Figure 4.5 Vertical profiles of (a) the cumulative leaf area (CLA) measured (circles) and approximated by Equation 4.1 (solid line), and the relative photosynthetic photon flux density (RPPFD) calculated by Equation 4.4 (broken line); (b) leaf-area-based R20 at Tamb = 10, 20 , and 30 °C, which were estimated by Equation 4.5;
(c) leaf area density (LAD) given by Equation 4.2; and (d) land-area-based leaf R20 at each 1-m-strata (R20, canopy) at Tamb = 16.7 °C (annual mean temperature in 2012), in a 10-year-old hinoki cypress canopy. Sunlit and Shaded indicate contributions of sunlit and shaded leaves, respectively.
R20, canopy (mol m2 m1 s1)
0.0 0.5 1.0 1.5
0 1 2 3 4 5 6 7 CLA (m2 m2)
0 2 4 6 8
Depth from the canopy top, z (m)
0 1 2 3 4 5 6 7
RPPFD
0.0 0.5 1.0
Measured CLA
Approximated CLA
Calculated RPPFD
R20 (mol m2 s1)
0.0 0.5 1.0 1.5
0 1 2 3 4 5 6 7
Tamb = 10 C
Tamb = 20 C Tamb = 30 C
LAD (m2 m2 m1)
0 1 2 3
0 1 2 3 4 5 6 7
(a) (b)
(c) (d)
Sunlit
Shaded
LAI = 6.73 Tamb = 16.7 oC
Sunlit
Shaded
4.3.2 Response of canopy respiration to temperature increases
Monthly rates of nighttime canopy respiration per unit land area at present and in the future are presented in Figure 4.6. Seasonal patterns of monthly canopy R differed considerably between the variable leaf Q10 and fixed leaf Q10 models. In particular, canopy R during summer estimated by the fixed Q10 model was much higher than estimates by the variable Q10 model. The seasonal course by the fixed Q10 model showed a similar seasonal pattern in air temperature see Figure 4.2a . By contrast, incorporating the seasonal temperature acclimation of leaf R into the model i.e., the variable leaf Q10
model resulted in a relatively flat seasonal pattern of canopy R.
Annual total nighttime canopy R in a 10‐year‐old hinoki cypress stand in 2012 was estimated to be 3.72 and 4.72 Mg C ha 1 year 1, with the variable leaf Q10 and fixed leaf Q10 models, respectively Table 4.1 . Difference in canopy R estimates between the models became greater with the future temperature increase. Specifically, the canopy R in 2012, 2070 RCP2.6 , and 2070 RCP8.5 which were estimated by the fixed Q10 model were 1.27, 1.36, and 1.50 times as large as estimates by the variable Q10 model, respectively.
Responses of the monthly canopy R to the future temperature increase also differed between the models Figure 4.6 . The monthly canopy R predicted by the fixed Q10 model increased in all months as the future temperature increased. This tendency was more evident in summer than in winter.
Conversely, the prediction by the variable Q10 model showed that the extent of future increase in monthly canopy R became smaller as the monthly mean temperature increased. Moreover, during summer July‒September , canopy R appeared not to differ between the present and future climates.
Annual total nighttime canopy R in 2070 based on the RCP2.6 and RCP8.5 scenarios were estimated to be 4.04 and 4.25 Mg C ha 1 year 1, respectively, with the variable Q10 model Table 4.1 . Thus, when annual mean air temperature increases by 2.0 °C and 3.9 °C in the future, the variable Q10
model predicted that the canopy R would increase to 109% and 114% in comparison with the present climate, respectively. On the other hand, according to predictions with the fixed Q10 model, the corresponding future increases in the canopy R were 117% and 135% of the present value, respectively.
Table 4.1 Annual total of nighttime canopy respiration in a 10‐year‐old hinoki cypress stand in 2012 and 2070 based on RCP2.6 and RCP8.5, predicted with the variable leaf Q10 and fixed leaf Q10 models
Future scenario Annual mean air temperature °C
Annual nighttime canopy respiration Mg C ha 1 year 1
Difference
% Variable Q10 model Fixed Q10 model
2012 present 16.7 3.72 4.72 21
2070 RCP2.6 18.7 4.04 109% 5.50 117% 27
Increase 2.0 0.32 0.78 59
2070 RCP8.5 20.6 4.25 114% 6.39 135% 33
Increase 3.9 0.53 1.67 68
Values in parentheses % indicate the relative values in the future compared with the present value Figure 4.6 Seasonal courses in monthly rates of nighttime canopy respiration per unit land area in a 10-year-old hinoki cypress stand at present (2012) and in the future (2070) based on the RCP2.6 and 8.5 scenarios. Predictions were calculated by the variable leaf Q10 model (a) and fixed leaf Q10 model (b). The variable leaf Q10 model considered the seasonal temperature acclimation in leaf R (i.e., Q10 and canopy R20
vary seasonally), as well as vertical variations in leaf R and leaf area. The fixed leaf Q10 model ignored the seasonal acclimation but considered vertical variations.
(a) Variable leaf Q10 model
Month
1 2 3 4 5 6 7 8 9 10 11 12
Nighttime canopy respiration (Mg C ha1 month1 )
0.0 0.2 0.4 0.6 0.8 1.0 1.2
2012 Present 2070 RCP2.6 2070 RCP8.5
(b) Fixed leaf Q10 model
Month
1 2 3 4 5 6 7 8 9 10 11 12
4.3.3 Response of stem maintenance respiration to temperature increases
Equation 4.11 showed the smallest values of the AIC, RMSE, and MAE among three models Equation 4.10‒4.12 and, hence, was selected as the best model with fixed stem Q10 Table 4.2 . These three values of Equation 4.14, in which the seasonally variable stem Q10 was incorporated, became slightly better than those of Equation 4.11. All coefficients in Equations 4.11 and 4.14 were statistically significant Table 4.3 . Thus, the calibrated Equations 4.11 and 4.14 were used as fixed stem Q10 and variable stem Q10 models, respectively, for further simulation.
Monthly rates of stem maintenance respiration per unit land area at present and in the future are presented in Figure 4.7. From June to September, the monthly rates of stem Rm estimated by the variable Q10 model were lower than the rates by the fixed Q10 model. However, the seasonal pattern in monthly stem Rm did not greatly differ between the models compared with the case of canopy R.
Annual totals of stem Rm in a 50‐year‐old hinoki cypress stand in 2012 estimated by the variable Q10 and fixed Q10 models were 1.54 and 1.53 Mg C ha 1 year 1, respectively, which were almost comparable Table 4.4 . However, annual totals of stem Rm in the future that predicted by the fixed Q10
model were slightly larger than predictions by the variable Q10 model. Specifically, the stem Rm in 2070 based on the RCP2.6 and RCP8.5 estimated by the fixed Q10 model were 1.01 and 1.04 times as large as estimates by the variable Q10 model, respectively.
Future annual totals of stem Rm in response to 2.0 °C and 3.9 °C increases in temperature were estimated to be 1.68 and 1.80 Mg C ha 1 year 1, respectively, with the variable Q10 model Table 4.4 . Thus, the variable Q10 model predicted that the stem Rm would increase to 109% and 117% in comparison with the present climate, respectively. The corresponding future increases in the stem Rm
predicted by the fixed Q10 model were 111% and 123% of the present value, respectively.
Table 4.2 Akaike information criterion AIC values of models for estimating the natural logarithm of daily stem CO2 efflux per stem surface area log Es‐d and the root mean square error RMSE and the mean absolute error MAE between observed log Es‐d and values predicted by each model n 609
Model Independent variable AIC RMSE MAE
Equation 4.10 Ts, Di, dD 620 0.40 0.31
Equation 4.11 Ts, log Di 1 , dD 530 0.37 0.29
Equation 4.12 Ts, log Di 1 620 0.40 0.31
Equation 4.14 Q10,stem×Ts/10, log Di 1 , dD 510 0.36 0.29 Abbreviations: Ts, daily mean stem temperature; Di, daily diameter increment; dD, annual diameter increment; Q10,stem, the short‐term temperature sensitivity of stem CO2 efflux
Table 4.3 Results of models for estimating the natural logarithm of daily stem CO2 efflux per stem surface area, log Es‐d
Coefficient Estimate Standard Error t‐value P‐value
Equation 4.11
h 0.054 0.003 21.4 0.001
i 0.359 0.021 17.5 0.001
j 0.119 0.012 10.0 0.001
C 1.990 0.035 56.1 0.001
Equation 4.14
l 0.365 0.016 22.2 0.001
i 0.333 0.021 16.0 0.001
j 0.126 0.012 10.6 0.001
C 1.873 0.038 48.8 0.001
Equations 4.11 and 4.14 were the models with fixed stem Q10 and the variable stem Q10, respectively.
Abbreviations: Ts, daily mean stem temperature; Di, daily diameter increment; dD, annual diameter increment; Q10,stem, the short‐term temperature sensitivity of stem CO2 efflux
Table 4.4 Annual total of stem maintenance respiration in a 50‐year‐old hinoki cypress stand in 2012 and 2070 based on RCP2.6 and RCP8.5, predicted with the variable stem Q10 and fixed stem Q10 models
Future scenario Annual mean air temperature °C
Annual stem maintenance respiration Mg C ha 1 year 1
Difference
% Variable Q10 model Fixed Q10 model
2012 present 16.7 1.54 1.53 1
2070 RCP2.6 18.7 1.68 109% 1.70 111% 1
Increase 2.0 0.14 0.17 18
2070 RCP8.5 20.6 1.80 117% 1.88 123% 4
Increase 3.9 0.26 0.35 26
Values in parentheses % indicate the relative values in the future compared with the present value Figure 4.7 Seasonal courses in monthly rates of stem maintenance respiration per unit land area in a 50-year-old hinoki cypress stand at present (2012) and in the future (2070) based on the RCP2.6 and 8.5 scenarios. Predictions were calculated by the variable stem Q10 model (Equation 4.11) (a) and fixed stem Q10
model (Equation 4.14) (b).
(a)
Variable stem Q10 modelMonth
1 2 3 4 5 6 7 8 9 10 11 12
Stem maintenance respiration (Mg C ha1 month1 )
0.00 0.05 0.10 0.15 0.20 0.25 0.30
2012 Present 2070 RCP2.6 2070 RCP8.5
(b)
Fixed stem Q10 modelMonth
1 2 3 4 5 6 7 8 9 10 11 12