Concluding Remarks

Aerial net primary production calculated from papyrus biomass dynamics at the experimental site at Lake Naivasha was 2.06 kg C m^-2 yr^-1 (Muthuri et al. 1989). This corresponds to 21% of the gross carbon fixation and 64% of the net carbon fixation predicted by the WIMOVAC model which calculates canopy carbon uptake and loss from a process-based model of leaf photosynthetic characteristics and canopy microclimate. Clearly, large amounts of the carbon fixed by the aerial shoots are partitioned to the below-ground, or more strictly speaking below-water, component of the papyrus plants. Estimates of below-ground standing biomass show that as much as 60% of the total plant biomass is in this component which consists of rhizomes (> 98% of dry weight) and roots (Jones and Muthuri, 1997). In a well established papyrus swamp community the below-ground component appears to show little net change in mass over time (Muthuri, 1985) so that all carbon gained from the aerial shoots is lost in turnover, either in respiration or as detritus. Our scaled estimate of rhizome respiration at 1.56 kg C m^-2 accounts for more than the total amount of carbon we estimate as going below ground. This would suggest that the detritus builds up very slowly in these swamps and that on an annual basis they are a minor sink for carbon. Of course, the detritus, as it accumulates, must be a significant sink for carbon over extended periods of time but we have little or no information on the rate at which it accumulates. It is this rate of accumulation which will be a measure of the activity of these wetlands as carbon sinks.

Figure 9 shows a schematic summary of carbon balance in the papyrus swamp based upon combined information from model simulation, eddy covariance and respiration measurements. The overall C budget suggests a net deficit for the papyrus ecosystem of approximately 0.96 kg C m^-2 y^-1 during the time of field measurements reported here. When the eddy covariance flux measurements alone were used to calculate the balance, based upon daytime carbon uptake and night-time carbon loss, the net deficit was 1.0 kg C m^-2 y^-1. It is still unclear from our data whether the main source of the C loss is structural stand material, or peat deposits below the papyrus. The draw-down of the lake during our measurements exposed substantial amounts of the detritus in the form of peaty deposits to aerobic conditions and it would therefore be expected that this is a major source of carbon. Further work examining the dynamics of above and below ground biomass is necessary to elucidate the source of the lost C.

Although the study site is thought to show relatively little seasonality, and measurements were made at contrasting times of the season, such short term measurements cannot give a wholly adequate picture of the C balance of the ecosystem over a longer period. It is possible, for example, that papyrus shows cycles of productivity with a long periodicity associated with ageing, weather or other factors. Lake water levels at Naivasha are known, for example, to show an unpredictable pattern due to variable rates of influx from tributary rivers and below ground aquifers and the changeable rates of water loss associated with evapo-transpiration and commercial abstraction from the lake.

When considering the carbon budget of the Papyrus swamp, the respiratory losses of carbon are clearly very significant. Scaled measurements of canopy respiration (6.28 kg C m^-2 yr^-1) compare well with predictions based on the McCree (1970) equation (6.17 kg C m^-2 yr^-1). Here we assume the low-end estimate suggested by McCree (1970) of 25% of gross assimilation (2.37 kg C m^-2 yr^-1) for growth associated respiration and an equivalent of 1% of above ground dry weight for maintenance respiration (3.8 kg C m^-2 yr^-1). However this comparison is less favourable if root/rhizome material is included in the maintenance respiration term (9.1 kg C m^-2 yr^-1) and a comparison made to canopy plus root/rhizome respiration measurements (7.84 kg C m^-2 yr^-1). This and the non-uniform distribution of nitrogen in culm, umbel and root/rhizome, shown in Table 1, suggests that alternative parameterisation or a more complex treatment than that proposed by McCree (1970) is needed to successfully account for respiration in papyrus when direct measurements are not available to parameterize scaling functions within the model.

An interesting conclusion from the modelling of year-round carbon exchange is that although mid-day maxima of canopy assimilation in June and December at Naivasha appear similar, the predicted net C gain in June is substantially higher. This results primarily from a decrease in air and canopy temperature associated with the rainy season and a concomitant decrease in respiration. At the ambient air temperatures measured at the study site model predictions indicate that an increase in air temperature leads to increased respiratory losses which appear to be greater in magnitude than corresponding increases in C uptake by canopy photosynthesis. These results indicate the marked sensitivity to temperature of the net carbon balance of the swamp. Furthermore, given that the temperatures at Lake Naivasha are well below the optima for C₄ photosynthesis (Jones, 1987) it might be expected that an increase in temperature would result in an increase in net canopy carbon gain. However the opposite is predicted here as a result of two causes. First, with a dense canopy (LAI 5-8) most of the foliar elements will not be light saturated and when strictly light limited, photosynthesis in C₄ species is largely independent of temperature . Secondly an increase in temperature will substantially increase respiratory losses for such a large canopy and this would act to decrease net carbon gain. This is consistent with the observation that papyrus standing dry matter production measured at Lake Naivasha is among the highest values recorded, even though mean temperatures are lower than at other sites at which production has been measured. There is a strong negative relationship between standing above ground dry matter and annual mean temperature for four locations in East Africa (Jones, pers. comm.) (Figure 10) and while this does not necessarily imply that there is a similar pattern of net productivity it is highly suggestive that this might be the case.

Both model prediction and productivity measurements suggest that rising temperatures may result in substantial decreases in C sequestration and inputs to stored carbon in the underlying peat deposits. Further, if the large difference between net production estimated from biomass dynamics and modeled gas exchange can be adequately explained by the large respiratory load from submerged root and rhizome material then this suggests that a decrease in the net supply of available carbon from the shoot, with increased, temperature, could cause a prolonged negative carbon balance and possible loss of papyrus stands.

Unfortunately there is currently little evidence for the magnitude or direction of temperature acclimation effects in papyrus and whilst these results provide an indication of some of the more direct effects of temperature the long term effects of acclimation may be of similar magnitude and opposite direction.

[ Home ] [ Up ]

Send mail to humph@essex.ac.uk with questions or comments about this web site.
Copyright © 1998 WIMOVAC Ltd.
Last modified: August 19, 1997