Totomatix: a novel automatic set-up to control diurnal, diel and long-term plant nitrate nutrition

Abstract

Background: Stand-alone nutritional set-ups are useful tools to grow plants at defined nutrient availabilities and to measure nutrient uptake rates continuously, in particular that for nitrate. Their use is essential when the measurements are meant to cover long time periods. These complex systems have, however, important drawbacks, including poor long-term reliability and low precision at high nitrate concentration. This explains why the information dealing with diel dynamics of nitrate uptake rate is scarce and concerns mainly young plants grown at low nitrate concentration.

Scope: The novel system detailed in this paper has been developed to allow versatile use in growth rooms, greenhouses or open fields at nitrate concentrations ranging from a few micro- to several millimoles per litres. The system controls, at set frequencies, the solution nitrate concentration, pH and volumes. Nitrate concentration is measured by spectral deconvolution of UV spectra. The main advantages of the set-up are its low maintenance (weekly basis), an ability to diagnose interference or erroneous analyses and high precision of nitrate concentration measurements (0·025 % at 3 mm). The paper details the precision of diurnal nitrate uptake rate measurements, which reveals sensitivity to solution volume at low nitrate concentration, whereas at high concentration, it is mostly sensitive to the precision of volume estimates.

Conclusions: This novel set-up allows us to measure and characterize the dynamics of plant nitrate nutrition at high temporal resolution (minutes to hours) over long-term experiments (up to 1 year). It is reliable and also offers a novel method to regulate up to seven N treatments by adjusting the daily uptake of test plants relative to controls, in variable environments such as open fields and glasshouses.

Fig. 1.

Overview of the culture system showing (A) a solution tank with peripheral equipment and (B) the ‘Totomatix’ analytical set-up. Key: arrows = flow direction; 1 = solution tank; 2 = cooling coil; 3 = pump; 4a = injection line to plants; 4b = drainage from plants; 5 = manual flow valve; 6a,b = bypass to analytical set-up; 7 = manual flow valve; 8 = charcoal filter; 9 = flow-through cell with pH probe; 10 = filter; 11 = micro-pump; 12 = spectrophotometer with flow-through cuvettes; 13 = acid stock solution; 14 & 18 = motorized syringe drives; 15 & 19 = eight-way motorized selection valves; 16 & 20 = manual three-way valves; 17 = nitrate stock solution; 21 = deionized water inlet; 22 = on/off solenoid valve; 23 = flowmeter; 24 = siphon with level sensor.

Fig. 2.

UV spectra of optically active nutrients (190–300 nm; 2 nm spectral resolution; 1 cm optical path): NO₃ and Fe-EDTA (left axis), molybdate and Cl (right axis), as indicated in the key. Vertical dotted lines denote the wavelengths selected for spectral deconvolution.

Fig. 3.

Time course changes of (A) nitrate concentration C_m, (B) mean spectral error E and (C) transformed error E_t of four nutrient solutions. Note the vertical axis breaks and uneven scales in (A) and (B). Conditions: four tanks of 500 L nutrient solution (pH = 5), each feeding 11 peach trees in a hydroponic orchard; nitrate analyses = 1- h interval; nitrate set points = 1 mm (pink and blue) or 0·025 mm (red and black); optical paths = 0·5 mm (pink and blue) or 10 mm (red and black). Other major nutrients (mm): K = 3, Ca = 3·5, Mg = 1·5, H₂PO₄ = 1, SO₄ = 6−. Micronutrients as in Table 2.

Fig. 4.

Temporal changes in the measured nitrate concentration (C_m, in percentage of set point C_sp) and in the volume of injected stock nitrate (I_N, mL). (A) Sequence of events recorded at 3-h time intervals over a 16-h period; (B) the same C_m data plotted together with intermediate analyses made in the absence of regulation at 20-min intervals over the period 7–10 h. Conditions: 27 plants feeding on 80 L of full solution (C_sp = 0·5 mm NO₃) in a glasshouse, solution temperature = 23 °C, 32 d after sowing.

Fig. 5.

Drift in the measured nitrate concentration after the final harvest of 92 tomato plants grown at C_sp = 3·0 mm and V_sp = 33 L. Conditions: growth room with lights off, nitrate, pH and volume regulations off, air and solution temperature = 20 °C. Regression line: [NO₃] = 2·99987 + 0·00069 × time (R² = 0·952).

Fig. 6.

Effect of the time interval between analyses (min) on the precision of plant nitrate uptake rate (ΔF, μmol h⁻¹) calculated from eqn (15) with n = 92 plants, C_sp = 3 mm (A,B) and 0·1 mm (C,D). In (A) and (C), effect of volume precision ΔV_sp = 0·04 L (dotted line), 0·08 L (thin solid line) and 0·12 L (thick solid line), with V_sp = 66 L. In (B) and (D), effect of total volume V_sp = 33 L (dotted line), 66 L (thin solid line) and 99 L (thick solid line), with ΔV_sp = 0·08 L.

Fig. 7.

Temporal variation of (A) individual plant nitrate uptake rate F and (B) water additions per tank. Conditions: growth room with 12 h photoperiod, 370 µmol m⁻² s⁻¹ photosynthetic photon flux density during day, air and solution temperature = 20 °C, n = 92 plants, V_sp = 30 L, C_sp = 3·0 mm. Other major nutrients (mm): K = 3, Ca = 3·5, Mg = 1·5, H₂PO₄ = 1, SO₄ = 4·5. Micronutrients as in Table 2. In (A), F was calculated either from eqns (2) and (9) (open symbols, dashed line) or from eqns (9) and (17) (closed symbols, solid line). The shaded areas denote the night periods.