Rate of soil drying and previous water deficits influence the relationship between CO2 assimilation and tree water status in potted lychee (Litchi chinensis Sonn.)

Abstract

The effects of water deficits on CO2 assimilation were investigated in young Tai So (Mauritius) lychee (Litchi chinensis Sonn.) trees growing in pots to test the usefulness of pot experiments for the development of irrigation strategies for this crop. Plants were first grown for various periods without water and measured on a single day. In the other experiments, the rate of drying was varied by growing the plants in sand or in a clay soil. The onset of water deficits was much slower in the clay soil with greater water holding capacity. Some plants were allowed to wilt, rewatered and given a second drying cycle to examine if they had adapted during the previous water deficit. There was a decline in leaf water potential (φL), stomatal conductance (gs) and net CO2 assimilation rate (A) measured at 0900 hours as plants went without water for 0, 5, 15, 19 or 22 d. Net CO2 assimilation rate fell to 18% of the value in well-watered plants when φL declined to ?3.2 MPa. At no time did internal CO2 concentration (C1) decrease with water deficits, and rose when when φL was about ?3.0 MPa or lower. These results suggest a strong reduction in the biochemical activity in the leaves as they dried. Leaf water potential of plants growing in a clay soil with high water holding capacity fell from about ?0.8 MPa in well watered plants to ?4.5 MPa in wilted plants after 17 d without water. Net CO2 assimilation rate fell from 3 to 8 µmol CO2 m?2 s?1 in well watered plants to almost nil as droughted plants wilted. Tree water status was restored within 2 d of rewatering, whereas gas exchange took 10 d to recover fully. In contrast, φL of plants growing in sand with lower water holding capacity fell to ?3.4 MPa after 4 d without water. As tree water status declined, A fell from 4 to 12 µmol CO2 m?2 s?1 in well watered plants to almost nil. Trees were dried to a φL of ?3.0 MPa when they wilted, rewatered and given another drying cycle. The water status of these trees was initially above that of well watered plants reflecting lower gs and CO2 exchange in these plants was always below the rates in well watered plants, even before tree water status started to decline again. In the second drying cycle, A declined to zero as φL fell to ?3.5 MPa and plants wilted. Asymptotic relationships were shown between relative A (A dry/A wet) and φL. The values of φL for relative A of 0.5 and zero were ?1.5 and ?2.5 MPa for the short drying cycle in sand and shifted to ?2.0 and ?3.0 MPa in a subsequent drying cycle, and to ?2.4 and ?3.5 MPa for the long drying cycle in a clay soil. Results collected from potted plants with a rapid onset of drought and no previous water deficits may not necessarily be applicable to field grown trees which experience slow drying and intermittent drought.