Climatic responses

It is evident from the above yearly budgets that the ecosystem is very sensitive to changes in the environmental conditions (Malhi et al., 1999). The balance between GPP and R $_{t}$ is responsible for the observed variability of NEE. If GPP or R $_{t}$ changes, both affects NEE and may lead to a situation when NEE even changes sign. That means that the region should turn to be a net source of CO $_{2}\protect$ to the atmosphere. A warming climate in the Carpathian Basin may enhance ecosystem respiration in an exponential fashion (see eq.

). If it is not compensated with GPP (e.g. because of drought which causes high VPD and low soil water content leading to reduced photosynthesis), NEE becomes positive. If this happens, this behaviour may act as a positive feedback to global warming.

Calculating monthly sums of NEE, GPP and R $_{t}$ , we may try to find relationship between the monthly mean climat variability and the variability in the measured carbon fluxes.

**Figure:** Monthly sums of NEE, R $_{t}$ , GPP and PPFD, monthly average temperatures, monthy sums of precipitation and monthly average daytime VPD during 1997, 1998 and 1999. See text for details.
$\includegraphics{resultv2.eps}$

Mean annual air temperature was 9.1 $^{o}$ C in 1997, 9.6 $^{o}$ C in 1998 and 9.6 $^{o}$ C in 1999. Annual precipitation was 695.3 mm in 1997, 844.4 mm in 1998 and 784.9 mm in 1999. As it was described in table

, total respiration in 1997 was lower than in the other years, while it was about the same for 1998 and 1999. This can be explained by the lower annual average temperature (see eq.

). Mean air temperature was about the same in 1998 and 1999, which resulted in very similar respiration rates.

In order to gain more insight into the processes governing the carbon balance, we can investigate the relation between the monthly sums of NEE, R $_{t}$ and GPP, and the environmental forces.

Figure

shows (from top to bottom) the monthly sums of NEE, R $_{t}$ , GPP (negative values represent carbon uptake by the vegetation), and PPFD, the monthly average temperature and the monthly sums of precipitation as measured by a nearby meteorological station (Szentgotthárd; the overplotted solid line is the normal values between 1961 and 1990 for both plots), and the monthly average daytime vapor pressure deficit for 1997, 1998 and 1999.

We may deduce more linkage between the variability of the carbon fluxes in monthly scale from figure

Vapor pressure deficit does not seem to affect photosynthesis essentially. This is demonstrated with July and August 1998. VPD during August was considerably higher than during July, but rates of GPP was equal, which led to very similar rates of GPP. As we saw it before in fig.

, photosynthesis is only affected by higher VPD values (VPD1.5 kPa), but not in every cases. VPD may affect GPP in a daily or weekly scale, but it doesn't seem to govern GPP in monthly or annual time scale.

Extreme events may be explained by other linkages. Since variability of NEE is expected to be derived by its subcomponents, we seek relationships between the climate forces and GPP and R $_{t}$ . GPP is expected to be controlled by the monthly sums of PPFD. R $_{t}$ is mainly derived by temperature, but soil CO $_{2}\protect$ efflux may also be important. The following case studies reflect the climate responses of the biosphere.

In August 1999 NEE was unusually low. This is caused by the lower rates of PPFD which is resulted in lower GPP (compared to other years). Low PPFD is tied to the excess precipitation (cloudiness). In contrast, during July 1999 NEE was quite large. This is caused by the high rates of GPP compared to GPP measured during July 1997 and 1998.

High respiration sum occured during July 1998. This is caused by above average precipitation during average temperature conditions which supports microbial respiration in the soil. In contrast, during July 1997 above average precipitation occured together with below average temperature. This results in normal respiration rates.

High NEE occured during May 1998 which is caused by reduced respiration. This can be explained by the occurence of average temperature (during July 1997 and 1999 temperature was above average) and normal GPP.

We should note that not every extreme monthly event can be explained by such reasoning.

All of the above confirms the complexity of linkages between NEE and the environmental factors. The sensitivity of carbon sequestering ability of the region is also evident from the data. Further measurements are needed to confirm the observed linkages and to find others, and to provide forecasts about the carbon balance of the region under a changing climate (IPCC website, 2001).