Coordination of functional traits in legume and non-legume tree species of tropical dry forests in northwestern Mexico

Abstract

Introduction:

Tropical dry forests (TDFs) are characterized by marked seasonality in water availability, which regulates the ecophysiological functioning of plant species and shapes resource acquisition and use strategies. We assessed the coordination of functional traits associated with water, carbon, and nitrogen use in dominant leguminous and non-leguminous tree species of a TDF in northwestern Mexico.

Methods:

Structural, hydraulic, isotopic, and stomatal traits were analyzed, including specific leaf area (SLA), leaf thickness (LT), leaf Nitrogen content (Leaf N), δ¹⁵N, intrinsic water-use efficiency (WUE_i), stomatal conductance (gₛ), stomatal density (SD), and the leaf water potential gradient (ΔΨ_leaf). Spearman correlation analyses, non-parametric tests, and principal component analysis (PCA) were used to evaluate patterns of functional coordination among groups and species.

Results:

Significant functional differences between leguminous and non-leguminous species were detected, particularly in traits associated with foliar resource use and water regulation. Non-leguminous species showed greater functional convergence in traits associated with leaf structure and water regulation, whereas leguminous species occupied a broader region of multivariate trait space, although differences in multivariate dispersion between groups were not statistically significant. The PCA explained 64% of total trait variability and indicated that functional differentiation between legume and non-legume species was primarily associated with foliar resource use and water regulation. Correlations among WUEi, ΔΨ_leaf, LT, and gₛ further indicated coordination among hydraulic, stomatal, and structural traits involved in gas-exchange regulation and maintenance of plant water status.

Discussion:

Functional strategies were not organized into strictly conservative or acquisitive categories, but rather along continuous gradients defined by different combinations of hydraulic, stomatal, structural, and resource-use traits. These findings support the interpretation that coordination among multiple functional traits underlies the diversity of ecological strategies.

1 Introduction

Tropical dry forests (TDFs) are extensive and diverse worldwide, with a seasonality marked by rainfall availability that regulates phenology, plant growth, carbon and water exchange at multiple scales (Miles et al., 2006; Murphy and Lugo, 1986; Sanchez-Azofeifa et al., 2005). The functional dynamics of tropical dry forests, including photosynthesis, evapotranspiration, and primary productivity, depend largely on the frequency, intensity, and seasonal distribution of rainfall (Lasky et al., 2014; Maass and Burgos, 2011; Rojas-Robles et al., 2023). Due to the strong coupling between precipitation and ecosystem functional processes, tropical dry forests (TDFs) are particularly sensitive to alterations in the hydrological cycle. As climate change intensifies variability in water availability, understanding vegetation functional traits becomes essential to elucidate ecosystem responses to hydrological shifts (Allen et al., 2015).

In Mexico, TDFs cover approximately 14% of the national territory and are especially prominent along the Pacific coast (Dirzo et al., 2011; Lopez-Toledo et al., 2023). The TDF is strongly dominated by leguminous and non-leguminous, evergreen and deciduous tree species that coexist and occupy complementary functional niches that influence ecosystem dynamics (Álvarez-Yépiz et al., 2008; Palomo-Kumul et al., 2021). Leguminous species stand out for their potential for biological nitrogen fixation, which can modify N availability and the functional composition of the plant community (Sprent and Gehlot, 2010). By altering soil nitrogen availability, leguminous species can increase ecosystem productivity (direct effect) and, at the same time, indirectly modify soil nitrogen cycling processes, including mineralization, nitrification, and nitrogen losses through changes in microbial communities and litter quality (Menge et al., 2017). Leguminous species tend to have low δ¹⁵N and high leaf nitrogen and carbon content, along with high SLA (Craine et al., 2009; Powers and Tiffin, 2010). However, these traits often vary along a successional gradient in TDF, where acquisitive pioneer species prioritize higher stomatal density and gas exchange rates over the conservative, high water-use efficiency strategies typically observed in late-successional or mature-forest species (Powers and Tiffin, 2010). In contrast, many non-leguminous species tend to be associated with conservative syndromes with functional traits such as lower gₛ and SLA, and greater LT. Furthermore, in non-leguminous species, C/N ratios are high and δ¹⁵N is relatively higher, suggesting tolerance to prolonged droughts and greater dependence on soil fertility, at the cost of lower rates of resource acquisition (Pineda-García et al., 2016; Reich, 2014; Santiago and Wright, 2007).

The theory of functional trait coordination posits that traits do not operate in isolation but co-vary in coherent patterns, organizing species along a continuum of resource-acquisitive and resource-conservative strategies (Reich, 2014; Reich et al., 2003; Wright et al., 2004). This coordination reflects fundamental trade-offs between productivity and persistence, which dictate how plants allocate resources across different organs and environments, and can be categorized into two primary strategies: acquisitive vs. conservative.

An acquisitive species maximizes rapid resource capture (i.e., light, carbon dioxide, nutrients, and water) and tends to maintain high rates of gas exchange and growth, but with lower structural investment leaf area per unit of dry mass, often, lower tolerance to prolonged stress. In terms of functional traits, acquisitive species tend to exhibit high specific leaf area (SLA; leaf area per unit of dry mass), leaf nitrogen content, and rates of photosynthesis (assimilation; A) and stomatal conductance (g_s) often coupled with greater nutrient resorption but lower water-use efficiencies (as WUE_i = A/g_s; Powers and Tiffin, 2010; Wright et al., 2004). In contrast, a conservative species minimizes resource losses and protects its tissues, investing in long-lived, structurally robust leaves characterized by greater leaf thickness (LT) and lower specific leaf area (SLA), resulting in lower photosynthetic capacity (low A), prioritizing persistence under resource-limited conditions (i.e., drought, low fertility) while maintaining moderate growth rates (Gorné et al., 2020).

The coordination of functional traits associated with contrasting resource-use strategies can be summarized along two major axes: (i) a leaf economics spectrum (LES), which organizes the co-variation between SLA/LT/foliar N, gas exchange rates, and leaf longevity; and (ii) a hydraulic continuum (HC), which integrates gₛ, stomatal density (SD), ΔΨ_leaf (the difference in water potential between pre-dawn and midday), WUE_i, and the δ¹³C signal (Choat et al., 2018; Reich, 2014). Together, the two axes (LES and HC) suggest modes of resource-use for predicting WUE_i and allow anticipation of how the plant species coordinate the water-carbon balance through stomatal control of gas exchange, as well as its influence on nitrogen dynamics in the soil–plant system (Penuelas et al., 2008; Yu et al., 2024).

In this work, morphological, stomatal, hydraulic, and isotopic functional traits were measured to describe water- and Nitrogen-use strategies in dominant tree species of the TDF. To describe water relations, we used the difference of pre-dawn and mid-day leaf water potential (ΔΨ_leaf) as it summarizes the magnitude of the daily decline in water status (Pérez-Harguindeguy et al., 2013; Ratzmann et al., 2019). Foliar Carbon-13 isotopic composition (δ¹³C) was used for approximating WUE_i in C3 plants (Farquhar et al., 1989; Seibt et al., 2008), while foliar nitrogen-15 isotopic composition (δ¹⁵N) provides information on sources, fractionations, and the relative availability of Nitrogen in the soil and plant (Craine et al., 2009). To characterize the structural investment of the leaf, SLA, LT and SD were used, and the gₛ, which reflects the degree of stomatal opening and is therefore associated with the balance between CO₂ gain and water loss (Hetherington and Woodward, 2003; Pérez-Harguindeguy et al., 2013). In the present study, we evaluated the coordination of functional traits in dominant leguminous and non-leguminous tree species from a tropical dry forest in northwestern Mexico, with emphasis on their relationships with water-use efficiency and nitrogen acquisition and use strategies. We hypothesize that dominant leguminous and non-leguminous tree species differ in their resource-use strategies, with non-leguminous species tending to exhibit more conservative patterns of trait coordination, whereas leguminous species tend to display more acquisitive patterns of trait coordination associated with water- and nitrogen-use strategies. We further hypothesize that these contrasting resource-use strategies emerge from the coordination of multiple structural, stomatal, water-relation, and isotopic functional traits rather than from individual traits acting independently. Consequently, different species are expected to exhibit distinct patterns of trait coordination that contribute complementarily to ecosystem functioning under seasonal water limitation.

2 Materials and methods

2.1 Study area

The study site is located within the Monte Mojino private reserve, managed by Naturaleza y Cultura Sierra Madre A.C., within the Sierra de Álamos–Río Cuchujaqui Flora and Fauna (APFF) Protection Area of the National Commission for Natural Protected Areas (CONANP) in southern Sonora, Mexico. The study site coordinates are 26.9968 N, −108.7892 W at an elevation ranging of 360 to 390 m above sea level (m a.s.l.). The vegetation type is tropical dry forest, with a mean annual temperature of 24 °C and a mean annual precipitation of 712 mm, more than 70% of which is associated with the North American Monsoon and concentrated between July and September (Rojas-Robles et al., 2023; Verdugo et al., 2024).

2.2 Species selection and sampling design

Species selection was based on the Importance Value Index (IVI), with the aim of including tree species that are structurally and functionally dominant in the tropical dry forest (TDF) of Álamos, Sonora. The IVI provides an ecologically meaningful representation of the plant community by integrating species abundance, dominance, and frequency within the ecosystem (Álvarez-Yépiz et al., 2008). The IVI was calculated from a forest inventory conducted using eight National Forest and Soil Inventory (INFyS)-type clusters following the methodology described by Hollinger (2008). Each cluster covered an area of 1.25 ha and consisted of four circular plots of 0.1 ha (radius = 17.84 m) arranged in an inverted Y configuration. Plot 1 was located at the center of the cluster, whereas plots 2, 3, and 4 were located to the north, southeast, and southwest, respectively. Each plot contained nested subplots of different sizes for vegetation sampling. Within the 1,000 m² subplots, all trees with a diameter at breast height (DBH) ≥ 20 cm were recorded; within the 400 m² subplots, individuals with DBH ≥ 7.5 cm were measured; and within the 80 m² subplots, all individuals with DBH ≥ 1 cm and < 7.5 cm were recorded. Species frequency was calculated based on its presence across the four plots within each cluster, whereas density and frequency were estimated at the cluster level, resulting in n = 8 for the IVI calculation.

Based on the IVI, eight dominant tree species were selected, representing two functional groups: legumes and non-legumes. The legume species included Lysiloma divaricatum, Lysiloma watsonii, Acacia cochliacantha, Brongniartia alamosana, and Haematoxylum brasiletto, whereas the non-legume species included Croton flavescens, Bursera grandiflora, and Bursera laxiflora. Together, these eight species accounted for more than 60% of the community IVI (Table 1). For functional trait characterization, four individuals per species were sampled.

2.3 Leaf functional traits

Functional trait characterization was conducted during the summer growing season (July–September 2023), when the selected tree species exhibited peak physiological activity. To determine specific leaf area (SLA), leaves free of visible damage were collected and immediately scanned fresh using a flatbed scanner against a high-contrast background to improve leaf edge definition. The resulting images were processed in the ImageJ® software, each scan was calibrated using a metric ruler scanned in the same image. Leaf thickness (LT) was indirectly estimated from fresh leaf mass and projected leaf area. Fresh mass was converted to an estimated leaf volume assuming a leaf tissue density of 1 g cm⁻³, following a volumetric approximation. Because leaf tissue density was not measured independently and may vary among species, LT should be interpreted as an indirect structural proxy rather than a direct anatomical measurement. After scanning and weighing, leaves were oven-dried at 65 °C for 48 h, and dry mass was measured on an analytical balance (precision: 0.0001 g). Finally, SLA (cm² g⁻¹) was calculated as the ratio of projected leaf area measured from fresh leaves to leaf dry mass (g).

2.4 Functional traits of the hydraulic continuum (HC)

This subsection primarily describes the functional traits associated with plant water relations and water use strategies within the framework of the Hydraulic Continuum. Additionally, foliar δ¹⁵N and foliar nitrogen concentration (FNC, mg g⁻¹) were included as complementary isotopic indicators of nitrogen acquisition and availability, and were not considered hydraulic traits. After projected leaf area had been measured and dry mass determined for SLA calculation, the same dried leaf material was used for carbon-13 (δ¹³C) and Nitrogen-15 (δ¹⁵N) stable isotope analyses, as well as foliar Nitrogen concentration (FNC, mg g⁻¹). Dried leaves were ground to a fine homogeneous powder using a Retsch MM400 rotary bead mill (Retsch GmbH, Haan, Germany). Approximately 3.5 mg of each sample were weighed and encapsulated in tin capsules for shipment to the University of Wyoming Stable Isotope Facility¹ for isotopic and elemental analysis. Carbon isotope discrimination (Δ¹³C) was calculated from δ¹³C values using Equation 1 (Ma et al., 2023):where Δ¹³C is calculated based on the difference between the isotopic ratio of δ¹³C in leaf tissue (δ product) and that of atmospheric CO₂ (δ substrate), approximately −8‰. The denominator (1 + δ product / 1,000) corrects for the isotopic composition of the plant tissue, normalizing values relative to natural isotopic fractionation. Intrinsic water-use efficiency (WUE_i) was then estimated from Δ¹³C values using Equation 2 (Seibt et al., 2008):

Here, Ca (426 ppm) corresponds to the mean atmospheric CO₂ concentration measured locally at the MX-Aog AmeriFlux site during the study period²; b′ is the isotopic discrimination due to Rubisco activity (~27‰); a is the isotopic discrimination due to diffusion through air (4.4‰); and 1.6 is the ratio of diffusivities of CO₂ and water vapor, adjusting the equation to reflect leaf gas exchange.

An additional set of leaves from all eight species was collected specifically for stomatal density measurements and immediately preserved in FAA solution (formaldehyde–glacial acetic acid–alcohol) in sealed tubes in the field until laboratory analysis. Leaf epidermal impressions were made using clear commercial nail polish applied to the adaxial surface; once dry, the nail polish peels were removed with dissection forceps. Each peel was mounted on a glass slide and examined under a Primo Star optical microscope (Carl Zeiss Microscopy GmbH, Göttingen, Germany) at 40 × magnification. Photographs were taken with an Axiocam ERc 5 s digital camera (Carl Zeiss, Göttingen, Germany) and processed in ImageJ® to quantify stomatal density (SD): the number of stomata per unit area (mm⁻²).

Stomatal conductance (g_s) was measured using a LI-600 N porometer/fluorometer (LI-COR Inc., Lincoln, USA), which provides real time measurements. To maximize stomatal aperture and ensure data representativeness, and were taken between 08:00 and 10:00 h, when light and humidity conditions favor maximum g_s before critical water stress levels develop.

To standardize sampling and reduce within-canopy microclimatic variability, sun-exposed healthy leaves from the outer crown were selected, as these represent the most physiologically active and inter-individually comparable leaf tissue. Immediately after collection, leaves were placed in sealed plastic bags (kept moist and shaded) and stored in a cooler until measurement within the next 2 h. Leaf water potential (Ψ) was subsequently measured using a Scholander-type pressure chamber (PMS-1000, PMS Instrument Co., Oregon, USA). To capture the daily dynamics of plant water status, readings were taken at two key time points: pre-dawn (05:00 h) and midday (12:00 h). The diurnal water potential difference (ΔΨ_leaf, MPa) was calculated using Equation 3:

Pre-dawn values assumes full plant rehydration (minimum stress), while midday values reflect maximum transpirational-water-loss.

Table 2 summarizes the functional traits evaluated in this study, including their ecological significance and corresponding measurement units.

2.5 Statistical analyses

To evaluate functional trait variation between legume and non-legume species groups, Wilcoxon rank-sum tests were performed. Comparisons among individual species were conducted using Kruskal-Wallis tests because most variables did not meet the assumption of normality, as assessed by the Shapiro–Wilk test (p < 0.05). When significant interspecific variation was detected, Dunn’s post hoc test with Bonferroni correction was applied to identify which species pairs differed significantly. Together, these analyses allowed us to identify functional trait differences between legume and non-legume species groups and among individual species. Patterns of trait coordination associated with acquisitive and conservative resource-use strategies were subsequently evaluated using correlation analyses and principal component analysis (PCA).

Spearman correlation analyses were performed separately for each functional group to identify and quantify the direction and strength of relationships among functional traits. This non-parametric test was chosen due to the lack of normality in the data and its ability to detect associations regardless of linearity. Correlation analysis provided a broader picture of how functional traits relate to one another, helping to interpret the observed patterns. Additionally, a principal component analysis (PCA) was performed to reduce data dimensionality and visualize patterns of variation among functional variables. Both the Spearman correlation analyses and the PCA were performed using individual-level observations, with each sampled tree considered an independent observational unit. When different leaves were required for specific measurements (e.g., stomatal density), all samples were collected from the same individual tree under the same sampling protocol, allowing trait values to be associated with the corresponding individual for multivariate analyses. PCA identified the components explaining the greatest proportion of total variance, facilitating species grouping based on functional traits and providing a visual tool for interpreting relationships among evaluated variables. Finally, to visualize and quantify individual-level variation in functional traits between the two functional groups, nine leaf traits were compared among individuals belonging to the non-legume and legume functional groups in the Álamos TDF (Figure 1). In this analysis, each sampled tree was considered an observational unit, whereas interspecific differences were evaluated separately using Kruskal–Wallis tests. For each trait, a set of monotonic transformations was evaluated to approximate normality within each group. Homogeneity of variances was then assessed on the selected transformation using the Brown-Forsythe test (the median-centered version of Levene’s test). The decision rule was as follows: if normality held in both groups and variances were homogeneous, a one-way ANOVA (factor: Group) was applied; if normality held but variances were heterogeneous, Welch’s ANOVA was used because it provides robust inference under unequal variances and unbalanced sample sizes while remaining relatively robust to moderate departures from normality; if normality could not be achieved after transformation, a Mann–Whitney test (Wilcoxon rank-sum test) was applied. This analytical framework was used specifically for comparisons between the two functional groups and therefore differs from the non-parametric analyses used for interspecific comparisons. All tests were two-tailed with α = 0.05. All statistical analyses were conducted in R (R Core Team, n.d., version 4.5.2). To quantify the visual dispersion of legume and non-legume species in the PCA ordination, multivariate dispersion was evaluated using a permutational analysis of multivariate dispersions (PERMDISP; betadisper function, vegan package) based on Euclidean distances calculated from the PCA scores. Differences in the mean distance to the group centroid were assessed using permutation tests (999 permutations).

Figure 1

3 Results

3.1 Differentiation along LES and HC

Functional trait variability among the studied species is presented in Table 3. Traits associated with the leaf economic spectrum (LES) reflected contrasting resource acquisition and conservation strategies across species and functional groups. Wilcoxon tests revealed significant differences between leguminous and non-leguminous species for leaf Nitrogen concentration (N; p = 0.013) and SLA (p = 0.003). In contrast, C/N did not differ significantly between functional groups (p > 0.05), although significant interspecific variation was detected. Specifically, C/N was significantly higher in L. watsonii compared to B. alamosana (p = 0.003), suggesting a more conservative foliar resource-use strategy in L. watsonii. SLA also exhibited significant interspecific differences, with Bursera grandiflora showing significantly higher values than L. divaricatum (p = 0.025). The high SLA values observed in B. grandiflora indicate greater leaf area per unit of dry mass, a characteristic commonly associated with acquisitive resource-use strategies. SD did not differ significantly between functional groups (p > 0.05), but significant differences among species were observed between A. cochliacanthla and B. alamosana (p = 0.040), with A. cochliacanthla presenting higher SD values.

Traits associated with water status regulation, particularly WUE_i, ΔΨ_leaf, and gₛ, also showed contrasting patterns among species and functional groups. Wilcoxon tests revealed significant differences in ΔΨ_leaf between leguminous and non-leguminous species (p < 0.001), whereas WUE_i and gₛ did not differ significantly between functional groups (p > 0.05). Among leguminous species, WUE_i was significantly higher in L. watsonii (180 mmol mol⁻¹) compared to A. cochliacanthla (122.77 mmol⁻² mol⁻¹; p = 0.025), indicating greater intrinsic water-use efficiency in the former species. In addition, L. watsonii differed significantly from B. laxiflora (p = 0.038), which exhibited the lowest WUE_i values among the evaluated species. Significant interspecific differences in ΔΨ_leaf were detected between A. cochliacanthla and B. grandiflora (p = 0.010). A. cochliacanthla exhibited the highest ΔΨ_leaf values (1.59 MPa), indicating a steeper water potential gradient relative to the other study species. gₛ did not differ significantly between functional groups (p > 0.05), although significant interspecific differences were observed between C. flavescens and L. watsonii (p = 0.043), as well as between L. divaricatum and L. watsonii (p = 0.002). L. watsonii exhibited the lowest stomatal conductance values among all evaluated species, suggesting tighter stomatal regulation relative to the other species analyzed.

Comparisons of leaf functional traits between non-leguminous and leguminous tree species are presented in Figure 1. Significant differences between functional groups were observed for SD (F₁,₃₀ = 28.34, p < 0.001), ΔΨ_leaf (F₁,₃₀ = 16.80, p < 0.003), SLA (F₁,₃₀ = 10.87, p < 0.002), leaf N (F₁,₃₀ = 7.51, p < 0.001), and δ¹⁵N (F₁,₃₀ = 7.01, p < 0.01). Non-leguminous species exhibited higher values of SD, SLA, leaf N, and δ¹⁵N, whereas leguminous species showed higher ΔΨ_leaf values, indicating a greater daily leaf water potential gradient and suggesting contrasting strategies in the regulation of leaf water status. In contrast, g_s (F₁,₃₀ = 0.02, p = 0.89), WUE_i (F₁,₃₀ = 0.00, p = 0.95), LT (F₁,₃₀ = 1.36, p = 0.25), and C/N (F₁,₃₀ = 0.46, p = 0.50) did not differ significantly between functional groups.

3.2 Functional traits correlations

The strength and direction of correlations among functional traits are shown in Figure 2. In leguminous species (Figure 2A), foliar Nitrogen was strongly and positively correlated with δ¹⁵N (ρ = 0.99, p < 0.001), whereas C/N showed negative correlations with both foliar Nitrogen (ρ = −0.46, p = 0.039) and δ¹⁵N (ρ = −0.46, p = 0.040). Leaf N was also negatively correlated with leaf thickness (LT; ρ = −0.53, p = 0.017).

Figure 2

Traits associated with water-use regulation exhibited significant coordination among leguminous species. WUE_i was positively correlated with ΔΨ_leaf (ρ = 0.71, p < 0.001), stomatal conductance (gₛ; ρ = 0.58, p = 0.007), and LT (ρ = 0.74, p < 0.001). In addition, ΔΨ_leaf was positively correlated with LT (ρ = 0.60, p = 0.005), while gₛ also showed a positive relationship with LT (ρ = 0.49, p = 0.030). Conversely, C/N was negatively correlated with gₛ (ρ = −0.48, p = 0.033), suggesting coordinated variation between leaf stoichiometry and gas exchange traits.

In non-leguminous species (Figure 2B), foliar Nitrogen was also strongly and positively correlated with δ¹⁵N (ρ = 0.99, p < 0.001). In contrast, C/N showed negative correlations with both foliar Nitrogen (ρ = −0.69, p = 0.013) and δ¹⁵N (ρ = −0.70, p = 0.011). Traits associated with leaf structure and gas exchange also exhibited significant relationships. gₛ was negatively correlated with both LT (ρ = −0.58, p = 0.048) and stomatal density (SD; ρ = −0.65, p = 0.022). In addition, SLA was positively correlated with ΔΨ_leaf (ρ = 0.61, p = 0.034), whereas LT showed a strong positive correlation with SD (ρ = 0.94, p < 0.001).

3.3 Multivariate coordination of leaf economic and hydraulic strategies

Principal Component Analysis (PCA) explained 64.0% of the total variability in the functional traits of the studied species (Figure 3). The first principal component (PC1) accounted for 41.5% of the total variance and was primarily associated with traits related to foliar resource use, particularly SLA, leaf Nitrogen content (Leaf N), δ¹⁵N, and ΔΨ_leaf. The second principal component (PC2) explained 22.5% of the variance and was mainly associated with physiological traits related to gₛ, intrinsic water-use efficiency (WUE_i), and C/N ratio. Variables such as δ¹⁵N and WUE_i showed intermediate loadings across both axes, suggesting that their variation is distributed throughout trait space rather than being associated with a single functional gradient.

Figure 3

The PCA revealed functional differentiation between leguminous and non-leguminous species (Figure 3). Non-leguminous species, particularly B. grandiflora and C. flavescens, PC1 was primarily associated with SLA, Leaf N, and ΔΨ_leaf describing a gradient of functional differentiation related to leaf structure, resource acquisition, and water regulation. Non-leguminous species were associated with trait combinations characterized by higher SLA and foliar nitrogen concentration and lower leaf thickness, whereas leguminous species occupied a broader functional space, indicating greater functional variability that may contribute to their success under the highly variable environmental conditions characteristic of tropical dry forests. PC2 was primarily influenced by C/N ratio, although its contribution to the separation between functional groups was comparatively limited. Along this axis, species such as L. divaricatum and L. watsonii occupied opposite positions, indicating variation in leaf stoichiometry within the same functional group rather than differentiation between groups. Similarly, gₛ was mainly associated with PC2 and contributed primarily to variation within the leguminous species group.

Partial overlap between both groups, represented by species such as B. laxiflora, indicates that some species share similar trait combinations and occupy intermediate positions in multivariate space rather than forming discrete clusters. Collectively, these results suggest that functional differentiation between leguminous and non-leguminous tree species in the Sonoran TDF is primarily associated with traits related to leaf-level water and nutrient use, whereas variation in C/N ratio, gₛ, and leaf thickness (LT) contributes more strongly to within-group heterogeneity.

4 Discussion

4.1 Functional trait coordination in leguminous and non-leguminous trees

This study aimed to characterize and compare the coordination of functional traits associated with water- and nitrogen-use in dominant tree species of the tropical dry forest (TDF) of Northwestern Mexico. We integrated functional traits within the conceptual framework of the leaf economics spectrum (LES) and the hydraulic continuum (HC) (Gorné et al., 2020; Mencuccini et al., 2019). This approach allowed us to evaluate how TDF tree species coordinate ecophysiological traits in response to the marked water seasonality characteristic of the TDF, where water availability occurs in pulses associated with the North American Monsoon season (Verdugo et al., 2024; Verduzco et al., 2015). Our results revealed functional differences between leguminous and non-leguminous tree species, primarily associated with SLA, Leaf N, LT, δ¹⁵N, and ΔΨ_leaf which gave insights on how plant groups fit within the conceptual framework of the LES and HC at the TDF (Lavorel and Garnier, 2002; Reichstein et al., 2014; Violle et al., 2007).

Although functional differentiation between leguminous and non-leguminous species was evident, the groups were not completely segregated in multivariate space (Wright et al., 2021). Partial overlap among species indicates that some trait combinations were shared between groups, whereas differences in key hydraulic, structural, and nutrient-related traits contributed to their functional differentiation. In other words, the ecophysiological functional response of TDF species does not depend solely on their taxonomic or plant functional grouping (i.e., leguminous or non-leguminous), but on different combinations of hydraulic, structural, and nutrient-related traits associated with water, carbon, and nutrient use carbon, consistent with a multidimensional conceptual frameworks of vegetation functional strategies (Choat et al., 2018; Penuelas et al., 2008).

Leguminous species occupied a broader region of the multivariate space than non-leguminous species. Although the mean distance to the group centroid was slightly greater for leguminous species (2.65) than for non-leguminous species (2.15), multivariate dispersion did not differ significantly between groups (PERMDISP, F = 0.684, p = 0.442). Therefore, the broader visual distribution of leguminous species should be interpreted as a tendency rather than as evidence of significantly greater functional variability. In contrast, non-leguminous species showed greater convergence in traits associated with leaf structure and water regulation, particularly SLA, Leaf N, and ΔΨ_leaf. These patterns are consistent with previous studies conducted in TDFs, where seasonal water availability acts as an environmental filter promoting different combinations of functional traits associated with resource acquisition and persistence under drought conditions (Méndez-Alonzo et al., 2012; Palomo-Kumul et al., 2021).

The patterns of functional differentiation observed in our study indicate that species traits do not respond independently to environmental variability, but rather through coordinated combinations of hydraulic, stomatal, and leaf structural traits. In the tropical dry forest of northwestern Mexico, the regulation of gas exchange, maintenance of plant water status, and nutrient acquisition appear to be integrated components of ecophysiological responses to the strong seasonality and pulsed nature of water availability that characterize this ecosystem. The coordination observed among WUE_i, ΔΨ_leaf, stomatal conductance, leaf thickness, and nutrient-related traits is consistent with the trade-offs among hydraulic safety, hydraulic efficiency, and drought avoidance reported for tropical tree species by Santiago et al. (2018). These patterns highlight that multiple functional traits operate as coordinated trait combinations rather than as independent attributes. Likewise, the influence of climatic variability on plant water relations documented in seasonally dry tropical forests of Mexico (Palomo-Kumul et al., 2021) supports the idea that coordinated trait responses are critical for coping with fluctuations in water availability.

Some species may share functional trait combinations across groups while still maintaining their overall association with a particular functional group. This is the case of Bursera laxiflora which remained within the non-leguminous cluster and had high SLA (121.43 cm² g⁻¹), large leaves, and low intrinsic water-use efficiency indicate a greater investment in resource acquisition than is typically expected for strongly conservative species. B. laxiflora showing a pattern of resource acquisition species may be related to the broad ecological distribution of the species across TDF and seasonally arid environments of northwestern Mexico, where persistence under variable water availability likely requires a flexible combination of hydraulic and structural traits (De-Nova et al., 2012; Pineda-García et al., 2016).

4.2 Hydraulic strategies and regulation of gas exchange

The differences observed in hydraulic, stomatal, and structural traits indicate the existence of multiple water regulation strategies among the evaluated species. For example, Lysiloma watsonii combined high intrinsic water-use efficiency with low stomatal conductance, whereas Acacia cochliacantha was characterized by a greater daily water potential gradient and higher stomatal density. These contrasting trait combinations suggest different mechanisms for coping with seasonal water limitation. However, the results show that these strategies are not distributed homogeneously within each functional group, particularly in leguminous species, which exhibited wide variability in their trait combinations.

In general, non-leguminous species showed relatively high values of SLA and stomatal density, as well as greater functional convergence in traits associated with water use (Table 3). High values of SLA and stomatal density suggest that non-leguminous species share coordinated structural and hydraulic traits that contribute to maintaining plant water status under the pronounced seasonality characteristic of tropical dry forests, a pattern consistent with the integration of hydraulic regulation and drought-response mechanisms reported for tropical tree species (Choat et al., 2018; Santiago et al., 2018). In contrast, leguminous species exhibited a wider range of values for stomatal conductance, ΔΨ_leaf, and intrinsic water-use efficiency (Table 3), reflecting greater variability in trait combinations associated with water regulation. This pattern suggests that functional differences between groups do not correspond to discrete categories of resource conservation or acquisition but rather express a continuous gradient of ecophysiological responses shaped by trait coordination and environmental conditions (Umaña et al., 2023).

Leguminous species exhibited substantial variability in the coordination of hydraulic, stomatal, and structural traits, indicating that different combinations of functional attributes can support plant performance under seasonal water limitation. Within this broader pattern, A. cochliacantha stood out by combining relatively low intrinsic water-use efficiency, a pronounced daily water potential gradient, and high stomatal density. These trait combinations suggest a distinct mechanism of water regulation compared with other leguminous species. This pattern indicates that different combinations of functional traits can support plant performance under seasonal water limitation, suggesting the existence of multiple mechanisms of water regulation within this group. This differentiation may be related to its ecological role as a pioneer species commonly associated with disturbed habitats and early successional stages in tropical dry forests (Álvarez-Yépiz et al., 2008). Pioneer species in seasonally dry environments frequently exhibit trait combinations that favor rapid resource acquisition and carbon gain during periods of favorable water availability (Quesada et al., 2009). Consequently, the pattern observed for A. cochliacantha suggests that variation among leguminous species is influenced not only by taxonomy but also by differences in ecological roles during succession.

On the other hand, L. watsonii, a semi-caducifolious species and that is mostly present in mature forest (Álvarez-Yépiz et al., 2008), exhibited the highest WUE_i values (180.69 mmol mol⁻¹) and lower gₛ values (0.05 mmol m⁻² s⁻), indicating a very distinct performance within the legumes group. The combination of reduced stomatal conductance and high intrinsic water-use efficiency suggests a stricter regulation of water loss while maintaining carbon assimilation under water-limited conditions corresponding to a highly conservative functional strategy. Similar patterns have been reported in seasonally dry and drought-prone ecosystems, where species exhibiting tighter stomatal control tend to achieve greater water-use efficiency and enhanced tolerance to prolonged dry periods (Choat et al., 2018; Pineda-García et al., 2016; Santiago et al., 2018). These results highlight the existence of contrasting water-use strategies within leguminous species and reinforce the idea that functional differentiation in TDF is strongly influenced by species-specific responses to water availability. In this context, higher ΔΨ_leaf values in some leguminous species indicate a steeper daily leaf water potential gradient between predawn and midday, possibly associated with differences in hydraulic conductance, tissue capacitance, and stomatal regulation rather than with a greater intrinsic water extraction capacity per se (Choat et al., 2018; Ratzmann et al., 2019). Similarly, the observed relationship between stomatal density and gₛ suggests that the regulation of gas exchange depends on the coordination among multiple traits and not solely on the number of stomata per unit leaf area.

Non-leguminous species, particularly B. grandiflora and C. flavescens, were characterized by relatively high SLA and elevated foliar nitrogen concentrations. These traits are generally associated with increased photosynthetic potential and resource capture during periods of water availability in seasonally dry environments (Falcão et al., 2022; Pineda-García et al., 2016; Song et al., 2016) which may suggest that these traits, despite of being rather acquisitive, are key to define a strategy to capture resources when available while scaping from the extreme conditions imposed by seasonality.

Taken together, our results indicate that the functional differences observed in the TDF emerge from the coordination among multiple hydraulic, stomatal, and structural traits rather than from variation in single traits. This pattern highlights the importance of evaluating trait coordination to understand how species respond to the strong seasonal fluctuations in water availability that characterize tropical dry forests. Recent studies have emphasized that plant ecological strategies emerge from coordinated trait syndromes operating across multiple levels of plant organization, allowing species to balance resource acquisition, water regulation, and stress tolerance under variable environmental conditions (Burton et al., 2020; Fontana et al., 2021; Sanaphre-Villanueva et al., 2022).

4.3 Nitrogen dynamics and functional strategies

Nitrogen dynamics also showed contrasting patterns among species belonging to the two functional groups evaluated. The positive correlation between foliar nitrogen concentration and δ¹⁵N observed in both leguminous and non-leguminous species suggests a close association between nitrogen availability and the processes of nitrogen acquisition and assimilation (Craine et al., 2009). The capacity for biological nitrogen fixation characteristic of several leguminous species can influence ecosystem nitrogen availability and contribute to nutrient cycling processes, potentially affecting community composition and ecosystem functioning (Caldera et al., 2023; Menge et al., 2017). However, our results suggest that leguminous species do not exhibit a uniform pattern of nitrogen use. Although positive relationships between Leaf N and δ¹⁵N were observed across both functional groups, leguminous species displayed greater variation in both traits, suggesting differences in nitrogen acquisition and assimilation among species. This variability may reflect contrasting degrees of dependence on biological nitrogen fixation and soil-derived nitrogen sources within the group (Menge et al., 2017; Sprent and Gehlot, 2010). In contrast, non-leguminous species exhibited more similar values of Leaf N and δ¹⁵N, indicating a more consistent pattern of nitrogen use among species that also suggest non nitrogen fixation capacities of this group.

4.4 Functional implications for ecosystem functioning

Taken together, our results show that the coexistence of species with contrasting combinations of hydraulic, stomatal, and nitrogen-related traits contribute to the functioning of the tropical dry forest under conditions of pronounced water seasonality. The observed coordination among these traits indicates that species employ different ecophysiological strategies to regulate water use, maintain carbon assimilation, and acquire nutrients under fluctuating environmental conditions. These findings highlight the importance of trait coordination in shaping functional diversity and ecosystem functioning in seasonally dry tropical forests. Therefore, rather than evidencing strictly conservative or acquisitive functional groups, our results indicated the existence of continuous gradients of functional coordination, where different species occupy intermediate positions within multivariate space (Díaz et al., 2016; Reich, 2014).

This functional diversity could contribute to ecosystem stability and resilience by enabling complementary ecophysiological responses to variations in water and nutrient availability. In this regard, the high functional variability observed among leguminous species could represent an important source of ecological resilience by expanding the range of physiological responses available under changing environmental conditions (Craine et al., 2002; Mori et al., 2013). Conversely, the greater functional convergence observed among non-leguminous species suggests that these species share a more consistent set of trait combinations associated to maintain water balance and nutrient use under seasonal drought. This pattern indicates that non-leguminous species may contribute to ecosystem functioning through relatively consistent ecophysiological responses to the environmental conditions characteristic of tropical dry forests.

5 Conclusion

This study demonstrates that the coordination of functional traits organizes distinct strategies of water, Carbon, and Nitrogen use in dominant species of the tropical dry forest of northwestern Mexico. While non-leguminous species showed greater functional convergence in traits associated with water regulation and nutrient use, leguminous species exhibited greater functional heterogeneity and dispersion in multivariate space, reflecting a broader diversity of ecophysiological responses to seasonal resource availability. The results suggest that functional differences among species cannot be described by discrete conservative or acquisitive syndromes, but rather by continuous gradients of trait coordination shaped by different combinations of hydraulic, stomatal, and Nitrogen-related traits. In this context, species such as A. cochliacantha and L. watsonii displayed highly contrasting ecophysiological strategies within the same functional group, highlighting the importance of within-group variability in the functional organization of the ecosystem.

The observed patterns were consistent with the leaf economics spectrum and the hydraulic continuum, demonstrating that the functional composition of the community influences the coupling between water, carbon, and nitrogen cycles. Furthermore, the coexistence of multiple functional strategies could contribute to the stability and resilience of canopy functioning in the face of the climatic variability characteristic of tropical dry forests. Overall, the functional trait-based approach provides a mechanistic basis for understanding the ecophysiological functioning of the TDF and establishes for future studies aimed at evaluating ecosystem responses to changing environmental conditions.

References

• 1

  AllenC. D.BreshearsD. D.McDowellN. G. (2015). On underestimation of global vulnerability to tree mortality and forest die-off from hotter drought in the Anthropocene. Ecosphere6, 1–55. doi: 10.1890/ES15-00203.1

  • CrossRef
  • Google Scholar

• 2

  Álvarez-YépizJ. C.Martínez-YrízarA.BúrquezA.LindquistC. (2008). Variation in vegetation structure and soil properties related to land use history of old-growth and secondary tropical dry forests in northwestern Mexico. For. Ecol. Manag.256, 355–366. doi: 10.1016/J.FORECO.2008.04.049

  • CrossRef
  • Google Scholar

• 3

  BurtonJ. I.PerakisS. S.BrooksJ. R.PuettmannK. J. (2020). Trait integration and functional differentiation among co-existing plant species. Am. J. Bot.107, 628–638. doi: 10.1002/AJB2.1451,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 4

  CalderaA.Álvarez-YépizJ. C.YépezE. A.de los Santos-VillalobosS. (2023). Symbiotic bacteria associated with legume tree species in a Mexican tropical dry forest. Braz. J. Bot.46, 445–451. doi: 10.1007/s40415-023-00875-y

  • CrossRef
  • Google Scholar

• 5

  ChoatB.BrodribbT. J.BrodersenC. R.DuursmaR. A.LópezR.MedlynB. E. (2018). Triggers of tree mortality under drought. Nature558, 531–539. doi: 10.1038/S41586-018-0240-X,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 6

  CraineJ. M.ElmoreA. J.AidarM. P. M.BustamanteM.DawsonT. E.HobbieE. A.et al. (2009). Global patterns of foliar nitrogen isotopes and their relationships with climate, mycorrhizal fungi, foliar nutrient concentrations, and nitrogen availability. New Phytol.183, 980–992. doi: 10.1111/J.1469-8137.2009.02917.X,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 7

  CraineJ. M.TilmanD.WedinD.ReichP.TjoelkerM.KnopsJ. (2002). Functional traits, productivity and effects on nitrogen cycling of 33 grassland species. Funct. Ecol.16, 563–574. doi: 10.1046/J.1365-2435.2002.00660.X

  • CrossRef
  • Google Scholar

• 8

  De-NovaJ. A.MedinaR.MonteroJ. C.WeeksA.RosellJ. A.OlsonM. E.et al. (2012). Insights into the historical construction of species-rich Mesoamerican seasonally dry tropical forests: the diversification of Bursera (Burseraceae, Sapindales). New Phytol.193, 276–287. doi: 10.1111/J.1469-8137.2011.03909.X,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 9

  DíazS.KattgeJ.CornelissenJ. H. C.WrightI. J.LavorelS.DrayS.et al. (2016). The global spectrum of plant form and function. Nature529, 167–171. doi: 10.1038/NATURE16489,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 10

  DirzoR.YoungH. S.MooneyH. A.CeballosG. (eds.) (2011). Seasonally Dry Tropical Forests: Ecology and Conservation. Washington, DC: Island Press. doi: 10.5822/978-1-61091-021-7

  • CrossRef
  • Google Scholar

• 11

  FalcãoH. M.MedeirosC. D. B.FonsêcaM. B.do Espírito-SantoM. M.SantosM. G.AlmeidaJ. S. (2022). Variation in the water use and gas exchange of two Brazilian tropical dry forest phytophysiognomies in response to successional stage. J. Arid Environ.206:104831. doi: 10.1016/J.JARIDENV.2022.104831

  • CrossRef
  • Google Scholar

• 12

  FarquharG. D.EhleringerJ. R.HubickK. T. (1989). Carbon Isotope Discrimination and Photosynthesis. Annu. Rev. Plant Biol.40, 503–537. doi: 10.1146/ANNUREV.PP.40.060189.002443

  • CrossRef
  • Google Scholar

• 13

  FontanaS.RasmannS.de BelloF.PomatiF.MorettiM. (2021). Reconciling trait based perspectives along a trait-integration continuum. Ecology102:e03472. doi: 10.1002/ECY.3472,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 14

  GornéL. D.DíazS.MindenV.OnodaY.KramerK.MuirC.et al. (2020). The acquisitive–conservative axis of leaf trait variation emerges even in homogeneous environments. Ann. Bot.129, 709–722. doi: 10.1093/AOB/MCAA198,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 15

  HetheringtonA. M.WoodwardF. I. (2003). The role of stomata in sensing and driving environmental change. Nature424, 901–908. doi: 10.1038/nature01843,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 16

  HollingerD. Y. (2008). “Defining a landscape-scale monitoring tier for the north American carbon program,” in Field Measurements for Forest Carbon Monitoring: A Landscape-Scale Approach, ed. HooverC. M. (New York, NY: Springer Science + Business Media), 3–16.

  • Google Scholar

• 17

  LaskyJ. R.UriarteM.BoukiliV. K.ChazdonR. L. (2014). Trait-mediated assembly processes predict successional changes in community diversity of tropical forests. Proceedings of the National Academy of Sciences of the United States of America, 111, 5616–5621. doi: 10.1073/PNAS.1319342111

  • CrossRef
  • Google Scholar

• 18

  LavorelS.GarnierE. (2002). Predicting changes in community composition and ecosystem functioning from plant traits: revisiting the holy grail. Funct. Ecol.16, 545–556. doi: 10.1046/j.1365-2435.2002.00664.x

  • CrossRef
  • Google Scholar

• 19

  Lopez-ToledoL.Prieto-TorresD. A.BarrosF. D. V.RibeiroN.PenningtonR. T. (2023). Editorial: seasonally dry tropical forests: new insights for their knowledge and conservation. Front. For. Glob. Change6:1350375. doi: 10.3389/ffgc.2023.1350375

  • CrossRef
  • Google Scholar

• 20

  MaW. T.YuY. Z.WangX.GongX. Y. (2023). Estimation of intrinsic water-use efficiency from δ13C signature of C3 leaves: assumptions and uncertainty. Front. Plant Sci.13:1037972. doi: 10.3389/FPLS.2022.1037972,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 21

  MaassM.BurgosA. (2011). Water Dynamics at the Ecosystem Level in Seasonally Dry Tropical Forests (pp. 141–156). doi: 10.5822/978-1-61091-021-7_9

  • CrossRef
  • Google Scholar

• 22

  MencucciniM.RosasT.RowlandL.ChoatB.CornelissenH.JansenS.et al. (2019). Leaf economics and plant hydraulics drive leaf: wood area ratios. New Phytol.224, 1544–1556. doi: 10.1111/NPH.15998,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 23

  Méndez-AlonzoR.PazH.ZuluagaR. C.RosellJ. A.OlsonM. E. (2012). Coordinated evolution of leaf and stem economics in tropical dry forest trees. Ecology93, 2397–2406. doi: 10.1890/11-1213.1,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 24

  MengeD. N. L.BattermanS. A.HedinL. O.LiaoW.PacalaS. W.TaylorB. N. (2017). Why are nitrogen-fixing trees rare at higher compared to lower latitudes?Ecology98, 3127–3140. doi: 10.1002/ECY.2034,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 25

  MilesL.NewtonA. C.DeFriesR. S.RaviliousC.MayI.BlythS.et al. (2006). A global overview of the conservation status of tropical dry forests. J. Biogeogr.33, 491–505. doi: 10.1111/j.1365-2699.2005.01424.x

  • CrossRef
  • Google Scholar

• 26

  MoriA. S.FurukawaT.SasakiT. (2013). Response diversity determines the resilience of ecosystems to environmental change. Biol. Rev. Camb. Philos. Soc.88, 349–364. doi: 10.1111/BRV.12004,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 27

  MurphyP. G.LugoA. E. (1986). Ecology of tropical dry forest. Annu. Rev. Ecol. Syst.17, 67–88. doi: 10.1146/ANNUREV.ES.17.110186.000435,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 28

  Palomo-KumulJ.Valdez-HernándezM.IslebeG. A.Cach-PérezM. J.AndradeJ. L. (2021). El Niño-southern oscillation affects the water relations of tree species in the Yucatan peninsula, Mexico. Sci. Rep.11:10451-. doi: 10.1038/s41598-021-89835-8

  • CrossRef
  • Google Scholar

• 29

  PenuelasJ.SardansJ.OgayaR.EstiarteM. (2008). Nutrient stoichiometric relations and biogeochemical niche in coexisting plant species: effect of simulated climate change. Pol. J. Ecol.56, 613–622.

  • Google Scholar

• 30

  Pérez-HarguindeguyN.DíazS.GarnierE.LavorelS.PoorterH.JaureguiberryP.et al. (2013). New handbook for standardised measurement of plant functional traits worldwide. Aust. J. Bot.61, 167–234. doi: 10.1071/BT12225

  • CrossRef
  • Google Scholar

• 31

  Pineda-GarcíaF.PazH.MeinzerF. C.AngelesG. (2016). Exploiting water versus tolerating drought: water-use strategies of trees in a secondary successional tropical dry forest. Tree Physiol.36, 208–217. doi: 10.1093/TREEPHYS/TPV124,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 32

  PowersJ. S.TiffinP. (2010). Plant functional type classifications in tropical dry forests in Costa Rica: leaf habit versus taxonomic approaches. Funct. Ecol.24, 927–936. doi: 10.1111/j.1365-2435.2010.01701.x

  • CrossRef
  • Google Scholar

• 33

  QuesadaM.Sanchez-AzofeifaG. A.Alvarez-AñorveM.StonerK. E.Avila-CabadillaL.Calvo-AlvaradoJ.et al. (2009). Succession and management of tropical dry forests in the Americas: review and new perspectives. For. Ecol. Manag.258, 1014–1024. doi: 10.1016/J.FORECO.2009.06.023

  • CrossRef
  • Google Scholar

• 34

  RatzmannG.ZakharovaL.TietjenB. (2019). Optimal leaf water status regulation of plants in drylands. Sci. Rep.9:3768. doi: 10.1038/s41598-019-40448-2,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 35

  R Core Team. (n.d.). RA Lenguaje y entorno para computación estadística. Fundación R para computación estadística. - Referencias - Publicaciones de investigación científica. (n.d.).

  • Google Scholar

• 36

  ReichP. B. (2014). The world-wide ‘fast–slow’ plant economics spectrum: a traits manifesto. J. Ecol.102, 275–301. doi: 10.1111/1365-2745.12211

  • CrossRef
  • Google Scholar

• 37

  ReichP. B.WrightI. J.Cavender-BaresJ.CraineJ. M.OleksynJ.WestobyM.et al. (2003). The evolution of plant functional variation: traits, spectra, and strategies. Int. J. Plant Sci.164, S143–S164. doi: 10.1086/374368

  • CrossRef
  • Google Scholar

• 38

  ReichsteinM.BahnM.MahechaM. D.KattgeJ.BaldocchiD. D. (2014). Linking plant and ecosystem functional biogeography. Proc. Natl. Acad. Sci. USA111, 13697–13702. doi: 10.1073/PNAS.1216065111,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 39

  Rojas-RoblesN. E.YépezE. A.Álvarez-YépizJ. C.Sánchez-MejíaZ. M.Garatuza-PayanJ.Rivera-DíazM. A. (2023). Net ecosystem production during secondary ecological succession: lessons from the tropical dry forest. Madera Bosques29:e2912368. doi: 10.21829/MYB.2023.2912368

  • CrossRef
  • Google Scholar

• 40

  Sanaphre-VillanuevaL.Pineda-GarciaF.DattiloW.Pinzon-PerezL. F.Ricano-RochaA.PazH. (2022). Above- and below-ground trait coordination in tree seedlings depend on the most limiting resource: a test comparing a wet and a dry tropical forest in Mexico. PeerJ10:e13458. doi: 10.7717/PEERJ.13458/SUPP-10

  • CrossRef
  • Google Scholar

• 41

  Sanchez-AzofeifaG.QuesadaM.RodríguezJ.NassarJ.StonerK.CastilloA.et al. (2005). Research priorities for Neotropical dry forests1. Biotropica37, 477–485. doi: 10.1046/j.0950-091x.2001.00153.x-i1

  • CrossRef
  • Google Scholar

• 42

  SantiagoL. S.De GuzmanM. E.BaralotoC.VogenbergJ. E.BrodieM.HéraultB.et al. (2018). Coordination and trade-offs among hydraulic safety, efficiency and drought avoidance traits in Amazonian rainforest canopy tree species. New Phytol.218, 1015–1024. doi: 10.1111/NPH.15058,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 43

  SantiagoL. S.WrightS. J. (2007). Leaf functional traits of tropical forest plants in relation to growth form. Funct. Ecol.21, 19–27. doi: 10.1111/J.1365-2435.2006.01218.X

  • CrossRef
  • Google Scholar

• 44

  SeibtU.RajabiA.GriffithsH.BerryJ. A. (2008). Carbon isotopes and water use efficiency: sense and sensitivity. Oecologia155, 441–454. doi: 10.1007/S00442-007-0932-7,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 45

  SongM.DjagbleteyG.NkrumahE. E.HuangM. (2016). Patterns in leaf traits of leguminous and non-leguminous dominant trees along a rainfall gradient in Ghana. J. Plant Ecol.9, 69–76. doi: 10.1093/JPE/RTV038

  • CrossRef
  • Google Scholar

• 46

  SprentJ. I.GehlotH. S. (2010). Nodulated legumes in arid and semi-arid environments: are they important?Plant Ecol. Divers.3, 211–219. doi: 10.1080/17550874.2010.538740

  • CrossRef
  • Google Scholar

• 47

  UmañaM. N.Salgado-NegretB.NordenN.SalinasV.GarzónF.MedinaS. P.et al. (2023). Upscaling the effect of traits in response to drought: the relative importance of safety–efficiency and acquisitive–conservation functional axes. Ecol. Lett.26, 2098–2109. doi: 10.1111/ELE.14328

  • CrossRef
  • Google Scholar

• 48

  VerdugoL.BojórquezA.GalazO.Romo-LeónJ. R.Sánchez-MejíaZ. M.YépezE. A.et al. (2024). Assessing two decades of landscape greenness in relation to temperature and precipitation in a tropical dry forest of northwestern Mexico. Ecol. Indic.166:112369. doi: 10.1016/J.ECOLIND.2024.112369

  • CrossRef
  • Google Scholar

• 49

  VerduzcoV. S.Garatuza-PayánJ.YépezE. A.WattsC. J.RodríguezJ. C.Robles-MoruaA.et al. (2015). Variations of net ecosystem production due to seasonal precipitation differences in a tropical dry forest of Northwest Mexico. J. Geophys. Res. Biogeosci.120, 2081–2094. doi: 10.1002/2015JG003119

  • CrossRef
  • Google Scholar

• 50

  ViolleC.NavasM.-L.VileD.KazakouE.FortunelC.HummelI.et al. (2007). Let the concept of trait be functional!Oikos116, 882–892. doi: 10.1111/j.0030-1299.2007.15559.x

  • CrossRef
  • Google Scholar

• 51

  WrightC. L.de LimaA. L. A.de SouzaE. S.WestJ. B.WilcoxB. P. (2021). Plant functional types broadly describe water use strategies in the Caatinga, a seasonally dry tropical forest in Northeast Brazil. Ecol. Evol.11, 11808–11825. doi: 10.1002/ECE3.7949

  • CrossRef
  • Google Scholar

• 52

  WrightI. J.ReichP. B.WestobyM.AckerlyD. D.BaruchZ.BongersF.et al. (2004). The worldwide leaf economics spectrum. Nature428, 821–827. doi: 10.1038/NATURE02403,

  • Pubmed Abstract
  • CrossRef
  • Google Scholar

• 53

  YuX.LiuY.NiuS.ZhaoW.FuC.ChenZ. (2024). “Structure, functions, and interactions of dryland ecosystems,” in Dryland Social-Ecological Systems in Changing Environments. Eds. B. Fu, M. Stafford-Smith, (Singapore: Springer), 69–107. doi: 10.1007/978-981-99-9375-8_3

  • CrossRef
  • Google Scholar