Leaf economics traits in wine grapes
Data files
Feb 03, 2025 version files 68.98 KB
-
Martin_et_al._Functional_Ecology._Wine_grape_trait_data.xlsx
65.05 KB
-
README.md
3.93 KB
Abstract
Resource acquisitive plant species are expected to show stronger trait integration vs. resource conservative species, due to simultaneous selection for multiple resource requirements including light, water, and nutrients. While this hypothesis has been invoked to predict interspecific differences in trait variation and integration, it has not been tested to explain intraspecific trait variation (ITV) and trait integration among varieties of crop species.
We quantified nine leaf physiological, water-use, chemical, and morphological traits related to acquisition and use of light, CO2, water, and nutrients, across six varieties of wine grapes (Vitis vinifera L.), in order to quantify the extent of ITV and trait integration among one of the world’s most common and economically important perennial crops. This dataset was also used to test the hypothesis that within a crop species, resource acquisitive varieties express stronger trait integration vs. resource conservative varieties.
All leaf traits varied significantly across wine grape varieties, and formed an intraspecific resource acquisitive-resource conservative axis of variation within wine grapes. Consistent with hypotheses on trait variation and integration, wine grape varieties expressing resource acquisitive trait syndromes were associated with stronger trait integration vs. those expressing resource conservative trait syndromes. Specifically, varieties expressing greater values of light-saturated photosynthesis (Asat), stomatal conductance (gs), maximum carboxylation (Vcmax) and electron transport (Jmax) rates, leaf nitrogen concentrations, and leaf area, expressed a ~45-65% increase in the number of significant bivariate trait correlations compared to resource conservative varieties.
However, within all varieties we detected strong and consistent integration among leaf physiological traits, indicating a mechanistic physiological basis that governs an intraspecific Leaf Economics Spectrum in wine grapes.
Strong trait integration in resource acquisitive wine grape varieties, support the hypothesis that “fast trait” plants have simultaneously been selected to optimize multiple rates of resource uptake, through multiple suites of traits. Our work clarifies the mechanisms by which resource acquisitive species, particularly crops, are able to capture multiple limiting resources to enhance their growth performance. This study also addresses a gap in our knowledge regarding the magnitude of intraspecific variation in trait integration.
README: Wine grape trait data
https://doi.org/10.5061/dryad.9p8cz8wt6
Description of the data and file structure
We quantified nine leaf physiological, water-use, chemical, and morphological traits related to acquisition and use of light, CO2, water, and nutrients, across six varieties of wine grapes (Vitis vinifera L.), in order to quantify the extent of intraspecific trait variation and integration among one of the world’s most common and economically important perennial crops. This dataset was also used to test the hypothesis that within a crop species, resource acquisitive varieties express stronger trait integration vs. resource conservative varieties.
Files and variables
File: Martin_et_al._Functional_Ecology._Wine_grape_trait_data.xlsx
Description: Leaf functional traits of wine grapes
Variables
Dataset column name | Description |
---|---|
unique | unique value identifier |
Variety | wine grape variety identity |
Row | Spatial location of sample (planting row) |
Position | Spatial location of sample (individual vine position) |
Plant | Unique plant identifier |
Vcmax | Maximum carboxylation rates (*μ*mol m-2 s-1) |
Jmax | Maximum electron transport rates rates (*μ*mol m-2 s-1) |
Rd | Leaf dark respiration (*μ*mol CO2 m-2 s-1) |
Vcmax_SE | Standard error of maximum carboxylation rates (*μ*mol m-2 s-1) |
Jmax_SE | Standard error of maximum electron transport rates rates (*μ*mol m-2 s-1) |
Rd_SE | Standard error of leaf dark respiration (*μ*mol CO2 m-2 s-1) |
leaf.area.cm2 | Leaf area (cm2) |
leaf.area.m2 | Leaf area (m2) |
leaf.dry.mass.g | Leaf dry mass (g) |
lma.g.m2 | Leaf mass per area (g m-2) |
leafn | Leaf nitrogen concentrations on a per leaf mass basis (% dry mass) |
leafc | Leaf carbon concentrations on a per leaf mass basis (% dry mass) |
leafn.g.m2 | Leaf nitrogen concentrations on a per leaf area basis (g m-2) |
amax | Light-saturated photosynthesis on a per leaf area basis (*μ*mol CO2 m-2 s-1) |
amass | Light-saturated photosynthesis on a per leaf mass basis (mol CO2 g-1 s-1) |
E | Evapotranspiration rates on a per leaf area basis (mol H2O m-2 s-1) |
gsw | Stomatal conductance (mol H2O m2 s-1) |
Tleaf | Leaf temperature for physiological measurements (degrees C) |
RHcham | Relative humidity during leaf physiological measurements (%) |
VPDleaf | Vapour pressure deficit during leaf physiological measurements (kPA) |
Code/software
R Statistical Software (no specialized packages).
Access information
Other publicly accessible locations of the data:
- TRY Functional Trait Database
Data was derived from the following sources:
- Field-based measurements
Methods
Leaf trait determination
For each individual leaf we used a LI-6800 Portable Photosynthesis System (Licor Bioscience, Lincoln, Nebraska, USA) to execute a 90-point A-Ci curve using the Dynamic Assimilation Technique (DAT) (Saathoff & Welles 2021; McClain & Sharkey 2023). All A-Ci curves and other physiological data collection was conducted while leaves were attached to the plant. During these curves CO2 assimilation rates on a per leaf area basis (Aarea; μmol CO2 m-2 s-1) were logged every 4-seconds across continuously ramping CO2 concentrations, with a ramp rate of 100 μmol mol-1 min-1 (consistent with recommendations by Stinziano et al. 2019; McClain & Sharkey 2023) beginning at 5 μmol mol-1 and concluding at 1700 μmol mol-1. For each A-Ci curve, conditions in the 6 cm2 aperture leaf chamber were otherwise held constant with a photosynthetic photon flux density (PPDF) of 1500 μmol m-2 s-1 of photosynthetically active radiation (PAR; 400-700 nm) (cf. Martin et al. 2022a), relative humidity at 50%, leaf vapour pressure deficits at 1.7 KPa, and leaf temperatures at 25° C. All leaves sampled were allowed a 4-5 minute acclimation period to initial chamber conditions, determined by stable rates of Aarea prior to running the full DAT program. Coupling the acclimation period and total logging time, each DAT-based A-Ci response curve took ~10 minutes to complete. In accordance with guidance by Saathoff and Welles (2021), CO2 and H2O sensor matches were performed using the LI-6800 range match function between measurements of every five leaves. This function records the offsets between the reference vs. sample infrared gas analyzers in the LI-6800 during a reference CO2 ramp. Match offsets are then regressed against sample CO2 concentrations as a polynomial function, and this function is then used to correct for match offsets in real time during the DAT procedure (Saathoff & Welles 2021).
For each leaf we also collected a steady state value of light saturated Aarea (hereafter referred to as Asat) and gs (mol H2O m-2 s-1). This was taken at the same environmental conditions as above but with CO2 concentrations held constant at 420 ppm, and recorded after gas exchange parameters stabilized generally after ~5-10 minutes, according to stability criteria based on the rate of change in Asat and gs. Intrinsic water use efficiency (WUEintr, mol H2O m2 s-1) was calculated as Asat/ gs. All physiological measurements including both Asat measurements and A-Ci curves were taken between 6:00 am and 12:00 pm in order to avoid mid-day stomatal closure.
Following collection of field data, leaves were collected and transported to the University of Toronto Scarborough for leaf morphological and chemical trait evaluation. In the lab, leaves were scanned for leaf area (cm2) using a LI-3100C leaf area meter (Licor Bioscience, Lincoln, Nebraska, USA), dried at 65 °C for 48 hours and weighed; LMA (g m-2) was then calculated as mass/ area. Dried leaves were then ground into a fine powder using a MM400 Retsch ball mill (Retsch Ltd., Hann, Germany), and ~0.1 grams of powdered tissue was analyzed for leaf C and N concentrations using a LECO CN 628 elemental analyzer (LECO Instruments, Ontario, Canada).
Data analysis – A-Ci curve fitting
All statistical analyses were performed using R Statistical Software v. 4.2.0 (R Foundations for Statistical Computing, Vienna, Austria). We used the ‘fitaci’ function in the ‘plantecophys’ R package (Duursma 2015) to fit the Farquhar, von Caemmerer and Berry (FvCB) model to each individual A-Ci curve, and estimate rates of Vcmax and Jmax, along with their standard errors, for each individual leaf. Models were fit using non-linear least square regression following the methods of Duursma (2015), such that Vcmax and Jmax were corrected to 25 °C, and mesophyll conductance was assumed to be infinite such that Vcmax and Jmax are considered apparent. Once these values were extracted from all 180 individual A-Ci curves, they were merged with our additional trait data for further analysis.