Jhm. Thornley et Mgr. Cannell, Dynamics of mineral N availability in grassland ecosystems under increased[CO2]: hypotheses evaluated using the Hurley Pasture Model, PLANT SOIL, 224(1), 2000, pp. 153-170
The following arguments are outlined and then illustrated by the response o
f the Hurley Pasture Model to [CO2] doubling in the climate of southern Bri
tain.
1. The growth of N-limited vegetation is determined by the concentration of
N in the soil mineral N pools and high turnover rates of these pools (i.e.
, large input and output fluxes) contribute positively to growth.
2. The size and turnover rates of the soil mineral N pools are determined o
verwhelmingly by N cycling into all forms of organic matter (plants, animal
s, soil biomass and soil organic matter - 'immobilisation' in a broad sense
) and back again by mineralisation. Annual system N gains (by N-2 fixation
and atmospheric deposition) and losses (by leaching, volatilisation, nitrif
ication and denitrification) are small by comparison.
3. Elevated [CO2] enriches the organic matter in plants and soils with C, w
hich leads directly to increased removal of N from the soil mineral N pools
into plant biomass, soil biomass and soil organic matter (SOM). 'Immobilis
ation' in the broad sense then exceeds mineralisation. This is a transient
state and as long as it exists the soil mineral N pools are depleted, N gas
eous and leaching losses are reduced and the ecosystem gains N. Thus, net i
mmobilisation gradually increases the N status of the ecosystem.
4. At the same time, elevated [CO2] increases symbiotic and non-symbiotic N
-2 fixation. Thus, more N is gained each year as well as less lost. Effecti
vely, the extra C fixed in elevated [CO2] is used to capture and retain mor
e N and so the N cycle tracks the C cycle.
5. However, the amount of extra N fixed and retained by the ecosystem each
year will always be small (ca. 5-10 kg N ha(-1) yr(-1)) compared with amoun
t of N in the immobilisation-mineralisation cycle (ca. 1000 kg N ha(-1) yr(
-1)). Consequently, the ecosystem can take decades to centuries to gear up
to a new equilibrium higher-N state.
6. The extent and timescale of the depletion of the mineral N pools in elev
ated [CO2] depends on the N status of the system and the magnitude of the o
verall system N gains and losses. Small changes in the large immobilisation
-mineralisation cycle have large effects on the small mineral N pools. Cons
equently, it is possible to obtain a variety of growth responses within 1-1
0 year experiments. Ironically, ecosystem models - artificial constructs -
may be the best or only way of determining what is happening in the real wo
rld.