Climate

Weather measurements, made over the experimental period showed that temperatures were close to the 10 year mean values given by Muthuri et al., (1985). The mean daily temperature was 16.6 + 0.2 °C with a daily maximum of 25.5 + 0.3 °C and minimum of 8.3 + 0.3 °C. The average relative humidity for the site was 70.5 + 1.2 %, with a daily minimum of 36.8 + 1.2 %, and a maximum of 94.8 + 0.7%. Maximum daily solar radiation at the site was essentially constant, with an average of 1920 + 162 µmol m^-2 s^-1 (not shown). The predominant wind direction was north westerly and corresponded to a breeze from lake to land during the day with a south easterly breeze from land to lake in the evenings. Wind speeds were generally low and did not exceed 4.8 m s^-1 (not shown).

Ecosystem carbon flux using eddy covariance

Eddy covariance CO₂ flux measurements for 21 non-consecutive days are shown in Figure 1. Data was chosen for those days in the periods (4 - 21 August, 1995 and 11 - 17 March, 1996) for which continuous Eddy covariance measurements and meteorological data were available for the entire day. The maximum rate of ecosystem CO₂ gain (where negative values by convention indicate net gain of CO₂) was 45.1 µmol m^-2 s^-1, and the maximum rate of loss 43.3 m mol m^-2 s^-1.

When numerically integrated, the measured rates indicated an average loss of C from the ecosystem of 2.88 g C m^-2 day^-1 with a total loss over the measurement period of 63.2 g C m^-2. Although peak C loss at night and peak C gain during the day were broadly comparable the duration of the net loss period each day was longer than the period of gain resulting in the overall net loss balance. Frequently the period of net carbon gain was not evident until 3-4 hours after sunrise and finished several hours before sunset.

Respiration

The specific respiration rate of juvenile culms was 0.74 + 0.02 mg C g^-1 C h^-1, while mature and senescent culms had slightly lower respiration rates of 0.62 + 0.02 mg C g^-1 C h^-1 and 0.51 + 0.01 mg C g^-1 C h^-1 respectively. An average culm specific respiration rate (0.59 + 0.05 mg C g^-1 C h^-1) was determined based upon the relative weight contribution of each culm age class to total canopy culm dry weight. The umbel respiration rate shown in Table 1 is based upon bracteole dark respiration rates measured using the IRGA and bracteole area values from Jones and Muthuri (1985). Specific rhizome respiration was 0.1 + 0.01 mg C g^-1 C h^-1. Detritus respiration measured on a ground area basis was 8.8 + 0.3 µmol CO₂ m^-2 s^-1 and consisted of respiration from root/rhizome (3.4 + 0.2 µmol CO₂ m^-2 s^-1) and microbial fractions (5.4 + 0.35 µmol CO₂ m^-2 s^-1) as indicated in Table 1.

(a) Light response of photosynthesis

The response of bracteole photosynthesis to varying light levels was predicted using WIMOVAC (Figure 2(i)). Measured values are based on the data of Jones (1987). Although predicted and measured points differ, the scatter of predicted points shows no obvious systematic variation and the R² value (0.86) calculated from Figure 2(ii) suggests a relatively high degree of conformity.

Figure 3(i) indicates an average bracteole dark respiration rate of 2.4 µmol CO₂ m^-2 s^-1 at 28° C. This is consistent with measurements made on other C₄ plants and somewhat higher than the original value of 0.8 µmol CO₂ m^-2 s^-1 proposed by Collatz (1992). The light response curve (Figure 2(i)) indicates an apparent quantum yield for the model of 0.04 mol mol^-1 which is in keeping with Collatz’s (1992) original parameterisation.

(b) A/C_i response

Figure 3(i) shows the measured A/C_i response of ten bracteoles selected at random from different umbels. The solid line in Figure 3(i) indicates the A/C_i response predicted using the model and assuming light saturating conditions (> 1500 µmol m^-2 s^-1) and a constant VPD of 0.5 kPa. The inset of Figure 3(i) shows the close conformity (R² = 0.75) between predicted (A’) and measured (A) bracteole photosynthesis in response to atmospheric CO₂ concentration under identical temperature and VPD conditions. The measured A/C_i response curve shows no systematic variation from the predicted solid line that assumes a fixed VPD and temperature. A component of the variation apparent in the figure will have arisen from small fluctuations in temperature and VPD between the measurement conditions and those used to predict the solid line. Figure 3(ii) shows that the stomatal model (g_s’) also provided a close prediction (R² = 0.89) of the measured response (g_s) of stomatal conductance to atmospheric CO₂.

(c) Diurnal patterns of photosynthesis

A comparison of the predicted and measured diurnal values of bracteole photosynthesis values (Figure 4(i)) showed a strong conformity. The greatest variations were typically associated with dawn and dusk measurements and may have arisen due to low sun angle in the early morning and water stress effects at dusk.

Light interception by the canopy

Simply separating the canopy model of radiation distribution into culm and umbel elements gave a close agreement between the predicted and the measured extinction of light in the canopy (Figure 5(i)). Measurements and model predictions indicated that 90% of light interception occurred within the umbel component of the canopy with the culms receiving 10% or less of total incident light above the canopy. The low light level, vertical orientation and high associated respiration rate of the culm elements ensure that their relative contribution to overall canopy gross photosynthesis is low and usually represent a net C loss to the canopy as a whole.

Model Predictions

Figure 6(i) illustrates the predicted daily course of gross canopy CO₂ exchange for late June (D_j = 180), when photon flux was at its lowest for the site and late December (D_j = 360) when photon flux was highest (Figure 6(iii)). The model prediction assumes a leaf area index (LAI) of 5 (Jones and Muthuri, 1985) and the ten year mean meteorological values of Jones and Muthuri (1985). On both dates, peak gross canopy photosynthesis was predicted to be similar but higher temperatures in December (Figure 6(iv)) increased respiration (Figure 6(ii) and led to lower net canopy C gain with values of 10 g C day^-1 and 7.32 g C day^-1 for June and December respectively.

Total aerial respiratory C losses for the June day were estimated to be 8 g C day^-1 from the culm and 7.2 g C day^-1 from the umbel. Respiration from detritus including roots and rhizome was estimated at 15.2 g C day^-1 with 11.2 g C day^-1. Corresponding figures for culm, umbel and soil on the December day were 10.8, 9.2 and 12 g C day^-1 respectively.

When these daily cycles were integrated (Figure 7(i)) the gross carbon gain by the canopy was found to be almost constant at 0.8 kg C m^-2 month^-1 with a total of 9.44 kg C m^-2 for the year. Net C gain by above ground structures varied from a minimum at the start of the year (0.23 kg C m^-2 month^-1) to a maximum in July (0.31 kg C m^-2 month^-1) (Figure 6(i)). The total annual net C gain was 3.16 kg C m^-2. Cumulative annual respiration was 4.12, 3.36 and 2.92 kg C m^-2 for detritus, culm and umbel components respectively. Growth and maintenance of culm structures were predicted to represent a net loss of 2.52 kg C yr^-1 whilst umbel elements represented a net gain of 5.68 kg C yr^-1.

Eddy covariance

[ Home ] [ Up ] [ List of Figures ] [ Figure 1 ] [ Figure 2 ] [ Figure 3 ] [ Figure 4 ] [ Figure 5 ] [ Figure 6 ] [ Figure 7 ] [ Figure 8 ] [ Figure 9 ] [ Figure 10 ] [ Table 1 ]

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