Environmental Engineering Reference
In-Depth Information
Temperate grasslands i ll the semiarid and arid centers of North America and Asia,
while temperate deciduous forests were the natural climax vegetation across most of
Atlantic and Eastern Europe, southeastern North America, and parts of northeastern
Asia. Extensive boreal forest (taiga) spreads across North America from the Pacii c
coast to the Atlantic and in Europe from Scandinavia to the Sea of Okhotsk. Treeless
tundra is the least biodiverse biome found along the northern fringes of North
America and Eurasia. The perpetually dark and cold abyssal, nutrient-poor waters of
the open ocean are the aquatic counterpart of desert. Highly diverse marine biomes
are limited to relatively small areas of coral reefs and kelp forests, to the regions of
upwelling (where rising cold water replenishes surface nutrients) along the western
coast of the Americas and Africa, and to estuaries of major rivers.
Before citing any numbers for phytomass densities I must stress several important
yet often ignored facts, none of them more critical than our surprisingly limited
basis for accurate assessments of global phytomass. Houghton, Hall, and Goetz
(2009, 1) put it best: “Our knowledge of the distribution and amount of terrestrial
biomass is based almost entirely on ground measurements over an extremely small,
and possibly biased sample, with many regions still unmeasured. Our understanding
of changes of terrestrial biomass is even more rudimentary.” This conclusion is true
even when we take into account the latest advances in quantifying carbon in tropical
forests (Saatchi et al. 2011), although the creators of many global vegetation models
and the even more complex simulations of biosphere-atmosphere interactions want
us to believe that their understanding has become deep enough to recreate faithful
virtual versions of Earth's biomass
in silico
Besides the limited basis of actually measured phytomass stocks, we have to
reckon with a large variability (spatial and temporal) of phytomass densities. The
possibilities of bias resulting from destructive sampling of small plots were noted
in the last chapter, but even the best available large-scale mapping of forest ecosys-
tems frequently ends with very different totals (Houghton et al. 2001). The range
of phytomass densities spans three orders of magnitude (the difference between the
northernmost tundras and giant trees in the rain forest of the Pacii c Northwest),
and even the difference between averages in cold and hot deserts, on the one hand,
and the richest forests on the other is more than 30-fold.
Spatial differences in average forest density are often an order of magnitude even
within 1 ha (100
100 m) vegetation patches, while even the highest resolution
that is now readily available for satellite vegetation imaging is 250 m, lumping
all phytomass within 6.25 ha into a single category; the other two monitoring reso-
lutions make a single data point from areas of 500
×
×
500 m and 1 km
×
1 km.
