Environmental Engineering Reference
In-Depth Information
biomes and ecosystems, and the questionable validity
of extrapolating detailed local assessments to larger areas.
Errors in converting volume to mass and energy equiva-
lents are another problem. Conversion of green volume
of wood to dry matter phytomass ranges between 0.4
and 0.8, with typical rates between 0.55 and 0.60. As a
result, there is less consensus on global phytomass than
on global NPP. Vernadsky's (1926) estimate of the
Earth's green matter in the first edition of his pioneering
topic on the biosphere, corresponding to 10
13
-10
14
tC,
proved to be an enormous exaggeration. Whittaker and
Likens (1975) estimated continental phytomass in 1950
at 1.837 Tt, or 827 Gt C.
Olson, Watts, and Allison (1983) improved the accu-
racy by subdividing continents into 0
:
5
0
:
5
cells and
by collecting the best available data on climatic factors,
vegetated areas, and phytomass ranges on that scale.
Their range was 460-660 (mean of 560) Gt C. Models
of the global carbon cycle published during the 1990s
contained total continental phytomass values as low as
486 Gt C (Amthor et al. 1998) and as high as 780 Gt C
(Post, King, and Wullschleger 1997). Pilot Analysis of
Global Ecosystems opted for a surprisingly broad range
of 268-901 Gt C (Matthews et al. 2000), whereas a
comprehensive review of global NPP offered 652 Gt C
as its best estimate (Roy, Saugier, and Mooney 2001).
Disparities in the categorization of land cover and differ-
ences in assumed phytomass densities explain these dif-
ferences. Land cover assessments published since 1980
have used values as low as about 25 Gm
2
and as high
as 75 Gm
2
for the total area of the Earth's forests (Ema-
nuel, Shugart, and Stevenson 1985; Solomon et al. 1993;
Cramer et al. 1999; FAO 2005a). In contrast, oceanic
phytomass adds up to only 1-3 Gt C.
With 500-800 Gt C, the biosphere's phytomass binds
an equivalent of 62%-99% of the element's current
atmospheric content (809 Gt C in 2005). Uniform dis-
tribution of dry terrestrial phytomass over ice-free land
would produce a layer about 1 cm thick; the same pro-
cess in the ocean would add a mere 0.03 mm of phyto-
plankton (in both cases, I assume an average biomass
density of 1 g/cm
3
). I know of no better examples to
illustrate the evanescent quality of life, but a different def-
inition of phytomass would produce an even smaller ter-
restrial total. Most of the structural polymers that play
essential supportive, protective, and conductive roles are
not alive, and this reality makes it possible to argue for
both drastically reducing and greatly expanding the defi-
nition of the Earth's phytomass.
The first course would be to restrict the phytomass
definition to the living protoplasm but, as a closer look
at tree phytomass illustrates, it is difficult to offer reliable
large-scale corrective multipliers. The radial extent of the
cambial zone, the generator of tree growth, is difficult to
define because of the gradual transition to differentiating
xylem and phloem. Most of the conducting tissue in
trees is sapwood, with typically only 5%-8% of living cells
in conifers and 10%-20% of living cells in hardwoods.
Conversion of sapwood into nonconducting heartwood
involves death of the cytoplasm of all living cells in soft-
woods, but in hardwoods some cells in axial and radial
parenchyma remain alive for years and decades. And
there are substantial specific differences in the shares of
total phytomass made up of fresh leaves, buds, young
branches, and rapidly growing fine roots (Gartner 1995;
Waring and Running 1998). In addition, trees and
shrubs have dead branches and dead roots. If, in a strict
sense, no more than 15% of all forest phytomass were
