Estimating photosynthetic radiation use efficiency using incident light and photosynthesis of individual leaves

Abstract

It has been theorized that photosynthetic radiation use efficiency (PhRUE) over the course of a day is constant for leaves throughout a canopy if leaf nitrogen content and photosynthetic properties are adapted to local light so that canopy photosynthesis over a day is optimized. To test this hypothesis, 'daily' photosynthesis of individual leaves of Solanum melongena plants was calculated from instantaneous rates of photosynthesis integrated over the daylight hours. Instantaneous photosynthesis was estimated from the photosynthetic responses to photosynthetically active radiation (PAR) and from the incident PAR measured on individual leaves during clear and overcast days. Plants were grown with either abundant or scarce N fertilization. Both net and gross daily photosynthesis of leaves were linearly related to daily incident PAR exposure of individual leaves, which implies constant PhRUE over a day throughout the canopy. The slope of these relationships (i.e. PhRUE) increased with N fertilization. When the relationship was calculated for hourly instead of daily periods, the regressions were curvilinear, implying that PhRUE changed with time of the day and incident radiation. Thus, linearity (i.e. constant PhRUE) was achieved only when data were integrated over the entire day. Using average PAR in place of instantaneous incident PAR increased the slope of the relationship between daily photosynthesis and incident PAR of individual leaves, and the regression became curvilinear. The slope of the relationship between daily gross photosynthesis and incident PAR of individual leaves increased for an overcast compared with a clear day, but the slope remained constant for net photosynthesis. This suggests that net PhRUE of all leaves (and thus of the whole canopy) may be constant when integrated over a day, not only when the incident PAR changes with depth in the canopy, but also when it varies on the same leaf owing to changes in daily incident PAR above the canopy. The slope of the relationship between daily net photosynthesis and incident PAR was also estimated from the photosynthetic light response curve of a leaf at the top of the canopy and from the incident PAR above the canopy, in place of that measured on individual leaves. The slope (i.e. net PhRUE) calculated in this simple way did not differ statistically from that calculated using data from individual leaves.

Fig. 1. Relationship between net and gross CO₂ assimilation integrated over a day (Daily A_n and Daily A_g, respectively) and daily incident PAR of individual leaves of aubergine, grown with either low (Low N) or high (High N) nitrogen fertilization. For daily A_n, Y = 26·8X – 43, R² = 0·97 for high N; and Y = 22·0X – 13, R² = 0·96 for low N. For daily A_g, Y = 32·2X, R² = 0·96 for high N; Y = 27·3X, <R² = 0·93 for low N. The slope of the regressions increased with N fertilization (for daily A_n, F_1,86 = 23·2; P < 0·001; for daily A_g, F_1,86 = 26·1; P < 0·001).

Fig. 2. Relationship between net and gross CO₂ assimilation integrated over one hour (Hourly A_n and Hourly A_g, respectively) and hourly incident PAR for individual leaves and for a subset of the high nitrogen data shown in Fig. 1. Data for each hour (i.e. 0600–0700...1900–2000 h) are plotted using different symbols. For hourly A_n, Y = –3·2X² + 36·5X – 4·1; R² = 0·96; for hourly A_g, Y = –3·7X² + 40·3X; R² = 0·95. The regressions (not shown for clarity of graphs) had significant quadratic components (for hourly A_n, F_1,151 = 100, P < 0·001; for hourly A_g, F_1,151 = 107, P < 0·001).

Fig. 3. Relationship between net and gross CO₂ assimilation integrated over a day (Daily A_n and Daily A_g, respectively) and daily incident PAR for individual leaves of aubergine grown with low nitrogen fertilization, using incident PAR averaged over the day, rather than the instantaneous PAR, to model assimilation. For daily A_n, Y = –0·51X² + 48·9X – 77·8, R² = 0·99; for daily A_g, Y = –0·56X² + 52·6X, R² = 0·99. The regressions had significant quadratic components (for daily A_n, F_1,42 = 211; P < 0·001; for daily A_g, F_1,42 = 207; P < 0·001).

Fig. 4. Relationship between net and gross CO₂ assimilation integrated over a day (Daily A_n and Daily A_g, respectively) and daily incident PAR of individual leaves of aubergine, grown with high nitrogen fertilization. Data are for an overcast day and the subsequent clear day (PAR sensors were kept on the same leaves for both days). For daily A_n, Y = 29·0X – 58·7, R² = 0·97 for the overcast day; Y = 26·6X – 33·8, R² = 0·98 for the clear day. For daily A_g, Y = 45·9X, R² = 0·99 for the overcast day; Y = 32·2X, R² = 0·96 for the clear day. The slope of the regression increased with N fertilization for daily A_g (F_1,18 = 8·6; P = 0·009) but not for daily A_n (F_1,18 = 0·55; P = 0·47).

Fig. 5. Pattern of above‐canopy incident PAR during the overcast and clear days of Fig. 4 (A) and class frequency distribution of PAR values (B) represented by the number of minutes per day during which PAR was within a given class of values (i.e. 0–100, 100–200 . . . 1600–1700 µmol m^–2 s^–1).

Fig. 6. Class frequency distribution of the PAR incident on each of the six most illuminated leaves among those shown in Fig. 4, during the overcast (closed circles) and clear (open circles) days shown in Fig. 5. Each point represents the number of minutes per day during which the PAR incident on the leaf was within a given class of PAR. The response curve of instantaneous net CO₂ assimilation to incident PAR (A_n curve), as modelled for the same leaf, is also shown. The PAR value corresponding to the point where the dotted line is tangential to the A_n curve (arrows) is that at which the leaf achieves maximum instantaneous net PhRUE. Above and below this PAR value, instantaneous net PhRUE diminishes.

Fig. 7. Relationship between net CO₂ assimilation integrated over a day (Daily A_n) and daily incident PAR for hypothetical leaves (open squares). Daily A_n values for the hypothetical leaves were calculated from the photosynthetic properties of a leaf at the top of the canopy and from the above‐canopy incident PAR on the two or three brightest days for the Low and High N datasets, respectively. Lines are fits, imposing a zero intercept, to the estimates of daily A_n of the hypothetical leaves: Y = 22·0X, R² = 0·99 for Low N; Y = 25·3X, R² = 0·99 for High N. Closed circles are data for actual leaves, as in Fig. 1 (for Daily A_n), which are plotted for comparison. The slopes of the regressions for the hypothetical leaves were not statistically different to those obtained by fitting data for actual leaves as in Fig. 1 (F_1,44 = 0·01, P = 0·92 for Low N; F_1,45 = 1·56, P = 0·22 for High N).