Managing Climate Change Impacts on Crops: The Influence of Soil Tillage on a Triticale Crop under Water Stress Conditions

ORIGINAL ARTICLE Managing climate change impacts on crops: The influence of soil tillage on a triticale crop under water stress conditions Paula Madejón 1 | Elena Fernández-Boy 2 | Engracia Madejón 1 | Laura Morales-Salmerón 2 | María Teresa Domínguez 2 1 Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS), CSIC, Sevilla, Spain 2 Departamento de Cristalografía, Mineralogía y Química Agrícola, Universidad de Sevilla, Sevilla, Spain Correspondence Paula Madejón, Instituto de Recursos Naturales y Agrobiología de Sevilla (IRNAS), CSIC, Sevilla, Spain. Email: [email protected]c.es Funding information European Regional Development Fund (FEDER), Grant/Award Number: US-1260627; Spanish Ministry of Science and Innovation, Grant/Award Number: PID2021-122628OBI00 WASTE4DROUGHT; European funds; Ministry of Science and Innovation, Grant/Award Number: PRE2018-084467 Abstract Water limitations for agriculture will likely become crucial in the next decades in some regions such as the Mediterranean basin with the current climate change projections. In this context, recent evidence suggests that the application of conservation agriculture, which reduces the frequency and intensity of soil tillage, could confer a higher stability of agricultural systems against climate variability. However, not many experiments have addressed the interaction between tillage type and the resistance to drought in rainfed crops. In this work, we evaluated the resistance to drought of triticale (Triticale hexaploide L.) crops managed with different tillage systems: traditional tillage (TT), reduced tillage (RT) and no tillage (NT). A rainfall exclusion experiment was carried out in a typical wheat/legume Mediterranean rotation in SW Spain, in a long-term experiment established in 2008 comparing the three tillage systems. Grain yield and different variables related to plant ecophysiology, root development, biomass allocation and colonisation by arbuscular mycorrhizal fungi (AMF) were evaluated over one crop cycle. Tillage type had a significant influence on soil water storage (SWS), such that soils under NT had, on average, a 16% greater SWS than soils under RT or TT. Grain yield was significantly reduced by rainfall exclusion, in particular in the TT, where drought reduced grain yield by 31%. Gas exchange data also showed that plants in the TT system were more sensitive to drought, such that maximum photosynthesis rates were reduced by 25% because of rainfall exclusion in this tillage system. Drought had a negative impact on root biomass across the three tillage systems, especially in the RT, where a reduction in the root:shoot ratio was observed. The effect of tillage on mycorrhizal colonisation was more evident than the effect of drought; in general, conservation tillage systems (RT and NT) tended to have higher values for all AMF traits compared to the TT. In summary, the NT system tended to exhibit more favourable performance in terms of soil water retention, grain yield stability under drought conditions and mycorrhizal symbiosis, which suggests enhanced resource use efficiency in this system. KEYWORDS drought, rainfed crops, rainout shelters, soil water storage, tillage types, triticale Received: 27 December 2023 Revised: 9 September 2024 Accepted: 10 September 2024 DOI: 10.1111/aab.12947 This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2024 The Author(s). Annals of Applied Biology published by John Wiley & Sons Ltd on behalf of Association of Applied Biologists. Ann Appl Biol. 2025;186:143–156. wileyonlinelibrary.com/journal/aab 143 1|INTRODUCTION Global climate change has increased the risk of short-term extreme climate stresses in agriculture. Drought is emerging as one of the most significant abiotic stresses limiting crop growth and development (Cruz de Carvalho, 2008; Lafitte et al., 2007). Actually, global water shortage currently limits crop yields in more than 70% of arable lands. These water limitations will likely become more important in the next decades in some regions, such as the Mediterranean basin, with predictions of reduced rainfall (IPCC, 2022). In addition, increments of agricultural activities in unfertile areas to satisfy growing demands for food will likely increase the need for water and nutrient inputs in agricultural systems (Hemathilake & Gunathilake, 2022). There has been a significant rise in air temperature (almost 1.2C in the last two decades) and evapotranspiration in the Mediterranean region, and there is a high confidence that climate change has worsened heat waves and droughts, which are already affecting agriculture in this region (reviewed in Ali et al., 2022). Accordingly, understanding and improving plant survival and growth under restricted water availability is of central relevance in contemporary plant science, and particularly for Mediterranean crops given the adverse climate change projections for this region. Triticale (Triticale hexaploide L.) is a crop species bred from wheat (Triticum aestivum L.) and rye (Secale cereale L.) through intergeneric hybridisation and doubling of hybrid chromosomes (Ayalew et al., 2018). This crop is emerging as an important source of food and animal feed owing to its good nutritional quality, strong stress resistance, wide adaptability and high yield. Triticale is a valuable stress-tolerant cereal and a potential genetic resource for breeding winter and spring cereals (Blum, 2014). In fact, it is known for its higher resistance under drought stress compared to wheat (Giunta et al., 1993). Moreover, some authors reported that under different levels of drought, triticale yield showed a non-significant reduction (8%), while durum wheat was significantly decreased in comparison to the irrigated control (54%) (Alatrash et al., 2022). Major objectives for the improvement of triticale crops include, among others, increasing grain yield, shortening plant height, and improving water use efficiency and its tolerance/ resistance to various biotic and abiotic stresses such as drought (Mergoum et al., 2019). Therefore, it is necessary to study the effects of water stress on the physiological processes of this crop plant for its best management (Munjonji et al., 2017). Soil management can play a central role in water storage capacity in crops, particularly in rainfed systems. In a scenario of climate uncertainty and scarcity of water resources, some studies have suggested that conservation agriculture could increase the stability of agricultural systems against climate variability (Puig-Sirera et al., 2022). In this regard, reduced tillage (RT) or no tillage (NT) have been adopted in arid and semi-arid regions with positive effects on crop production by changing the soil environment (Du et al., 2022). Increases of soil water infiltration and retention under NT or RT are often related to improvements in grain yields and enhanced water use efficiency of the crop, especially under drought conditions (Du et al., 2022). Multiple studies in the Mediterranean region have concluded that no-till and low tillage not only increase soil moisture but also improve water use efficiency (Cantero-Martínez et al., 2007; Fernández-García et al., 2013; Morell et al., 2011). However, this effect is not always observed (Madejón et al., 2023). Tillage can have a profound effect on soil biological communities, which may have some implications for the resistance of crop plants to water stress. For instance, several works have shown that NT often promotes an increase in the abundance of fungi (Panettieri et al., 2020), including those establishing mycorrhizal symbiosis with plants (Brito et al., 2021; Sebbane et al., 2023), and this could be beneficial for plant water uptake under drought stress conditions (AbdelFattah & Asrar, 2012; Liu et al., 2015). The objective of this study was to evaluate the potential impact of drought predictions for the Mediterranean region on grain yield in a triticale crop, and to assess whether this impact can be minimised by the adoption of conservation tillage practices. The hypothesis of this work is that the type of tillage could have an influence on the resistance of triticale to drought conditions. Therefore, we could expect that conservation tillage would increase the crop resistance to reductions of water inputs, and that this increase could be verified through crop ecophysiological indices and changes in root architecture. To test these hypotheses, a rainfall exclusion experiment was carried out in a typical wheat/legume Mediterranean rotation in SW Spain, taking advantage of a long-term experiment established in 2008. Different variables related to plant ecophysiology, root development and arbuscular mycorrhizal fungi (AMF) colonisation and productivity were evaluated over one crop cycle. The use of rainout shelters to simulate drought conditions is a practical approach to test the influence of tillage on crop resistance to drought, as it helps to control experimental variables and allows for direct comparison between treatments. Moreover, this is a long-term experiment permitting the capture of the cumulative effects of different tillage systems on soil properties, which can influence the plant response to drought. Finally, this experiment could contribute valuable information to agricultural practices in Mediterranean regions and beyond. If conservation tillage is indeed found to enhance triticale drought resistance, this knowledge could be used to guide farmer choices of tillage methods to mitigate the effects of water scarcity on crop yields. 2|MATERIALS AND METHODS 2.1 |Experimental area and description of the experimental design The long-term tillage experiment was established in 2008 at the agriculture experimental farm ‘La Hampa’of the ‘Instituto de Recursos Naturales y Agrobiología de Sevilla’(IRNAS-CSIC), located at Coria del Río (Seville, SW Spain) to compare three tillage systems: traditional tillage (TT), reduced tillage (RT) and no tillage (NT). Three replicate plots of 300 m 2 were delimited for each tillage type in a completely randomised experimental design. See more details of the design of the experiment in previous works (Madejón et al., 2023; Panettieri 144 MADEJÓN ET AL. 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License et al., 2020). The tillage treatments were established in 2008 and continue to be carried out every year in all the plots. Briefly, TT consists of mouldboard ploughing with soil inversion (25–30 cm deep) and two chisel operations at 25 cm depth (0.57 m separation between chisels) followed by a disc harrowing of 12 cm depth; RT operation consists of one chisel operation at 25 cm depth followed by a disc harrowing of 5 cm depth. Finally, in NT sowing is done by direct drilling (no pre-sowing operations are carried out). Moreover, at least 30% of the soil surface under NT and RT remains covered by crop residues from the previous crop, while 10–15 % of crop residues from TT are left on the soil surface after harvest, these remaining over the summer to be buried by ploughing in the autumn.The soil texture at the site is sandy clay loam, and it is classified as Typic Xerofluvent (Soil Survey Staff, 2014). Soil chemical properties measured before the sowing of the triticale crop are shown in Table S1. The climate is Mediterranean, with mild rainy winters and hot and dry summers. Weather conditions during the study were recorded at a meteorological station located at the experimental farm (Figure 1). FIGURE 1 (a) Precipitation (blue bars) and temperature (red line) during the study period. Soil water storage (SWS) in the 0– 40 cm layer under different tillage and rainfall conditions, from sowing date to crop harvest (mean values ± SD), (b) No tillage (NT), (c) Reduced tillage (RT) and (d) Traditional tillage (TT). Rainfall exclusion treatment (EXC) is shown with yellow circles and control treatment (CONT) with green circles. MADEJÓN ET AL.145 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License In October 2020, a rainfall exclusion experiment was set up. For this purpose, rainout shelters were used. Each shelter covers a soil surface of 6.25 m 2 (2. 5 m 2.5 m) and is made of light metallic structure with six bands made of transparent acrylic that exclude approximately 30% of rainfall inputs, which is within the range of climate change projections for the Mediterranean region (IPCC, 2022). At each tillage plot (nine plots in total, three per tillage system), two rainout shelters were established (rainfall exclusion treatment). Rainout shelters were established in November 2020 and were over the plots throughout the experiment. They were only removed for tillage and planting, but immediately after these tasks they were placed in their corresponding plots. Likewise, at each tillage plot, two areas of equal size (2.5 m 2.5 m) were delimited on the ground, subjected to environmental precipitation conditions (control treatment). The number of total experimental plots was 36 (three tillage systems two rainfall levels six replicates). Soil moisture across the 0–40 cm depth was periodically recorded with a Frequency Domain Reflectometry (FDR) probe (Delta-T Devices) in fibreglass tubes installed at the centre of each plot in half the replicates (three tillage two rainfall levels  three replicates). A legume-triticale crop rotation was established at the experiment. Tillage operations described above for NT, RT and TT are performed during the autumn before seed sowing. Vicia faba L., was the crop established in the previous season (from November 2020 to May 2021; see details in Madejón et al., 2023), while triticale was cultivated over the 2021–2022 season. Tillage was conducted in September 2021. In the middle of November, in the NT treatment, spray with pre-emergence glyphosate (KARDA) at a rate of 6 L ha 1 was applied. Finally, in the middle of December 2021, triticale, ‘Saleroso R-2’variety, was sown with a density of 250 kg of seeds ha 1 . At the beginning of April 2022, due to the high amounts of weeds, it was necessary to treat the plots with a mixture of two herbicides (KINVARA at a rate of 0.33 g ha 1 and MONOLITH at a rate of 2.25 L ha 1 ). 2.2 |Plant germination and mycorrhizal root colonisation Seedling emergence was estimated 34 days after sowing, when it was observed that the majority of seedlings had emerged, by counting all the germinated seedlings in all the rainout shelters and in the control plots. At the end of March (28th), three complete plants (including roots) were sampled in a selection of 24 plots (only in four plots of each tillage system with and without rainfall exclusion). For the study of the colonisation by AMF, a sub-sample of secondary roots (diameter less than 2 mm) was obtained and treated following the staining method developedbyVierheiligetal.( 1998). The root material was hot digested with 10% (w/v) KOH until the roots were discoloured and then stained with 0.05% blue ink (Pelikan 4001) in vinegar. For all the measures, 30 fragments of the composite samples per plot were selected. They were further cut into 1 cm pieces and examined under a microscope (Olympus BX40). AMF colonisation and the abundance of arbuscules and vesicles were calculated following the method of Trouvelot et al. (1986). The variables studied were: degree of colonisation by AMF, indicated by the frequency of mycorrhizae in the root system (F%), the intensity of mycorrhizal colonisation in the root system (M%), intensity of mycorrhizal colonisation in root fragments (m%), abundance of vesicles structures used to store lipids and other elements (v%) and abundance of arbuscules, where carbon and nutrient exchange occurs (a%) (Smith & Read, 2008). 2.3 |Ecophysiological measurements In the middle of April 2022 (19th), measurements of gas exchange were conducted with a portable photosynthesis system (LI-6400, LiCor, Lincoln, NE, USA) to determine maximum rates of photosynthesis (AN, max), stomatal conductance (g s , max), intercellular concentration of carbon dioxide (Ci), and transpiration rate (E). Gas exchange was measured between 10:00 and 13:00 h on healthy, fully developed young leaves in a selection of 18 plots (three plots three tillage systems two rainfall levels), measuring three plants per plot. Measurement conditions were set to 350 μmol air s 1 , 430 μmol CO 2 air mol 1 , with a PAR of 1500 μmol photons m 2 s 1 . A Scholander-type pressure chamber (PMS Instrument Company, Albany, OR, USA) was used to measure leaf water potential. For these measurements, one leaf per plant and two plants per plot were selected. Fully developed leaves from the outer part of the plant canopy were selected for these measurements. The sampled leaves were stored in closed plastic bags with humid filter paper and kept in a portable cooler until leaf water potential was measured in the laboratory (Rodriguez-Domínguez et al., 2022). 2.4 |Plant traits and chemical analyses Three plants (shoots and roots) from each plot (only in four plots of each tillage system with and without rain exclusion, 24 plots) were sampled at the end of March 2022, coinciding with the sampling for mycorrhizal traits. Samples were freshly weighed, washed with distilled water and dried at 60C for at least 48 h. Plant shoots (stems, leaves, spikes) and roots (principal and secondary) were dried separately and weighed. Leaves mass fraction (LMF), stem mass fraction (SMF), fruit mass fraction (Spike, FMF), root mass fraction (RMF) and the shoot:root ratio were recorded. To analyse macro and micronutrient content in leaves, these were digested by wet oxidation with concentrated HNO 3 in a Digiprep Ms. block digester, and the extracts were determined by inductively coupled plasma spectrophotometry (ICP-OES: Varian ICP 720-ES, with axially viewed plasma). The accuracy of the analytical method was determined using a plant reference material: INCT-ONTL-5 (Tobacco leaves). Recovery rates for reference plant samples ranged between 95% and 105%. 146 MADEJÓN ET AL. 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License 2.5 |Plan production At the end of May (31st), when the spikes were at maturity, the aboveground part of the biomass was collected within two quadrats of 0.50 m 2 per plot. These samples included both triticale plants and weeds. Triticale spikes were further separated from stems. Plant material (spikes and weeds) was weighed fresh, and subsequently dried at 60C for 48 h. After drying the spikes, their grains were separated from the chaff using a Sheller (SAS GOD E, model V 05-204). Grain dry weight per plot was also recorded. 2.6 |Data analysis Soil water storage (SWS, mm), of the 0–40 cm layers (root zone), was calculated from soil water content measurements (θ) through the soil profile on 15 sampling dates between 17 September 2021 and 18 May 2022 following trapezoidal rules (Jia et al., 2013) according to the equation: SWS¼50θ1þX 3 i¼1 50θiþθiþ1 ðÞ: where θ i represents volumetric moisture (cm 3 cm 3 ) in the ith soil depth (1: 10 cm; 2: 20 cm; 3: 30 cm; 4: 40 cm); the soil depth interval was 10 cm. Linear mixed models were applied to test for the effect of tillage and rainfall exclusion (as well as their interactions) on SWS, seedling emergence and crop yield, mycorrhizal traits and ecophysiological variables, as well as on plant chemical traits, using the nlme package in R. The plot was included as a random term to account for repeated measurements of soil water content. Also, the random term included the spatially nested design, given that within each tillage plot, two control plots and two drought plots were delimited. Validation of the model assumptions was done by exploration of model residuals. When homogeneity of variance was not met, a variance coefficient was introduced in the model to account for heteroskedasticity among different factor levels, using the varIdent function of the nlme package (Pinheiro & Bates, 2000). No transformation of data was required to satisfy the assumptions of the analysis. Pearson's correlation analyses were used to explore the relationships between crop production and weed incidence, as well as among mycorrhizal variables and plant ecophysiological measurements. 3|RESULTS 3.1 |Soil water storage: Effect of rainfall exclusion and tillage Tillage type had a significant influence on SWS (linear mixed modelling, F=4.10, p=.044), and soils under NT had greater SWS than soils under RT or TT. These differences among the three tillage systems were especially noticeable from January 2022 to the end of the experiment in May (Figure 1). From September 2021 to May 2022, SWS values were, on average, 16% higher in soils under NT than in those under RT and TT. The impact of rainfall exclusion on SWS depended on the date of the sampling. The simulated drought reduced SWS significantly (p< .05, Ftest) in 5 out of 15 sampling dates, especially in the spring season (Figure S1). The greatest differences between control and exclusion plots were observed on 4 November 2021 for NT and 18 March 2022 for TT, when the exclusion treatment led to a reduction of SWS of 21 and 22 mm in the NT and TT treatments on average, respectively (Figure S1). In the case of RT, the maximum difference was found 1 month later (a reduction of 22 mm). These values represented reductions of 20%, 26% and 21% of SWS relative to control plots in the NT, RT and TT, respectively. At the end of the experiment, there was a significant tillage rainfall treatment interaction, so rainfall exclusion led to a reduction of SWS in the RT system, but not in the TT or the NT systems (Figure S1). 3.2 |Germination and grain production: Effects of rainfall exclusion and tillage Seedling emergence was not influenced by either tillage or rainfall exclusion (Table 1). In contrast, crop production, assessed by the measurement of the grain yield (Figure 2a), was significantly reduced by rainfall exclusion (F=5.2653, p=.0308), with yields observed in the exclusion treatment around 22% lower compared to the control. Among the tillage systems, only TT exhibited significant differences in grain yield (F=5.59, p=.045) because of rainfall exclusion (31% less in the exclusion treatment compared to control), whereas changes in grain yield because of rainfall exclusion were not significant in the conservation tillage systems (F=0.526, p=.489, and F=0.769, p=.406 for NT and RT, respectively). Despite the different impacts of rainfall exclusion, both NT and TT showed similar grain yield values in the exclusion treatment. Considering the scenario of reduced rainfall, both NT and TT maintained grain yields of 1614 and 1490 kg ha 1 , respectively (slightly higher on NT). Wild grasses (weeds) were also evaluated because they can affect the proper development of the crop (Figure 2b). The negative and significant correlation (r=0.57, p=.0003; n=36) between grain yield and biomass of wild grasses highlights the competition for TABLE 1 Number of germinated seeds according to the two experimental factors, tillage and rainfall conditions (mean values ± SE; N=6). Tillage Control Exclusion NT 115.2 ± 54.8 136.2 ± 57.1 RT 144.2 ± 72.8 150.0 ± 81.1 TT 115.2 ± 44.8 98.83 ± 20.8 Abbreviations: NT, no tillage; RT, reduced tillage; TT, traditional tillage. MADEJÓN ET AL.147 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License resources (nutrients, water and light) between the crop and weeds. The lower production of wild grasses because of herbicide application, which was more intense in the NT system, likely contributed to improve crop yield by reducing competition for resources. 3.3 |Effects of rainfall exclusion and tillage on biomass allocation and root mycorrhization Rainfall exclusion and tillage did not have any significant effects on the shoot mass of individual plants (F=0.279, p=.604; F=0.116, p=.891, respectively). Rainfall only had some impact on the patterns of biomass allocation to the roots in triticale plants, especially in the RT system. 3.3.1 | Root biomass fractions Rainfall exclusion had a significant negative effect on total root biomass and primary root biomass (F=5.550, p=.036 and F=6.896, p=.017, respectively) (Figure 3). Considering all the tillage systems, secondary roots were not significantly (p> .05, Ftest) affected by rainfall exclusion. However, when distinguishing between tillage types, there was a clear reduction in the growth of secondary roots under rainfall exclusion in the RT system, with nearly a 50% reduction in biomass. Rainfall exclusion also affected RMF, SMF, and the ratio of roots to shoots (F=26.01, p=.0003, F=19.31, p=.0009, and F=19.31, p=.0009, respectively) (Figure S3). Rainfall exclusion led to a reduction in RMF and the root:shoot ratio, but to an increase in SMF. When considering each tillage system separately, the increase in the shoot mass fraction was particularly high in the RT treatment (Figure S3). 3.3.2 | Mycorrhizal colonisation The traits studied for this purpose were: F%: Frequency of mycorrhiza in the root system, M%: Intensity of the mycorrhizal colonisation in the root system, m%: Intensity of the mycorrhizal colonisation in the FIGURE 2 (a) Grain yield and (b) biomass of wild grasses (weeds) in the three tillage systems and in control and rainfall exclusion plots (mean and SE). Green bars indicate control plots and yellow bars exclusion plots. The asterisk (*) indicates significant differences between the control and exclusion treatments for each tillage system (p< .05, Ftest). FIGURE 3 Biomass of different root fractions of plants (a) total root; (b) primary root; (c) secondary root) growing in the three tillage systems, and in control and rainfall exclusion plots (mean and SD). The asterisk (*) indicates significant differences (p< .05, Ftest) between the control and the rainfall exclusion treatments for each tillage system. DW: Dry weight. 148 MADEJÓN ET AL. 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License root fragments, a%: Arbuscule abundance in mycorrhizal parts of root fragments, v%: Vesicle abundance in mycorrhizal parts of root fragments. The high variability of AMF traits was an important observation in this experiment. Rainfall exclusion treatment had a significant effect (F=13.97, p=.003) on the frequency of mycorrhizal infection in the root system (F%), being slightly higher in the plants under the rainout shelters than in control plots. The interaction between rainfall exclusion and tillage was significant for this variable (F=7.92, p=.008), so in TT drought clearly increased F%, butnotinNTorRT. The effect of tillage was more evident on v%, the vesicle abundance (F=5.79, p=0.049). For this variable, plants in the conservation tillage systems (RT and NT) had higher values compared to the TT system (Figure 4d), which was a common trend for all AMF traits, including the abundance of arbuscule (F=4.64, p=.072). When only considering the control treatment, significantly (p< .05, Ftest) higher values of F% and v% (vesicle abundance) were observed in both conservation tillage systems (NT and RT) in comparison to TT. When considering the drought plots only, significant differences (p< .05, Ftest) among tillage systems were restricted to the abundance of arbuscules (a%) between RT and TT. 3.4 |Effects of rainfall exclusion and tillage on plant ecophysiology and nutrition 3.4.1 | Ecophysiological data The measured photosynthetic variables included maximum rates of photosynthesis (AN max ), intercellular concentration of carbon dioxide (Ci), stomatal conductance (g s ) and transpiration rate (E). The main results indicate a limited effect of rainfall exclusion or the tillage type. When distinguishing among tillage systems, maximum photosynthesis rates (AN_ max ), in the TT system were reduced by 25% because of rainfall exclusion, although because of data dispersion, this effect was not significant (F=2.12, p=.188; Figure 5a). For conservation tillage systems (NT and RT), similar values were recorded between exclusion and control plants. Moreover, AN max values were positively and significantly correlated to stomatal conductance (r=0.556, p< .001, n=18). For stomatal conductance (g s ), there were no differences because of rainfall exclusion or tillage. Intercellular concentration of carbon dioxide (Ci) was not affected by the rainfall exclusion, although there was a trend towards lower values in the control treatment, especially in the TT system, where a reduction of 50% of Ci in control plants was observed in comparison to plants from exclusion treatments. Ci FIGURE 4 Percentages of mycorrhizal colonisation in Triticale roots in the different treatments (mean values ± SE). Data in green indicate control plots and yellow data indicate exclusion plots. For each tillage, significant differences between control and exclusion are marked with an asterisk (p< .05, Ftest). MADEJÓN ET AL.149 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License values were also positively correlated with g s (r=0.575, p< .001, n=18). As with photosynthesis rates, rainfall exclusion led to some reduction in E across all tillage systems, particularly in RT and TT (around 25% and 22% lower, respectively), although this trend was not significant (F=1.04, p=.312). These values were also significantly correlated with g s (r=0.809, p< .001, n=18). FIGURE 5 (a) Maximum rates of photosynthesis (AN max ), (b) stomatal conductance (g s max ), (c) intercellular concentration of carbon dioxide (Ci), (d) and transpiration rate (E) and (e) leaf water potential in the three tillage systems in control (green points) and exclusion rainfall (yellow points) plots (mean and SE). 150 MADEJÓN ET AL. 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License Leaf water potential (Ψ leaves ) values were not significantly affected by rainfall exclusion in any of the tillage systems (F=0.002, p=.968). Surprisingly, in the TT system, water potential values were slightly more negative in control leaves compared to exclusion leaves, despite a higher standard deviation in the exclusion. Overall, Ψ leaves values did not show a consistent response to rainfall exclusion across the tillage systems (Figure 5e). 3.4.2 | Macro and micronutrients in leaves In general, nutrient contents in leavesweremoreinfluencedby rainfall exclusion than by tillage (Table 2). Macronutrients in leaves tended to present higher contents in the control treatment than under rainfall exclusion, with a significant decrease in N contents because of drought (F=9.14, p=.008). Also, there was a trend for reduction in leaf P under rainfall exclusion, although this effect was marginally non-significant (F=3.48, p=.081). In contrast, there was an increase in leaf K in the rainfall exclusion treatment (F=4.55, p=.049). For the studied micronutrients, there were no clear statistical differences because of rainfall exclusion, except for Fe, which presented significantly higher values in control leaves (F=14.27, p=.0018). Finally, we could not find any differences because of the tillage system in the studied macro and micronutrients. The main chemical properties of the soils taken just before the triticale sowing (Table S1)showedanincreaseinECforall three systems under rainfall exclusion, but an increase in N for the NT system only. For all tillage systems, contents of OM and available P were higher in the exclusion treatment than in the control (F=7.341, p=.019; and F=3.818, p=.074, respectively). 4|DISCUSSION 4.1 |Soil water storage Tillage type can affect the dynamics of water infiltration and retention through the soil profile. In our case, soils from NT showed higher water storage than soils from RT or TT. Thisisconsistentwithpreviouswork conducted at the site that showed that conservation tillage methods tend to improve water retention in the soil compared to TT practices (LópezGarrido et al., 2014). In addition, several field studies have reported a positive impact of the NT and RT on soil moisture, which ensures the absorption and utilisation of water by plants (Wang et al., 2007; Zhang et al., 2022). Although RT has also been proposed as a potential strategy to increase SWS in those works, we could not observe this effect in this study. These results could be related to the fact that NT produces sufficient crop residues to provide mulching on the soil surface, reducing losses via evapotranspiration and contributing to water conservation (Kühling et al., 2017), which is more evident at the first cm of soil (in this study SWS was estimated down to 40 cm depth). Moreover, NT can restore macropore connectivity and aggregation, increasing preferential flow from topsoil to depth (Strudley et al., 2008). The rainfall exclusion treatment had a statistically significant impact on SWS on certain dates, especially during the spring season. Although the magnitude of SWS reduction (in relation to control plots) was similar among tillage types, beneath the rainout shelters (exclusion treatment), NT had the highest mean values of SWS. This indicates that NT practices may have compensated for the reduction of rainfall inputs with a higher water retention within the soil. TABLE 2 Concentrations of macro and micronutrients in Triticale leaves according to the two treatments, tillage and rainfall conditions (mean values and SD in brackets). Tillage Rainfall N Ca K Mg Na P S Cu Fe Mn Zn % mg kg 1 NT Control 3.45 0.70 1.96 0.30 0.07 0.27 0.28 7.80 272.1* 57.10 27.55 (0.36) (0.11) (0.13) (0.06) (0.05) (0.02) (0.04) (0.80) (38.9) (5.58) (1.96) Excl. 2.92 0.64 2.25 0.27 0.23 0.24 0.25 7.71 214.1 58.49 27.62 (0.48) (0.08) (0.21) (0.05) (0.33) (0.04) (0.03) (2.71) (26.7) (4.61) (3.31) RT Control 3.15 0.69 2.13 0.29 0.10 0.25 0.25 6.96 247.2 61.87 24.25 (0.42) (0.09) (0.03) (0.05) (0.06) (0.03) (0.05) (0.94) (84.7) (13.37) (7.47) Excl. 2.90 0.66 2.31 0.31 0.06 0.24 0.24 7.07 224.2 56.01 26.15 (0.54) (0.04) (0.35) (0.03) (0.04) (0.03) (0.06) (0.75) (53.4) (10.49) (6.67) TT Control 3.51 0.68 2.17 0.29 0.11 0.28 0.29 7.43 231.0* 58.21 25.55 (0.11) (0.05) (0.12) (0.02) (0.14) (0.03) (0.01) (0.88) (12.4) (8.60) (3.39) Excl. 3.32 0.62 2.19 0.28 0.03 0.26 0.27 8.02 201.2 61.40 30.45 (0.24) (0.10) (0.14) (0.04) (0.03) (0.02) (0.03) (0.77) (16.6) (18.16) (8.82) Note: The asterisk (*) indicates significant differences between the control and the rainfall exclusion treatments for each tillage system (p< .05, Ftest). Abbreviations: NT, no tillage; RT, reduced tillage; TT, traditional tillage. MADEJÓN ET AL.151 17447348, 2025, 2, Downloaded from https://onlinelibrary.wiley.com/doi/10.1111/aab.12947 by Readcube (Labtiva Inc.), Wiley Online Library on [19/03/2025]. See the Terms and Conditions (https://onlinelibrary.wiley.com/terms-and-conditions) on Wiley Online Library for rules of use; OA articles are governed by the applicable Creative Commons License