Management effects

The site-specific nature of carbon storage is illustrated in Fig. 8.1 simulating the Terrestrial Ecosystem Model (TEM) (Melillo et al., 199g; Xiao et al., 1997, 1998; Tian et al., 1999, 200g; Felzer et al., 2004, 2005) for two sites and under three man-

Potential vegetation

(Temperate deciduous forest) Crops, rain-fed, no fertilization Crops, irrigated, fertilized

Potential vegetation

(Temperate deciduous forest) Crops, rain-fed, no fertilization Crops, irrigated, fertilized

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 (a) Year

Fig. 8.1. Simulated historical changes in reactive soil organic carbon (RSOLC) at agricultural sites in (a) Buffalo, New York, and (b) Bakersfield, California, under three hypothetical management scenarios. Note that fertilizer application did not occur until 1950 in the fertilized scenario and that cropland at the Bakersfield site was abandoned in 1965.

6,000

S 3,000

to cjJ

W 1,000

6,000

S 3,000

to cjJ

W 1,000

^â– abandoned

Potential vegetation

(arid shrubland) Crops, rain-fed, no fertilization Crops, irrigated, fertilized i j t i

N

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 Year

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 (a) Year

1900 1910 1920 1930 1940 1950 1960 1970 1980 1990 2000 Year

Fig. 8.1. Simulated historical changes in reactive soil organic carbon (RSOLC) at agricultural sites in (a) Buffalo, New York, and (b) Bakersfield, California, under three hypothetical management scenarios. Note that fertilizer application did not occur until 1950 in the fertilized scenario and that cropland at the Bakersfield site was abandoned in 1965.

agement regimes. Several observations are worth making. First, simulated reactive soil organic carbon (RSOLC) based on natural conditions (potential vegetation) varies by an order of magnitude between the sites. The arid Bakersfield site holds only about 1000 g RSOLC per square meter (g C/m2), whereas the Buffalo site under natural conditions was estimated to hold about 10,000 g C/m2.1 Second, cropping was estimated to significantly reduce carbon storage at both sites, but the reduction was far greater at the Buffalo site in absolute as well as percentage terms of RSOLC. Third, while it is often assumed that the difference between the carbon in currently degraded soil and that prior to degradation represents the potential amount of carbon that could be stored, that difference is largely irrelevant to estimates of increased storage when a different management practice is applied. In particular, the Buffalo site with the addition of fertilizer and irrigation only gains back somewhat more than half of the RSOLC lost when converted to crops. In contrast, irrigation and fertilization leads to an increase in RSOLC at the arid Bakersfield site several times that under natural conditions. Fourth, as illustrated for the Bakersfield site, if management is removed, carbon storage can change substantially. In this case, much of the modelled increase in RSOLC due to irrigation and fertilization was lost in just a few years once the site was aban doned. Interestingly, it appears that some of the additional carbon stored may remain even after being abandoned for as many as 35 years. Even though it fell after abandonment, RSOLC remained on the order of 80% above the natural level. The management regime (and abandonment) was set to represent the actual historical management at these sites. Abandonment of cropping at the Buffalo site would likely lead to a further increase in carbon, perhaps back to near the predisturbed level. Other management practices alone or in combination may lead to other results, but our conclusions from just these two sites are that the impact of different management practices on carbon storage can differ by an order of magnitude, and that the 'predisturbed' soil carbon level is not always a clear guide to how much carbon could be stored.

0 0

Post a comment