Although it is well-known that transpiration is the major driving force for plant nutrient uptake/transport, the coupling between these two processes has not been well-established. Calcium (Ca) is a good candidate to evaluate the coupling between transpiration (water use) and nutrient accumulation as Ca uptake is a passive process without relocation once incorporated, thus avoiding the confounding effects of active regulation and translocation. Here we developed a novel theoretical equation to use soil-to-plant calcium ratio (S_Ca/B_Ca) to predict plant water use efficiency (WUE) derived from δ¹³C. We tested this relationship with two conifer and one angiosperm tree species across their native habitats in China and a controlled greenhouse experiment with soil Ca concentrations manipulated. A linear relationship between WUE and S_Ca/B_Ca was found in all three species across their natural distribution ranges and in the greenhouse experiment. S_Ca/B_Ca was also strongly related to the precipitation of the growing season, and weakly with the mean annual precipitation. Our results suggest a strong linkage between water and Ca, and that S_Ca/B_Ca can be an index for assessing plant WUE. The close relationship between WUE and S_Ca/B_Ca also provides a model system to study the coupling between plant water use and passive nutrient uptake, and the response of this coupling to environmental change.

Field sampling across geographical locations

Soil and tree leaf samples were collected from five provinces in northern and northeastern China across a latitude range of 35°24′-48°35′N and a longitude range of 108°29′-131°18′E in July and early August of 2015 (Fig. 1; Table S1 in Supplementary Information). The entire region is under the strong influence of the East Asian Monsoon with warm and wet summers and cold and dry winters and springs. Dominant soils include Inceptisols, Endisols, Aridisols, and Alfisols. Three tree species including two conifers P. tabuliformis and P. orientalis, and one angiosperm species Q. mongolica were selected for the field study due to their commonness and wide distribution (Table S1). Samples for P. tabuliformis were obtained from natural forests in Liaoning, Hebei, Shanxi, and Shaanxi provinces (Fig. 1, Table S1), for Q. mongolica from Heilongjiang and Liaoning provinces, and for P. orientalis from Liaoning and Hebei provinces.

At least nine 20 x 20 m plots were established for each tree species in their selected natural distribution areas. All tree individuals inside the plot were recorded for diameter at breast height by species, and 2-3 tree individuals that were close to the average diameter at breast height were randomly chosen for taking soil and leaf samples. A composite mineral soil sample was collected with a corer (42 mm in diameter) along four perpendicular directions within 2 m from each chosen tree trunk to a depth of 150 mm. Soil cores in one plot were mixed into one soil sample. Meanwhile, a composite leaf sample was also collected with the same number of current-year and fully-developed leaves along four directions from the lower 1/3 crown for each chosen tree. Eight branchlets were randomly collected for each chosen tree by a ladder and a pole pruner [33].

Climate data acquisition

Monthly precipitation data of all the sampling sites were acquired from the WorldClim global climate and weather database (www.worldclim.org; WorldClim 2.1). Mean annual precipitations of 6 years from 2010 to 2015 (the sampling year) were then calculated and used for analyses. As leaf δ¹³C is mainly related to the environmental conditions of the leaf growing season, the growing season precipitation was also calculated as the total precipitation from April to June (the leaf expansion season before the sampling in 2015). Annual potential evapotranspiration and Aridity index data of all the sampling sites were extracted from the Global Aridity Index and Potential Evapotranspiration Climate Database v2 (www.cgiar-csi.org) . The aridity index was calculated as the ratio of mean annual precipitation and potential evapotranspiration.

Controlled greenhouse experiment

This experiment was carried out in a greenhouse at the Beishan Research Station of the Shenyang Agricultural University in 2018. Three-year-old P. tabuliformis saplings with similar sizes (~ 30-cm-high, 8 mm in basal diameter) were planted in 7.5 L pots (one sapling per pot) with 6 kg sandy soils of low water-extractable Ca concentrations (< 20 mg kg^-1) in early April. After one month of establishment, one litter of Ca-free nutrient solutions (510 mg L^-1 KNO₃, 240 mg L^-1 MgSO₄, 135 mg L^-1 KH₂PO₄, 850 mg L^-1 NaNO₃, 3 mg L^-1FeSO₄, 5.7 mg L^-1 H₃BO₃, 1.5 mg L^-1 MnCl₂, 0.05 mg L^-1 CuSO₄, 0.12 mg L^-1 ZnSO₄, 0.08 mg L^-1 H₂MoO₄) were applied to each pot at the beginning of the experiment. The pH of the nutrient solutions varied between 5 and 6. Then the potted saplings were randomly separated into seven groups (n = 8 for each group) and each group was assigned ramdomly for treatments with different concentration CaCl₂ solutions (0 mg L^-1, 832.5 mg L^-1, 1665 mg L^-1, 3330 mg L^-1, 6660 mg L^-1, 9990 mg L^-1 ,13320 mg L^-1). The plants were watered to 70% of field water capacity by weighing every five days for the non-peak growing season or every three days during the peak growing season. Physiological measurements and leaf tissue sampling were carried out in July and August (see below). At the end of the experiment, sampling height, diameter, and biomass were measured.

Soil water-extractable Ca

To obtain soil water-extractable Ca, soil samples were air-dried, ground with mortar and pestle, and sieved through 0.85 mm (20-mesh size) before 5 g of the sieved soil was extracted with 25 ml distilled water in a rotary shaker for 30 min. A 9.5 ml of the extractant was filtered with a filter paper to a 10 ml Eppendorf tube and spiked with 0.5 ml of 3% SrCl₂ and stored at 4℃ before analysis for Ca in a flame atomic absorption spectrometer (Z-2000, Hitachi, Japan) within 48 h.

Leaf Ca concentration and δ¹³C values

Fresh leaves were briefly dried in an oven at 105℃ for 0.5 h then dried to constant weight at 80℃ before leaf samples were ground with a ball mill to pass through a 100-mesh sieve. A subsample of 0.15 g of the ground leaf materials was then digested with HNO₃-HClO₄. All digestive solution was added with 2 ml 3% SrCl₂ before diluted with deionized water to 50 ml for Ca analysis. Another subsample of 7 mg of the ground leaf materials was analyzed for ¹³C in a mass spectrometer (IsoPrime 100 Isotope Ratio Mass Spectrometer, Germany). Intrinsic water use efficiency (iWUE) was then calculated based on δ¹³C (see Equation [2]).

Photosynthetic gas exchange measurements

Light-saturated leaf photosynthetic rates (A), stomatal conductance (g_s), and transpiration rates (T_r) of the tree saplings were measured with a Li-6400 portable photosynthesis measurement system during sunny days in July and August of 2018 (Li-Cor Inc, Lincoln, NE, USA). Three individual plants per treatment were randomly selected for the measurements. For each individual, three branches were measured to get the average of the individual. Needles were aligned to cover the entire leaf chamber for the measurements. All the measurements were done in the early morning (9:00 to 11:00) on sunny days. The PPFD was set at 1000 μmol m^-2 s^-1 for all the measurements. During the measurements, the CO₂ concentration was around 410 ppm, the leaf temperatures were around 37 ^oC, and the relative humidity was around 31%. Leaf water use efficiency during the measurement was calculated as both A/g_s and A/T_r. Leaf electron transport efficiency of the photosystem II (F_v/F_m) was also detected in the predawn using the Li-6400 portable photosynthesis measurement system for all the treatments.