Diverging temperature responses of CO2 assimilation and plant development explain the overall effect of temperature on biomass accumulation in wheat leaves and grains

Abstract

There is a growing consensus in the literature that rising temperatures influence the rate of biomass accumulation by shortening the development of plant organs and the whole plant and by altering rates of respiration and photosynthesis. A model describing the net effects of these processes on biomass would be useful, but would need to reconcile reported differences in the effects of night and day temperature on plant productivity. In this study, the working hypothesis was that the temperature responses of CO₂ assimilation and plant development rates were divergent, and that their net effects could explain observed differences in biomass accumulation. In wheat (Triticum aestivum) plants, we followed the temperature responses of photosynthesis, respiration and leaf elongation, and confirmed that their responses diverged. We measured the amount of carbon assimilated per "unit of plant development" in each scenario and compared it to the biomass that accumulated in growing leaves and grains. Our results suggested that, up to a temperature optimum, the rate of any developmental process increased with temperature more rapidly than that of CO₂ assimilation and that this discrepancy, summarised by the CO₂ assimilation rate per unit of plant development, could explain the observed reductions in biomass accumulation in plant organs under high temperatures. The model described the effects of night and day temperature equally well, and offers a simple framework for describing the effects of temperature on plant growth.

Keywords: Biomass; development; grain growth; photosynthesis; respiration; specific leaf area; temperature; thermal time; wheat..

Figure 1

Temperature responses (experiment 1) of leaf elongation rate (LER), daily net photosynthesis (P_N), daily dark respiration (R) and daily net CO₂ assimilation per day (A_N) expressed with time (d) or developmental time units (A_N.20°C, d_20°C). Dots: average values; error bars: confidence intervals (p =0.95); lines: regression from Eq.2. (a) LER (n > 8). (b) P_N (squares), R (triangles) and A_N (circles) (n > 4). (c) LER (black dots) and A_N (white dots) normalised by their respective values at 20 °C. Dashed line displays the temperature response of A_N under saturating light. (d) A_N.20°C.

Figure 2

Leaf elongation rate (LER, (a), net CO₂ assimilation per day (A_N, (b) or day at 20 °C (A_N.20°C, (c), leaf dry mass per area (LMA, averaged for leaves 4, 5, 6 and 7, (d) and the relationship between A_N.20°C and LMA (e) under four different temperature scenarios (T°_day/T°_night, experiment 2). Bars: average values (n = 6); error bars: confidence intervals (p =0.95). Means with the same letter indicate that there were no significant differences in a pairwise t-test.

Figure 3

Time courses of leaf chlorophyll amount (SPAD units) under different temperature regimes (experiment 3), 20/15 °C (blue), 20/20 °C (green), 25/15 °C (red) and 25/20 °C (orange). Time is expressed either as day (d, a) or developmental time (d_20°C, b). Dots: average values (n≥ 4). Error bar: average confidence intervals (p = 0.95). Lines are bilinear regressions with 3 parameters (SPAD₀, t_s, a_s). SPAD₀ is fixed and common to all treatments. Inset in a) Values of t_s. Bars: parameter value ± confidence interval calculated by bootstrap (p = 0.95). Inset in b) Values of t_s.20°C. Bars: parameter value ± confidence interval (p = 0.95).

Figure 4

Time courses of individual grain dry weight (GDW) under different temperature regimes (experiment 3), 20/15 °C (blue), 20/20 °C (green), 25/15 °C (red), 25/20 °C (orange). Time is expressed either as days (a, c) or developmental time (d_20°C, b). M.: grain maturity. Dots: average values (n≥ 4). Error bars: average confidence intervals (P = 0.95). Lines are logistic regressions with 3 parameters (W₀, t₀, λ). W₀ is fixed and common to all treatments. (a) λ and t₀ are free in each treatment. Inset in a) values of t₀± confidence interval (p = 0.95). (b) λ and t₀ are free in each treatment but with time expressed as developmental time (d_20°C). Inset in b) values of t_{0_20°C}±confidence interval (P = 0.95). (c) λ is the only free parameter in each treatment. t₀ (d) is calculated in each treatment from a single t_{0_20°C} value (d_20°C), common to all treatments.

Figure 5

Values of maximum grain growth rate (GGR_max, (a) estimated from regression displayed in Fig.4c (W₀ and t₀ fixed), expressed with time (black bars) or developmental time units (white bars), and the relationship between net CO₂ assimilation per d_20°C (A_N.20°C) and final individual grain weight or GGR_max.20°C in the 4 different temperature scenarios (b). (a) Bars: estimated parameter value. Error bar: confidence interval (P = 0.95). (b) Grey triangles: final grain weight. White circles: GGR_max.20°C. A_N.20°C values were measured in Experiment 2 and shown in Fig. 2 [see Supporting Information—Table S2].

Figure 6

Observed values vs. calculated values for the reduction in final grain weight between temperatures treatments. Observed data come from the literature [see Supporting Information—Table S1]. Dashed line is the model x = y.