of estimating BNPP from a focus on changes in root
biomass to a focus on annual nitrogen availability
(see Chapter 14) and the amounts of nitrogen allocated to aboveground litter and to wood (see
Chapters 2 and 16). Available nitrogen is calculated
using the following assumption:
where N m is measured nitrogen mineralization (see
Chapter 14), Np is the amount of nitrogen received
in precipitation (see Chapter 17) and Nl is the
amount of nitrogen lost to leaching (see Chapter
16). A sample calculation is illustrated in Table 4.2.
This method has not been widely used. Aber et
al. (1985) compared the nitrogen balance method
to a biomass method and concluded that for stands
with low nitrification the two methods compared
very well. For stands with high nitrification, Aber
et al. (1985) suggested that the larger estimates of
NFRP from the nitrogen balance method were better estimates than those from the biomass method.
The limitations of this method are related to the
assumptions that it makes about nitrogen allocation
and the accuracy of estimates of annual nitrogen
availability and nitrogen concentration in fine roots
(Aber et al. 1985; Nadelhoffer and Raich 1992).
Minirhizotrons
Root observation methods are among the oldest
methods to be used to evaluate root and rhizome
dynamics (B6hm 1979). However, it has only been
since the recent availability of small video cameras
that the method has been considered for estimating
BNPP (Upchurch and Ritchie 1983; Ferguson and
Smucker 1989; Cheng et al. 1990; Hendrick and
Pregitzer 1992).
The method involves placing transparent tubes
in the soil to a depth determined by the objective
of the study (Brown and Upchurch 1987). Observations of roots intersecting the tubes are made at
intervals and the images are recorded on videotape
(Brown and Upchurch 1987). The video images
must be converted to digital images for analysis,
which requires digitizing the root outline (Hendrick
and Pregitzer 1992) or counting the number of intersections of roots with the tube (Merrill and Upchurch 1994). While the potential for automating
the digitizing step exists, to date the problem of
extracting the root images from the background of
organic debris has proven to be very difficult
(Smucker 1993). Manually digitizing the root images is very time consuming.
Root length is measured at each sample date and
the change in root length is expressed as a proportion of the root length on the first sample date (Hendrick and Pregitzer 1992, 1993). The amount of
root length added as a proportion of the initial
amount is a turnover coefficient. Production can be
expressed either in terms of root length (Hendrick
and Pregitzer 1992) or in terms of biomass (Hendrick and Pregitzer 1993). NFRP or BNPP is calculated by assuming that the turnover in root length