https://orcid.org/0000-0003-4404-2563
Figures
Abstract
Yield reliability under diverse environments is important to address climate stress consequences in wheat production systems. Breeding for reliability under a changing climate remains a challenge in wheat. We assessed the performance of 18 hexaploid (Triticum aestivum L.) genotypes in three field environments at a location within the semi-arid Canadian Prairies over four years with a primary aim to establish knowledge of the phenotypic plasticity and yield reliability in the parental lines as it relates to heat and drought stress tolerance. We collected data on various physiological traits along with some agronomic and morphological attributes, uncovering significant variation across early seeded rainfed, early seeded irrigated, and late seeded rainfed (hot and dry) environments. Eight high yielding hexaploid genotypes ‘01S0263-28’, ‘AC Foremost’, AC Karma’, ‘Cutler’, ‘MN03358-4’, ‘Reeder’, ‘Stettler’, and ‘Superb’ showed higher grain Δ13C. Six of these genotypes ‘01S0263-28’, ‘AC Foremost’, ‘MN03358-4’, ‘Reeder’, ‘Stettler’, and ‘Superb’ showed higher water use efficiency under irrigated as well as hot and dry environment compared to the low yielding lines ‘Red Fife’,’8021-V2’ and ‘BW278’. Only four genotypes ‘01S0263-28’, ‘MN03358-4’, ‘Reeder’, and ‘Stettler’ were found with higher yield reliability index. The grain yield relationship with leaf rolling, glaucousness, and canopy temperature was found to be weak. The flag leaf stomatal numbers increased with water stress in high yielding lines which were otherwise low in stomatal numbers. Contrastingly, water stress significantly reduced the stomatal numbers in low yielding lines that were otherwise high in stomatal numbers. The results highlight the stomatal adaptability of different genotypes in response to drought. Taken together, these results provide baseline information that the genotypes with high grain Δ13C and WUE, and low stomata numbers are more yield reliable under variable field environments, and this information can guide the breeding of climate-resilient germplasm that expresses consistent and reliable grain yield production in the semi-arid Prairies.
Citation: Singh Sangha J, Wang W, Knox R, Ruan Y, Cuthbert RD, Isidro-Sánchez J, et al. (2025) Phenotypic plasticity of bread wheat contributes to yield reliability under heat and drought stress. PLoS ONE 20(3): e0312122. https://doi.org/10.1371/journal.pone.0312122
Editor: Xi Liang, University of Idaho, UNITED STATES OF AMERICA
Received: January 31, 2024; Accepted: October 1, 2024; Published: March 10, 2025
Copyright: © 2025 Sangha et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All relevant data are within the paper and its Supporting Information files.
Funding: This study was financially supported by the ‘Agriculture and Agri-Food Canada’, ‘National Research Council’, and ‘Canadian Wheat Alliance’. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.
Competing interests: The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.
Introduction
Wheat is a staple food for nearly 35% of world’s population. It is grown widely in North America, contributing approximately 10% of annual global wheat production and more than a third of the world’s export supply [1]. The demand for bread wheat (Triticum aestivum L.) is growing due to changing food habits and increasing global population [2]. Feeding the world thus needs a consistent increase in wheat yield over the coming decades [3].
Improving grain yield in wheat is a challenge under current climate patterns requiring integrating new technologies and traits in breeding [4–6]. The current rate of genetic gain and varietal development effort for wheat are not sufficient to tackle the challenges of global wheat production under severe heat and drought stress [1,2,7]. There is great variability in wheat yield due to genotype by environment interaction, an important consideration in varietal development [8]. With further destruction and conversion of forest lands for agriculture crop use being a poor option, increasing wheat production must rely on exploiting physiological plasticity in plants and identifying and incorporating new sources of tolerance or resilience to weather extremes [9,10]. Screening and selection within diverse germplasm based on heritable physiological traits has been useful in developing wheat lines for regions with water scarcity [11–13]. Such germplasm is a source for improving yield with less variability in diverse environments, with better adaptability and resilience to heat and drought stress.
Hexaploid wheat, one of the important field crops on the Canadian Prairies, is commonly grown under semiarid conditions representing more than 90% of bread wheat production in Canada [14]. With a relatively small proportion of acreage under irrigation, most wheat acreage is totally dependent on snow melt that enters the soil and on subsequent rains for growth and development on the Prairies, therefore, climate change induced heat and drought stress limits wheat productivity in the region [15,16]. Attaining the full potential of wheat grain yield requires further improvements in breeding for stress tolerance, adaptability, and resilience [17,18]. Developing wheat varieties with climate change resilience, especially for heat and drought stress is therefore imperative to global food security [19].
Heat and drought events could reduce up to 50% of the expected grain production [18,20]. The impact is more intense if these conditions occur during the grain fill period [21]. Heat and drought are realized in western Canada with huge impact on crops. During the past 100 years, at least 40 droughts have occurred in Canada and those observed in the 1930s, 1980s, early 21st century, and recently in 2021 left a huge impact on the wheat sector [22,23]. Drought in 2021 was more intense than previous drought events and felt throughout the Prairies [23] pressing the need to develop heat and drought tolerant cultivars with less damage potential under such conditions. Many spring wheat cultivars in North America have been identified as sensitive to heat and drought with a reduction in yield of up to 7.5% for every 1°C increase in temperature [1]. One option for addressing this issue is to have cultivars that maintain grain yield reliability under diverse growing conditions, which limits downward yield fluctuations to address yield losses under unpredictable environments [24].
Selection of wheat genotypes in diverse environments using a combination of agronomic, morpho-physiological, and quality traits could help germplasm development for reliable yield performance under heat and drought. Various traits such as stomatal density, stomatal conductance, canopy temperature, and stable isotope signatures especially under water stress are representative of drought stress response in plants [12,25–27]. Some of these traits such as Δ13C are important indicators of photosynthetic activity, WUE, and high grain yield in wheat [26,27]. These specific physiological traits also relate well with molecular and biochemical attributes of the plant making them more relevant candidate traits for genetic association compared to the broader traits of drought and heat tolerance [28]. Inclusion of these physiological parameters with selectable and heritable features in breeding strategies is thus important for abiotic stress adaptation [26,29,30].
We studied a panel of diverse hexaploid wheat genotypes, including the founder line ‘Red Fife’ and a few popular wheat cultivars in Canada (Table 1). These wheat genotypes carry a diversity of traits that have been used in developing mapping populations and identifying trait marker associations [31]. Except for a few reports, not much is known regarding the variation in morpho-physiological traits in the majority of these genotypes [32–35]. Using a combination of measurable traits in this field study, we investigated drought stress resilience and yield reliability in wheat under the semi-arid climate of the Canadian Prairies. The objectives are to i) determine the grain yield performance of a diverse panel of wheat genotypes that can lead to establishing knowledge of the genetic basis of heat and drought stress tolerance in Canadian wheat; ii) characterize the phenotypic plasticity of the wheat panel for physiological traits that potentially associate with yield reliability under variable environmental conditions, and iii) improve our understanding of the morpho-physiological heat and drought stress traits that are best to apply to wheat breeding.
Materials and methods
Plant material and field trials
We studied a panel of 18 hexaploid (Triticum aestivum L.) genotypes (Table 1) which are mostly bread wheats. These genotypes have been used in crossing within Agriculture and Agri-Food Canada’s (AAFC) breeding programs and are parents of genetic populations used to study traits such as disease resistance and abiotic stress tolerance. This panel also consists of three wheat lines from the USA (Reeder, MN03358-4 and 01S0263-28) and historical wheat lines that were used in decades past in Canadian breeding programs, including the founder line ‘Red Fife’. All the experiments were conducted at the research facilities of AAFC-Swift Current, SK. Therefore, no permits were required to use the facilities and complete this work. The research work was approved by the department.
In a multiyear field study from 2012 to 2014 and 2016, we seeded three replicated plots in three field environments on a Swinton loam soil (Orthic Brown Chernozem) near Swift Current, SK Canada using a randomized complete block design. The field environments were created as early rainfed (seeded early in May without irrigation, ES-RF), early irrigated (seeded in early may with irrigation, ES-IRR), and late rainfed (seeded in late May without irrigation, LS-RF). Each genotype was seeded in a plot of 3 m length and 1.2 m width having four planting rows, with 0.23 m between rows. The gap between plots was filled by seeding two rows of winter wheat. Each plot was sown at a rate of 1000 seeds per plot (adjusted for percentage of germination) with standard field practices.
Phenotypic data collection
Data collected included measurements or documentation of traits associated with agronomy and morpho-physiology of the plants of each wheat line/cultivar to differentiate genotypic characteristics and responses under irrigated and rainfed environments.
Agronomic traits.
For agronomic data, we collected harvested grain yield (GY, kg ha-1), kernels per spike (KPS), test weight (TW), thousand kernel weight (TKW, g), days to anthesis (DTA), days to maturity (DTM), and plant height (PH, cm) and percent grain protein concentration (GPC, %). Harvesting was conducted using a Wintersteiger Elite plot combine (Wintersteiger AG, Salt Lake City, UT, USA) at 18% plant moisture on a wet weight basis. Grain samples were dried to about 12% moisture before weighing plot yield, expressed in kg ha-1. The number of KPS were counted using five spikes per genotype collected from the field environment. The TW was measured using methods described by the Canadian Grain Commission that uses volume of grain expressed in kilograms per hectolitre (kg/hL) (https://grainscanada.gc.ca/en/grain-quality/) and TKW was determined by weighing 1,000 seeds in grams. The number of DTA was recorded when 50% spikes showed anthers protruding in a plot whereas the number of DTM was recorded when > 75% of the spikes in a plot showed total loss of green color. The PH was measured using a meter stick, recording an average of three plants per plot. For grain quality, we measured GPC by scanning whole grain with a visual near infra-red spectrometer (FOSS NIRSystems 6500) at 13.5% moisture basis.
Morpho-physiological traits.
Under morpho-physiological traits, we collected canopy temperature (CT), carbon isotope discrimination (Δ13C), water use efficiency (WUE), glaucousness (GLA), leaf rolling (LR), and flag leaf stomatal density converted to stomatal numbers (SN) per flag leaf.
Canopy temperature of plants was measured using an infrared thermal imaging camera (FLIR T620) between 11:00 a.m. and 1:00 p.m., at late grain-filling, by positioning the infrared sensor at a consistent distance above the canopy and angled at 45° from the plot’s north side. We then used the temperature images produced by the image processing software (FLIR Research) to compute the average temperature for the entire plot canopy.
We used finely ground grain from the wheat genotypes harvested in 2012 and 2013 for carbon isotope analysis at the isotope facility of AAFC’s Research Centre in Lethbridge AB, utilizing the Finnigan Delta V isotope ratio mass spectrometer (Thermo Electron, Bremen, Germany). The stable isotope analyses used 0.3 mg of sample flour and reference material weighed into silver capsules. The process involved calculating the amount of 13C isotope in relation to the international Vienna-Pee Dee Belemnite (V-PDB) standard [36]. We calculated Δ13C as described in the handbook Physiological Breeding II [37] as follows:
Where δa is the stable carbon isotope composition of the atmosphere and δp refers to the stable carbon isotope composition of the plant sample.
Using soil moisture, precipitation, and grain yield data for the year 2012, we calculated the WUE for all wheat genotypes. We determined the soil water content up to a depth of 100 cm in different plots both at the onset of seeding and during crop harvest using a PR2 Soil Moisture Profile Probe (Delta T Devices, UK). Throughout the growing season, we conducted moisture readings at soil depths of 10, 20, 30, 40, 60, and 100 cm. For each measurement instance, we recorded three readings per plot. Using the grain yield data from 2012, we determined WUE ((kg ha-1mm-1) with the following equation:
We used methods described in the handbook Physiological Breeding II [37] to collect GLA notes in the early morning during the grain-filling period using the scale from 0 (none) to 10 (total cover of white waxy bloom on the leaves and stem). For LR, we rated plants during the grain-filling period in the early morning (before 10:00 a.m.), using a scale from 0 to 3. The extent of leaf rolling of the most recent fully expanded leaf was estimated as 0 = no leaf rolling, 1 = leaf loosely rolled from the tip (<33%), 2 = leaf moderately rolled (34-66%), and 3 = leaf tightly rolled (>67%).
We analyzed total number of stomata per flag leaf using stomatal density (mm2) from each wheat genotype grown in water stress (WS, drought) and non-stressed (WW, well-watered) conditions in a glasshouse experiment conducted from March 2023 to June 2023. Plants were grown in 1 L plastic pots filled with a mixture of field and peat soil. The greenhouse temperature was set at 15˚C for the initial 30 d after seeding and thereafter at 22˚C with 16 h of light and 18˚C with 8 h of dark. Once the flag leaf fully expanded, we took impressions from both the abaxial and adaxial surfaces using clear nail polish in the dry impression method described by Ferris et al. 2002 [38]. Using a stereo microscope attached to a digital camera (Optica, Italy), we captured images of the impressions at 10x magnification, ensuring a consistent field of view and manually recorded the number of stomata within 1 mm2 field of view (number of sample counts = 9). We determined the stomatal numbers per flag leaf area by multiplying the average stomatal density in mm2 by the average leaf area (LA), where LA = [flag leaf length (mm) × flag leaf width (mm) × 0.75].
Data analysis
We analysed the data with Linear mixed model in R software (version 4.3.1) for variance components, with genotype effect considered as a fixed factor and environmental effect along with its interaction with genotype effect as random factors. Least square means were estimated in each experiment for each genotype for each trait using R software (version 4.3.1). Unless otherwise indicated, results were considered statistically significant if p < 0.05 or highly significant at p < 0.01. Figs in the manuscript were constructed with MS-Excel, JMP version 17.2 and SigmaPlot version 14.0 (Systat Software, San Jose, CA, USA). Using pairwise correlation analysis, we assessed the relationships between various traits within each environment and year, then visualized these correlations in a scatterplot matrix. Biplots of principal component analysis (JMP version 17.2) were performed to explain the variation influenced by different factors.
We computed broad sense heritability using the R package “inti”, following the equation provided below: https://cran.r-project.org/web/packages/inti/vignettes/heritability.html
where σ2P is the phenotypic variance component and σ2g is the genotypic variance component.
where ng is the number of samples, ne is the number of environments, nr is the number of replications.
For estimating phenotypic coefficient of variance (PCV, %) and genotypic coefficient of variance (GCV,%) we used the formulas:
where x is the sample mean.
We calculated yield reliability index (YRI) for grain yield in three field environments based on the method of Kataoka 1963 [39] as described in Doring et al. 2015 [24]:
where mGY is the mean grain yield, Sev is square root of the environmental variance, Z(P) is the percentile from the standard normal distribution for which the cumulative distribution function reaches the value P. Z(P) is 0.675 for P = 0.75; 0.840 for P = 0.80; 1.040 for P = 0.85; 1.280 for P = 0.90; and 1.645 for P = 0.95. For this study, we used Z(P) as 1.645 with the P value set as 0.95. A genotype with high YRI is considered more reliable for grain yield.
Results
Weather conditions during field trials
The long term historical (1980-2022), and trial years (2012, 2013, 2014, and 2016) precipitation and temperature data for different growing months at Swift Current are given in Fig 1. The rainfall in June for all four years was higher than the average of the last four decades. Late season drought occurred in 2012 and 2013 during the grain fill period of July and August with less precipitation than the long-term average. Less precipitation was received in July 2014 and August 2016 compared to the long-term average. A cooler growing season, favourable for vegetative growth, was experienced in 2013 with the highest temperature not exceeding 30oC.
Monthly average precipitation for the growing season of each year of the study is conveyed by colored bars with the scale in mm on the right-hand axis. The long-term average (1980-2022) precipitation is provided by the blue dotted line. The average temperature for the long term is provided by the solid black line, with the grey vertical streaks in the background indicating average daily extremes. Average daily temperature for each month for the years of study (2012-2014, 2016) is provided by solid-colored lines. The temperature scale is provided on the left-hand axis.
Phenotypic and genotypic variation in wheat traits
Data from various agronomic and morpho-physiological traits studied with hexaploid wheat lines showed considerable phenotypic and genotypic variation in the test lines. The mean squares for 13 of the 14 traits representing agronomic and morpho-physiological parameters studied in 18 hexaploid genotypes are presented in Table 2 whereas the estimates of σ2g, σ2p, PCV%, GCV%, and H2 for the data are given in Table 3. Genotype values for different agronomic and agronomic traits are given in Table 4.
Grain yield and other agronomic traits.
Analysis of the agronomic traits showed significant differences (p < 0.01) among the genotypes (G), and environments (field environments and years, E). The difference for G x E interaction was significant for all these traits (Table 2). Data on variance estimates showed that PCV estimates ranged from 1.7 for DTM to 13.6 for KPS and GCV estimates varied from 1.0 for DTM (LS-RFF) to 12.8 for KPS (LS-RF) in different field environments (Table 3). The highest σ2p and σ2g were found in GY and lowest were found in DTM and in DTA. Both σ2p and PCV estimates were higher than the σ2g and GCV of the agronomic traits, except for PH that showed the least difference in genotypic and phenotypic variance in all three environments. We observed low to high estimates of heritability for the agronomic traits. The H2 was the highest for PH in all field environments, whereas it was lowest for DTM (36%) followed by DTA (40%) under LS-RF environments.
With genotypes averaged over years, GY ranged from a minimum of 2818 kg/ha to a maximum of 4988 kg/ha under these environments (Table 3). The GY was highest under the ES-IRR field environment followed by ES-RF and lowest under LS-RF. Genotypic variation shows that the lowest GY was observed for ‘BW278’ in all three field environments, whereas the highest overall GY was recorded for ‘AC Karma’ under two field environments (ES-IRR and LS-RF). For the ES-RF field environment, GY was highest in the genotype ‘01S0263-28’ (Table 4). Similarly, smaller but significant variations were observed under these environments for other agronomic traits, KPS, TW, TKW, DTA, DTM, and PH. KPS was lowest in the LS-RF environment and TW was highest in LS-RF. Lower TKW was observed in both rainfed environments. Similarly, DTA was shorter in both rainfed environments, whereas DTM was shorter in the LS-RF environment and longer in the ES-IRR environment. The grain end use quality factor, GPC also increased significantly under the irrigated field environment, compared to the other two environments (ES-RF and LS-RF) ranging from a minimum of 11.3% to a maximum of 15.3% across field environments. Among genotypes, ‘AC Karma’ had the lowest GPC in all three field environments whereas ‘Lillian’ produced the highest GPC in all three environments (Table 4).
Morpho-physiological traits.
Analysis of the morpho-physiological traits in the wheat panel showed considerable variation among the genotypes and environments (Table 2).
Glaucousness and leaf rolling.
Both GLA and LR showed significant differences (p < 0.01) during field evaluation and showed genotypic and phenotypic variation (Table 2, Table 3). Both traits were significantly different for GxE interaction. The PCV and GCV estimates were higher for GLA and LR under all different field environments than other physiological traits (Table 3). The σ2p and σ2g values were low for both GLA and LR but σ2p estimates were higher than the σ2g and PCV estimates were higher than the GCV. The H2 was moderate to high for GLA. For LR, the trait was weakly expressed (17%) under ES-IRR conditions and among the lowest heritabilities for all the morpho-physiological traits studied in hexaploid wheat. The H2 for LR was moderate to moderately high for the other two environments.
GLA was expressed more in plants grown in LS-RF and ES-RF conditions compared to ES-IRR conditions (Table 3). However, the expression of GLA in wheat genotypes was not higher than three on a scale of zero to ten. During genotypic comparison, highest GLA was expressed in ‘AC Foremost’ in the LS-RF field environment, whereas the least expression was observed with ‘Vesper’ under irrigated conditions (Table 4). Similarly, genotypic variation was also observed for LR that was more evident in the LS-RF field environment with the highest LR recorded for Cutler (2.3). Although LS-RF produced the greatest range among cultivars for LR, ES-RF had the highest mean for LR. LR was least expressed in ES-IRR that affected parameters of variance estimates (Table 3).
Carbon isotope discrimination.
Genotypes differed significantly for grain Δ13C (p < 0.01), as did field environments (p < 0.01) and G x E interaction (p < 0.01) (Table 2). The variance components, σ2p, σ2g, PCV and GCV, were low for Δ13C in all three environments, although the σ2p were higher than σ2g and PCV estimates were higher than GCV. The heritability estimates were high in all three field environments.
When compared across field environments, grain Δ13C increased (p < 0.01) with decreasing water stress exhibited by lower mean values in the hotter later seeded rainfed environment (LS-RF) and early seeded rainfed (ES-RF) environments, compared to the cooler field environment with lower water stress (ES-IRR) (Table 3). When comparing genotypic variation under each environment, the lowest Δ13C values were observed with ‘Red Fife’ in the ES-RF (15.4) and ES-IRR (17.2) field environments, however, under the LS-RF environment ‘AC Cadillac’ had the lowest Δ13C (14.2) (Table 4). The genotype ‘Cutler’ was found with relatively high grain Δ13C in both rainfed environments (ES-RF = 18.1; LS-RF = 16.4), whereas the genotype ‘AC Foremost’ had the highest Δ13C (18.9) under irrigation (LS-IRR).
Canopy temperature.
Among all the morpho-physiological traits, we did not find significant genotypic differences (p > 0.05) in CT recorded at the grain fill stage in all three field environments, however, environment variation was significant at p < 0.01 level (Table 2). The highest CT was observed in ES-RF (28.1˚C) and lowest in LS-RF (25.4˚C) (Table 3). Variance estimates showed that σ2p, σ2g, PCV and GCV were the lowest for CT in all three environments and not reliable for explaining H2 differences in CT estimates. The calculated H2 for CT was also low under all three environments and no H2 was calculated for ES-IRR. The analysis of genotypes for CT differences showed that ‘01S0263-28’, ‘AC Karma’ and ‘MN03358-4’ were highest in CT (28.1) under the ES-RF environment, whereas the genotype ‘Superb’ recorded the coolest CT (25.4) under LS_RF environment (Table 4).
Water use efficiency (WUE).
Based on soil moisture data, crop season precipitation, and GY collected in 2012, we calculated WUE for all the hexaploid wheat genotypes grown in three field environments. The WUE of low yielding founder line ‘Red Fife’ was used as a reference for comparing the genotypes (Fig 2). Water use efficiency was higher in most high yielding modern hexaploid genotypes than the founder line, ‘Red Fife’ in all three field environments. Lines ‘01S0263-28’, ‘AC Foremost’, ‘MN03358-4’, ‘Reeder’, ‘Stettler’, and ‘Superb’ were consistently higher in WUE (p < 0.05) than ‘Red Fife’. However, WUE of ‘8021-V2’, ‘AC Barrie’, ‘BW278’, ‘AC Cadillac’, ‘AC Domain’, and ‘Lillian’ were found to be not significantly different (p < 0.05) from ‘Red Fife’ in any of the three field environments. Genotypes ‘Vesper’ and ‘RL4452’ showed significantly different WUE under the LS-RF environment, whereas ‘AC Karma’, ‘Carberry’, and ‘Cutler’ showed significantly higher WUE than ‘Red Fife’ under the ES-IRR environment.
ES-RF = early seeded rainfed, ES-IRR = early seeded irrigated, and LS-RF=late seeded rainfed. The results are based on three replicates for each environment (Mean ± SD).
Stomatal numbers on flag leaf.
The stomatal numbers (SN) combined for both sides on the flag leaf varied significantly (p < 0.05) among genotypes grown under well-watered (WW) and water stress (WS) treatments. On average, under WW conditions, 402,724 stomata were observed on the entire flag leaf, whereas this number increased to 426,159 under drought stress (p < 0.05) (Fig 3a). Generally, the stomatal numbers consistently increased in high yielding lines, whereas it was found to decrease in low yielding lines, ‘Red Fife’, ‘RL4452’, ‘8021-V2’, and ‘Lillian’ (Fig 3b). Genotypic variation of stomatal numbers showed that the lower SN under WW conditions was found with ‘AC Barrie’ (296,745) and the highest number recorded was for ‘Red Fife’ (620,937). In the case of WS, the lowest stomata number was recorded for ‘AC Barrie’ (348,944) and highest number was found with ‘RL4452’ (509,999). When compared to other genotypes, eight genotypes showed non-significant change in stomatal numbers with water stress on both sides of the flag leaf.
a) Average stomatal numbers of hexaploid flag leaf; b) genotypic variation in stomatal numbers combined for both surfaces of the flag leaf; c) percent change in stomatal numbers in well-watered and water stress conditions.
‘Stettler’ showed the least change in SN with change in water regime (WW = 370,832 to WS = 352,836). The % change (decrease) in SN was higher in ‘Red Fife’, followed by ‘RL4452’, whereas the % change was lower in ‘Lillian’ and ‘8021-V2’ (Fig 3c). A similar trend for the change in SN was observed on both the adaxial and abaxial leaf surfaces (S1 Fig). On average, the adaxial stomata increased from 229,464 under the well watered condition to 242,095 with water stress, whereas abaxial stomata increased from 173,260 to 184,065.
As with stomatal numbers, the flag leaf area of wheat genotypes varied with water availability (S2 Fig). The average leaf area increased from 29.9 cm2 under WS conditions to 30.4 cm2 with WW. Under WW conditions, the greatest flag leaf area was observed in ‘RL 4452’ (42.20 cm2), whereas smallest leaf area was recorded with ‘AC Domain’ (23.3 cm2). When water stress was applied, the greatest leaf area was observed for the genotype RL4452 (33.9 cm2) and smallest for Carberry (25.9 cm2).
Yield reliability index
Wheat genotypes with higher reliability index tend to be those with a higher grain yield and with low to moderate variance (S2) for the yield estimates across different environments. Four high yield hexaploid wheat genotypes such as ‘Reeder’, ‘Stettler’, ‘01S0263-28’, and ‘MN03358-4’ with low to moderate mean variance were found with comparatively high reliability index than other genotypes, with a YRI ranging from 2833 to 3147 (Table 5). The founder genotype ‘Red Fife’ (YRI = 2001) followed by ‘RL4452’ (YRI = 2106) were found to be among the least reliable hexaploid genotypes for grain yield among environments during this study. One line ‘BW278’ with low variance and low yield did not show high YRI. Another line, ‘AC Foremost’ is an example of a relatively modern high yielding genotype with a low reliability index due to a very high yield variance. Similarly, AC Karma showed comparatively high variance than other genotypes.
Relationship between various traits
Pairwise correlation analysis shows that many traits exhibited moderate to high correlations with grain yield of hexaploid wheat lines grown in three field environments. Among all genotypes grown in the ES-RF environment, a positive correlation of GY was observed with Δ13C, GLA, CT, TKW, KPS and WUE, whereas a negative correlation was observed with PH, GPC, DTM, and LR (Fig 4). The correlations of GY were very low (<-0.2) with DTA and TW. Under ES-IRR conditions, a similar trend was observed, except for a low negative correlation of GY with CT, TW, and LR. Under the LS-RF environment, the GY was found to have a high positive correlation with Δ13C and LR and a moderate to low positive correlation with CT, TKW, and WUE. GY was negatively correlated with GPC, PH, and DTM. The correlations of GY were very low with GLA, TW, DTA, and KPS. The Principal Component Analysis biplot of the hexaploid wheat traits shows that the first two principal components of the biplot reflected about 60% of the variance under ES-RF, 60% of the variance under ES-IRR, and 53% of the variance under LS-RF (Fig 4).
Discussion
This study helped to determine the variability in grain yield performance to establish knowledge of the genetic basis of heat and drought stress tolerance in Canadian wheat, and characterize the phenotypic plasticity of wheat to apply the morpho-physiological traits in breeding for yield reliability under variable environmental conditions. From the field evaluation of a panel of 18 diverse hexaploid wheat genotypes including some newer and older cultivars, and advanced breeding lines, we found extensive evidence of phenotypic plasticity in different wheat lines. This phenotypic plasticity was evident from the variation observed in some agronomic and morpho-physiological traits in wheat lines grown under different environments. Such variability in germplasm is essential for genetic enrichment in yield related traits in cultivars for diverse field environments as it provides understanding of the traits required for the selection of climate resilient genotypes [6,19]. This information on phenotypic and genetic variability of traits is potentially useful for improving the yield reliability of Canadian wheat. Our study is specifically relevant to the rainfed Prairies that supports over 90% of the wheat production in Canada, which has experienced frequent drought events in recent years [23].
Phenotypic and genetic variability in wheat
Grain yield and related traits.
The difference of 40% (2,170 kgs/ha) in harvested grain within 18 wheat lines grown in three field environments indicates the presence of diverse genetics among bread wheat lines for grain yield. In fact, all the agronomic traits except CT studied during four years presented significant variability among genotypes in different environments. The variance components, PCV and GCV that ranged from low (<10%) to high ( > 20%) estimates for the grain yield and other agronomic traits confirms that moderate phenotypic and/or genotypic variability exist in this wheat panel. Variation in agronomic traits is a key for cultivar improvement. In wheat, phenotypic and genetic variation for agronomic traits have been shown with a wide range of diversity [40,41]. Whereas the lines with high genetic variability in traits are considered more relevant for use in breeding, those with low genetic variability are still important in phenotypic plasticity [42]. These lines with low genetic variation for the traits are therefore useful to make contrasts for genetic assessment of various plant traits.
The PCV estimates for the agronomic traits were slightly higher than GCV indicating that there was environmental influence on these traits, as evident from the statistically significant GxE interaction. Since the difference between PCV and GCV was small, the magnitude of environmental influence was not high and most of the variation was due to existing genetic variation [43]. However, our study was already a multiyear environmental assessment and the results of the variability were based on consideration of GxE interaction in the model. Therefore, the results provide great potential for genetic assessment of yield related traits under tested field environments. Since response to selection for a quantitative trait is directly proportional to the function of its heritability and its genetic variance [44], the variance components observed in this study for the test material will enable wheat breeders to apply selections based on available genetic differences in the test lines. Low estimates of broad sense heritability indicate that genetic variance is low compared to the phenotypic variance. We found moderate to higher heritability estimates [45] in our results from 12 site years for the majority of the agronomic traits, which suggests that obtaining genetic gain for most of these traits through selection in different environments is possible [46].
Morpho-physiological traits
Our study shows a wide range in the estimates of variance components for the morpho-physiological traits when compared to agronomic traits that we studied, which provides the baseline for the presence of substantial genetic variability in the wheat panel for morpho-physiological evaluation [43,44]. The estimates for PCV and GCV were higher for GLA and LR, but the heritabilities were lower, especially under the ES-IRR environment. The difference between PCV and GCV for GLA and LR were also higher suggesting that these two traits were highly influenced by the environments [44]. In contrast, the PCV and GCV estimates for other morpho-physiological traits were low and the difference between the two variance components was small suggesting lesser impact of GxE interaction on these traits, especially in Δ13C, WUE, and leaf SN.
The high broad sense heritability estimates for various morpho-physiological traits in the wheat panel that we analysed in this study also suggest a low genotype-by-environment interaction for many of the morpho-physiological traits, which is also in agreement with reports from others [45]. Since a high heritability for the plant traits is an indication of lower environmental influence, the selection based on phenotypic expression of individual accessions for these traits may be easy due to relatively small contribution of the environment to the expression of the phenotype where most of the variation is rather genetically influenced [46]. The high heritability estimates of the morpho-physiological traits in different environments including under the hot and dry environments we tested suggest that any progress made in selection of these traits in wheat can be of high breeding value for genetic advancement of water stress tolerance [47], particularly for germplasm development for yield improvement at the location of our study. These results improve our understanding of physiological breeding that will further improve grain yield reliability in wheat.
Yield reliability in different environments
Yield reliability index helps to predict the minimum average yield for a genotype grown in diverse environments that a grower can expect from his crop [24,39,48]. In fact, reliability in yield is highly relevant to address yield fluctuations and losses under unpredictable environments [24]. The genotypes ‘Reeder’, ‘Stettler’, ‘01S0263-28’, and ‘MN03358-4’ were found with high YRI based on low to moderate variance estimates and high grain yield in different field environments across the years. The genotypes with high YRI will be useful for those environments with yield fluctuations [24]. While the YRI for the line ‘Red Fife’ was one of the lowest, another line ‘AC Foremost’ was less reliable for grain yield than ‘Red Fife’. The founder line ‘Red Fife’ is a tall cultivar, that was susceptible to severe drought, resulting in poor yield performance in dry environments [33]. There is some limited production of ‘Red Fife as a heritage cultivar, but not for large scale commercial growing. Interestingly, a number of modern hexaploid wheat cultivars and newer genotypes descended from ‘Red Fife’ were found with low mean variance for grain yield along with above-mean grain yield, that is to say have improved yield reliability. Consequently, this demonstration of improved reliability for grain yield through breeding along with the results of this study suggest further possibilities to increase genetic gain for climate resilience and yield reliability under water stress environments with physiological trait based selections as reported by others [18,49].
Potential traits for drought stress selections
One of the morpho-physiological traits we focused on was Δ13C because it showed mostly a positive relationship with grain yield and genotypic variation in wheat for this parameter is also well documented [5,25,27]. It is reported that the environmental factors such as the availability of water during the crop growing period and the intensity of drought can influence the relationship of Δ13C with grain yield [25,50,51]. However, our results showed a consistent positive relationship of Δ13C with grain yield, along with the genotype variation and a high heritability of this trait. Wheat growing areas in the Prairies, including the location of the current study, often receive sufficient moisture and ground recharge from winter snow until anthesis, with subsequent dry spells affecting crop growth and development. The relationship between grain yield and grain Δ13C did not change between early and late sowing, the latter of which creates hot and dry conditions for the crop due to a shortage of available moisture during grain filling. It seems that this phenomenon showing a positive GY-Δ13C relationship could be common in the Prairies although the intensity could vary in different years. Such as in growing year 2012, which is recognized a drought year [23], we found a strong GY-Δ13C relationship compared to the less dry year 2013. These observations indicate the potential of Δ13C to be used in selections for drought stress tolerance and yield related traits. Measuring Δ13C on a small sample of grain from an F2 plant could benefit breeding hexaploid for yield improvement.
Crop plants that show improved water relations, such as water uptake, and optimized transpiration, including higher water use efficiency are desirable for regions that experience regular water stress [49]. Therefore, we focused on WUE as another trait in assessing wheat lines and found that WUE has been increased in many cultivars in parallel to the improvements in grain yield traits since the introduction of ‘Red Fife’ in the early 20th century. The high yield modern hexaploid genotypes such as ‘Reeder’, ‘Stettler’, ‘01S0263-28’, and ‘MN03358-4’ displayed higher WUE compared to older low yielding genotypes ‘Red Fife’ and ‘RL4452’. WUE is a selection target for drought resilience in wheat. Some reports did reveal that drought tolerant crops with higher WUE could show lower grain yield [52]. Our findings are showing improvement in grain yield in lines with higher WUE and suggest that selecting wheat varieties with both high grain yield and the WUE is a possibility when developing drought tolerant wheat germplasm [51]. Many of the hexaploid lines such as ‘Reeder’, ‘AC Karma’, ‘Stettler’, ‘MN03358-4’, and ‘01S0263-28’ were selected and developed for rainfed semi-arid regions in the Prairies, which is likely the case for high WUE and grain yield than other genotypes. In fact, some of the high yielding genotypes in this study have performed well in the farmers fields over the years [53].
Stomata are important components in plant water relations and CO2 uptake for carboxylation in the chloroplast. A change in stomata density or the total number of stomata per leaf area with water stress can impact WUE and Δ13C content in the plant’s tissue thus altering yield [37,54,55]. We also observed variation in stomatal numbers among wheat genotypes grown under well-watered (no-stress) and water stressed conditions to find further evidence of phenotypic plasticity in hexaploid wheat. With the exception of four genotypes, our finding of increased stomatal numbers under water stress, is in agreement with the findings of other studies [reviewed in 56]. The decline in stomatal numbers of four genotypes, ‘8021-V2’, ‘Lillian’, ‘RL4452’, and ‘Red Fife’ after water stress is interesting, as this contrasting change in stomatal numbers per unit leaf area could have a relation with plants’ susceptibility response to drought. Reducing the stomatal numbers is considered important for improving WUE in plants to cope with water stress [54,55]. Perhaps the inherently reduced stomatal numbers and induced stomatal numbers with water stress in a genotype is unrelated and leads to different mechanisms of water stress response in plants. Thus, the reduction in stomatal density and numbers per flag leaf with water stress in low yielding lines ‘Red Fife’ and ‘RL4452’ may explain the plant’s attempt to adapt to a stressful environment, but with limits in the plasticity with other WUE traits. Consequently, the genotypes such as ‘Red Fife’ and ‘RL4452’ were more prone to water loss than other genotypes despite showing adaptive tendencies. The physiological mechanisms for this variation in stomatal number in hexaploid wheat is still not fully clear and needs further understanding.
Two lines, ‘Reeder’ and ‘Stettler’, that are found to be highly yield reliable, did not show a noticeable change in stomata numbers under both water regimes, whereas the others showed smaller change. In a recent published work [35], the genotype ‘Stettler’ showed lower stomatal conductance than another genotype ‘Superb’ under moisture stress (~60% of field capacity), which indicates the role of stomatal control to cope with water stress [35]. Additional information will be useful to understand the function of stomatal traits in relation with WUE in tested genotypes. Both ‘Reeder’ and ‘Stettler’ are found to be highly yield reliable and worth consideration as parents in crossing for incorporation of heat and drought tolerance in wheat for the Prairies. Other stomatal traits such as size and timing of opening and closing [5,55] may explain their role in regulating water stress mechanisms in hexaploid wheat.
Relationship of GY with different traits
The relationship between Δ13C and WUE is known to be complex. While some reports describe a lower Δ13C with higher WUE in plants, others demonstrate a positive increase in Δ13C with increasing WUE [5,50,57,58]. This association is influenced by the factors such as availability of water and nutrients, soil properties, and species [59]. The association of GY with Δ13C and WUE was strong in our study, although the GY was poorly correlated with stomatal numbers. Improvements in crop genetics through physiological breeding with selective parameters such as Δ13C and WUE is realistic based on the development of wheat cultivars ‘Drysdale’ and ‘Rees’ for drought prone environments in Australia [11]. Our results corroborate those reports providing a further line of evidence that physiological traits such as Δ13C and WUE could be reliable physiological markers for germplasm screening. The performance of conventionally bred cultivars such as ‘Reeder’, with increased WUE and corresponding Δ13C suggests the possibility of selecting these traits for improving yield reliability in hot and dry rainfed agri-zones such as Canadian Prairies. However, WUE is extremely cumbersome to measure in a breeding situation, whereas Δ13C is technically feasible, where laboratory facilities are available to process samples from breeding programs.
Interestingly, one of the important physiological traits, CT was not well correlated with GY in different field environments within our study. This is in contrast to many other reports that showed CT as a good proxy for GY prediction [12,30,54]. Many factors such as plant density, height, and growth stage are known to introduce confounding results to CT that would not accurately predict GY or other associated parameters [54,60,61]. Since the wheat genotypes in the panel have divergent phenology and morphological features, we couldn’t identify the confounding factors influencing the CT-GY relationship [62]. We also conducted an extended experiment in 2023, to measure CT of the genotypes at different times of the day (7:00 a.m., 10:00 a.m., 1:00 p.m., 4:00 p.m., 7:00 p.m.), but no factors were identified for the inconsistent behaviour of CT results. Nevertheless, lower CT was observed for high yielding genotypes such as ‘AC Foremost’ compared to other genotypes with low yield such as ‘Vesper’ and ‘Red Fife’. These observations imply that to obtain accurate CT measurements, timing of observations relative to phenological stage and perhaps different and more reliable tools are required for in a hot and dry (rainfed) environment.
Conclusion
Heat and drought are major limiting factors affecting crop productivity. Projected climate change and the possibility of continuous drought events in wheat growing regions will require that more and diverse wheat germplasm be evaluated to identify traits better adapted for improved water relations or other water saving mechanisms for reliable grain yield in hot dry regions, such as the Prairies. Development of germplasm with abiotic stress tolerance requires plasticity of traits that plants can utilize to adapt to changing weather conditions, without having significant yield penalty. Four yield reliable genotypes (Reeder, MN03358-4, 01S0263-28, and Stettler) found in this study are worth further investigation, due to showing less variation in yield in different field environments. In addition, wheat yield reliability could be improved through focusing on specific traits such as WUE and Δ13C and stomatal numbers per plant. Additional studies could help to better understand these findings. In fact, higher heritability of these traits suggests that measuring these physiological traits in an early generation of breeding would be feasible and a better means of enriching breeding populations for yield reliability before multi-environment yield trials are applied to the populations. For example, Δ13C can be determined on a single plant basis from grain of plants selected from large early generation breeding populations. Furthermore, genetic populations already developed from crosses of the genotypes evaluated in the present study can be used for genetic studies to find genes or QTL associated with yield reliability and traits such as Δ13C. Additional studies are underway to confirm the use of morpho-physiological traits, including Δ13C, WUE, and low stomatal numbers in selection of early generation wheat lines in diverse field environments through a physiological breeding approach.
Supporting information
S1 Fig. Stomatal numbers on adaxial and abaxial sides of the flag leaves.
Total number of stomata on flag leaf of hexaploid wheat grown in well-watered (WW) and water stress (WS) conditions in a glasshouse experiment. Data based on nine images taken on three independent flag leaves (Mean ± SE).
https://doi.org/10.1371/journal.pone.0312122.s001
(PDF)
S2 Fig. Flag leaf area of hexaploid wheat.
Mean flag leaf area of hexaploid wheat genotypes grown in well-watered (WW, blue solid line) and water stress (WS, red solid line) conditions in a glasshouse experiment. Data based on three randomly selected flag leaves for measurements of leaf area (Mean ± SE).
https://doi.org/10.1371/journal.pone.0312122.s002
(PDF)
Acknowledgments
We appreciate access to the climate data provided by Jason Nimegeers and information on wheat genotypes provided by Shawn Yates. We thank Dale Kern for stomatal counting. We are grateful to Dr. Benjamin Ellert and Brett Hill at AAFC-Lethbridge for timely analysis of stable isotopes. We also thank the Agriculture and Agri-Food Canada’s cereal staff at Swift Current, SK for providing field management and phenotyping support.
References
- 1. Zhang T, He Y, DePauw R, Jin Z, Garvin D, Yue X, et al. Climate change may outpace current wheat breeding yield improvements in North America. Nat Commun. 2022;13(1):5591. pmid:36180462
- 2. Curtis T, Halford NG. Food security: the challenge of increasing wheat yield and the importance of not compromising food safety. Ann Appl Biol. 2014;164(3):354–72. pmid:25540461
- 3.
Erenstein O, Jaleta M, Mottaleb KA, Sonder K, Donovan J, Braun HJ. Global trends in wheat production, consumption and trade. In: Reynolds MP, Braun HJ, editors. Wheat improvement: food security in a changing climate. Cham: Springer International Publishing; 2022. p. 47–66.
- 4. Bailey-Serres J, Parker JE, Ainsworth EA, Oldroyd GED, Schroeder JI. Genetic strategies for improving crop yields. Nature. 2019;575(7781):109–18. pmid:31695205
- 5. Condon AG. Drying times: plant traits to improve crop water use efficiency and yield. J Exp Bot. 2020;71(7):2239–52. pmid:31912130
- 6. Zandalinas SI, Fritschi FB, Mittler R. Global warming, climate change, and environmental pollution: recipe for a multifactorial stress combination disaster. Trends Plant Sci. 2021;26(6):588–99. pmid:33745784
- 7. Langridge P, Reynolds M. Breeding for drought and heat tolerance in wheat. Theor Appl Genet. 2021;134(6):1753–69. pmid:33715017
- 8.
Johansson E, Branlard G, Cuniberti M, Flagella Z, Hüsken A, Nurit E, et al. Genotypic and environmental effects on wheat technological and nutritional quality. In: Igrejas G, Ikeda T, Guzmán C, editors. Wheat quality for improving processing and human health. Cham: Springer; 2020. p. 171–204.
- 9. Brooker R, Brown LK, George TS, Pakeman RJ, Palmer S, Ramsay L, et al. Active and adaptive plasticity in a changing climate. Trends Plant Sci. 2022;27(7):717–28. pmid:35282996
- 10. Vishwakarma S, Zhang X, Lyubchich V. Wheat trade tends to happen between countries with contrasting extreme weather stress and synchronous yield variation. Commun Earth Environ. 2022;2(1):61.
- 11. Richards RA. Physiological traits used in the breeding of new cultivars for water-scarce environments. Agri Water Manage. 2006;80(1–3):197–211.
- 12. Li X, Ingvordsen CH, Weiss M, Rebetzke GJ, Condon AG, James RA, et al. Deeper roots associated with cooler canopies, higher normalized difference vegetation index, and greater yield in three wheat populations grown on stored soil water. J Exp Bot. 2019;70(18):4963–74. pmid:31089708
- 13. Mahpara S, Bashir MS, Ullah R, Bilal M, Kausar S, Latif MI, et al. Field screening of diverse wheat germplasm for determining their adaptability to semi-arid climatic conditions. PLoS One. 2022;17(3):e0265344. pmid:35303032
- 14. Statistics Canada [Internet]. Estimated areas, yield, production, average farm price and total farm value of principal field crops, in metric and imperial units; c2018 Jun 27 - [cited 2024 Jan 25]. Available from: https://www150.statcan.gc.ca/n1/tbl/csv/32100359-eng.zip
- 15. Carew R, Meng T, Florkowski WJ, Smith R, Blair D. Climate change impacts on hard red spring wheat yield and production risk: evidence from Manitoba, Canada. Can J Plant Sci. 2018;98(3):782–95.
- 16. Qian B, Zhang X, Smith W, Grant B, Jing Q, Cannon AJ. Climate change impacts on Canadian yields of spring wheat, canola and maize for global warming levels of 1.5 °C, 2.0 °C, 2.5 °C and 3.0 °C. Environ Res Lett. 2019;14(7):.
- 17. Griffiths CA, Reynolds MP, Paul MJ. Combining yield potential and drought resilience in a spring wheat diversity panel. Food Energy Secur. 2020;9(4):e241. pmid:33391733
- 18. Reynolds MP, Lewis JM, Ammar K, Basnet BR, Crespo-Herrera L, Crossa J, et al. Harnessing translational research in wheat for climate resilience. J Exp Bot. 2021;72(14):5134–57. pmid:34139769
- 19. Leigh FJ, Wright TIC, Horsnell RA, Dyer S, Bentley AR. Progenitor species hold untapped diversity for potential climate-responsive traits for use in wheat breeding and crop improvement. Heredity (Edinb). 2022;128(5):291–303. pmid:35383318
- 20. Reynolds M, Dreccer F, Trethowan R. Drought-adaptive traits derived from wheat wild relatives and landraces. J Exp Botany. 2007;58177–86.
- 21. Zhang J, Zhang S, Cheng M, Jiang H, Zhang X, Peng C, et al. Effect of drought on agronomic traits of rice and wheat: a meta-analysis. Int J Environ Res Public Health. 2018;15(5):839. pmid:29695095
- 22. Wheaton E, Kulshreshtha S, Wittrock V, Koshida G. Dry times: hard lessons from the Canadian drought of 2001 and 2002. Canadian Geographies / Géographies Canadiennes. 2008;52(2):241–62.
- 23.
Laforge J, Corkal V, Cosbey A. Farming the Future: Agriculture and climate change on the Canadian Prairies. Winnipeg (MB): The International Institute for Sustainable Development. 2021.
- 24. Döring TF, Annicchiarico P, Clarke S, Haigh Z, Jones HE, Pearce H, et al. Comparative analysis of performance and stability among composite cross populations, variety mixtures and pure lines of winter wheat in organic and conventional cropping systems. Field Crops Research. 2015;183235–45.
- 25. Condon AG, Richards RA, Rebetzke GJ, Farquhar GD. Breeding for high water-use efficiency. J Exp Bot. 2004;55(407):2447–60. pmid:15475373
- 26. Araus JL, Slafer GA, Royo C, Serret MD. Breeding for Yield Potential and Stress Adaptation in Cereals. Crit Rev Plant Sci. 2008;27(6):377–412.
- 27. Cabrera-Bosquet L, Albrizio R, Nogués S, Araus JL. Dual Δ¹³C/δ¹⁸O response to water and nitrogen availability and its relationship with yield in field-grown durum wheat. Plant Cell Environ. 2011;34(3):418–33. pmid:21062317
- 28. Sailaja B, Subrahmanyam D, Neelamraju S, Vishnukiran T, Rao YV, Vijayalakshmi P, et al. Integrated physiological, biochemical, and molecular analysis identifies important traits and mechanisms associated with differential response of rice genotypes to elevated temperature. Front Plant Sci. 2015;61044. pmid:26640473
- 29. Fleury D, Jefferies S, Kuchel H, Langridge P. Genetic and genomic tools to improve drought tolerance in wheat. J Exp Bot. 2010;61(12):3211–22. pmid:20525798
- 30. Reynolds M, Langridge P. Physiological breeding. Curr Opin Plant Biol. 2016;31162–71. pmid:27161822
- 31. Bokore FE, Cuthbert RD, Knox RE, Singh A, Campbell HL, Pozniak CJ, et al. Mapping quantitative trait loci associated with common bunt resistance in a spring wheat (Triticum aestivum L.) variety Lillian. Theor Appl Genet. 2019;132(11):3023–33. pmid:31410494
- 32. Ashe P, Shaterian H, Akhov L, Kulkarni M, Selvaraj G. Contrasting root and photosynthesis traits in a large-acreage Canadian durum variety and its distant parent of Algerian origin for assembling drought/heat tolerance attributes. Front Chem. 2017;5:121. pmid:29312927
- 33.
Hucl P, Baker RJ. A study of ancestral and modern Canadian spring wheats. Can J Plant Sci. 1987;67:87-97.
- 34. Willick IR, Lahlali R, Vijayan P, Muir D, Karunakaran C, Tanino KK. Wheat flag leaf epicuticular wax morphology and composition in response to moderate drought stress are revealed by SEM, FTIR-ATR and synchrotron X-ray spectroscopy. Physiol Plant. 2018;162(3):316–32. pmid:28857201
- 35. Sharma G, Brar GS, Knipfer T. Leaf water relations and osmotic adjustment of Canada Western Red Spring wheat cultivars subjected to drought. Funct Plant Biol. 2023;50(12):1037–46. pmid:37814368
- 36. Farquhar G, Ehleringer I, Hubick K. Carbon isotope discrimination and photosynthesis. Ann Rev Plant Physiol Plant Molecul Biol. 1989;40:503–37.
- 37.
Pask AJD, Pietragalla J, Mullan DM, Reynolds MP. Physiological Breeding II: A Field Guide to Wheat Phenotyping. Mexico: CIMMYT; 2012.
- 38. Ferris R, Long L, Bunn SM, Robinson KM, Bradshaw HD, Rae AM, et al. Leaf stomatal and epidermal cell development: identification of putative quantitative trait loci in relation to elevated carbon dioxide concentration in poplar. Tree Physiol. 2002;22(9):633–40. pmid:12069919
- 39.
Kataoka S. A stochastic programming model. Econometrika. 1963;31:181–96.
- 40. Royo C, Ammar K, Villegas D, Soriano JM. Agronomic, physiological and genetic changes associated with evolution, migration and modern breeding in durum wheat. Front Plant Sci. 2021;12:674470.
- 41. Pinto RS, Reynolds MP, Mathews KL, McIntyre CL, Olivares-Villegas J-J, Chapman SC. Heat and drought adaptive QTL in a wheat population designed to minimize confounding agronomic effects. Theor Appl Genet. 2010;121(6):1001–21. pmid:20523964
- 42. Castillo JM, Gallego-Tévar B, Figueroa E, Grewell BJ, Vallet D, Rousseau H, et al. Low genetic diversity contrasts with high phenotypic variability in heptaploid Spartina densiflora populations invading the Pacific coast of North America. Ecol Evol. 2018;8(10):4992–5007. pmid:29876076
- 43. Tuhina-Khatun M, Hanafi MM, Rafii Yusop M, Wong MY, Salleh FM, Ferdous J. Genetic variation, heritability, and diversity analysis of upland rice (Oryza sativa L.) genotypes based on quantitative traits. Biomed Res Int. 2015;2015290861. pmid:26258135
- 44. Hossain MdM, Azad MdAK, Alam MdS, Eaton TE-J. Estimation of Variability, Heritability and Genetic Advance for Phenological, Physiological and Yield Contributing Attributes in Wheat Genotypes under Heat Stress Condition. AJPS. 2021;12(04):586–602.
- 45. Johnson H, Robinson H, Comstock R. Estimates of genetic and environmental variability in soybean. Agron J. 1955;47314–8.
- 46. Maniee M, Kahrizi D, Mohammadi R. Genetic variability of some morpho-physiological traits in durum wheat (Triticum durum Desf.). J Appl Sci. 2009;91383–7.
- 47. Schmidt P, Hartung J, Bennewitz J, Piepho H-P. Heritability in Plant Breeding on a Genotype-Difference Basis. Genetics. 2019;212(4):991–1008. pmid:31248886
- 48.
Annicchiarico P, Genotype × Environment Interactions: Challenges and opportunities for plant breeding and cultivar recommendations. FAO plant production and protection papers, vol. 1(174). Rome: Food & Agriculture Organization. 2002. ISSN 0259-2517.
- 49. Pask AJD, Reynolds MP. Breeding for yield potential has increased deep soil water extraction capacity in irrigated wheat. Crop Science. 2013;53(5):2090–104.
- 50. Rebetzke GJ, Condon AG, Richards RA, Farquhar GD. Selection for Reduced Carbon Isotope Discrimination Increases Aerial Biomass and Grain Yield of Rainfed Bread Wheat. Crop Science. 2002;42(3):739–45.
- 51. Isidro-Sánchez J, Perry B, Singh A, Wang H, DePauw R, Pozniak C. Effects of seeding rate on durum crop production and physiological responses. Agron J. 2017;1091981–90.
- 52. Zhang Y, Yu X, Chen L, Jia G. Whole-plant instantaneous and short-term water-use efficiency in response to soil water content and CO2 concentration. Plant Soil. 2019;444(1–2):281–98.
- 53. McCallum BD, DePauw RM. A review of wheat cultivars grown in the Canadian prairies. Can J Plant Sci. 2008;88(4):649–77.
- 54. Rebetzke GJ, Rattey AR, Farquhar GD, Richards RA, Condon ATG. Genomic regions for canopy temperature and their genetic association with stomatal conductance and grain yield in wheat. Funct Plant Biol. 2012;40(1):14–33. pmid:32481083
- 55. Buckley CR, Caine RS, Gray JE. Pores for Thought: Can Genetic Manipulation of Stomatal Density Protect Future Rice Yields?. Front Plant Sci. 2020;101783. pmid:32117345
- 56. Shahinnia F, Le Roy J, Laborde B, Sznajder B, Kalambettu P, Mahjourimajd S, et al. Genetic association of stomatal traits and yield in wheat grown in low rainfall environments. BMC Plant Biol. 2016;16(1):150. pmid:27378125
- 57. Foulkes MJ, DeSilva J, Gaju O, Carvalho P. Relationships between δ 13 C, δ 18 O and grain yield in bread wheat genotypes under favourable irrigated and rain-fed conditions. Field Crops Research. 2016;196237–50.
- 58. Pronger J, Campbell DI, Clearwater MJ, Mudge PL, Rutledge S, Wall AM, et al. Toward optimisation of water use efficiency in dryland pastures using carbon isotope discrimination as a tool to select plant species mixtures. Sci Total Environ. 2019;665698–708. pmid:30780015
- 59. Tambussi EA, Bort J, Araus JL. Water use efficiency in C3 cereals under Mediterranean conditions: a review of physiological aspects. Ann Appl Biol. 2007;150:307–21.
- 60. Anderegg J, Aasen H, Perich G, Roth L, Walter A, Hund A. Temporal trends in canopy temperature and greenness are potential indicators of late-season drought avoidance and functional stay-green in wheat. Field Crops Research. 2021;274:108311.
- 61. Kunz K, Hu Y, Schmidhalter U. Carbon isotope discrimination as a key physiological trait to phenotype drought/heat resistance of future climate-resilient German winter wheat compared with relative leaf water content and canopy temperature. Front Plant Sci. 2022;13:1043458. pmid:36589131
- 62. Luan X, Vico G. Canopy temperature and heat stress are increased by compound high air temperature and water stress and reduced by irrigation – a modeling analysis. Hydrol Earth Syst Sci. 2021;25(3):1411–23.