Introduction

Global warming, a key driver of climate change, contributes to extreme weather events by raising the Earth’s temperature. By 2100, it is projected that global temperatures could increase by 1.4 to 4.4 °C, leading to an up to 40% rise in the annual wettest days across almost all continents1. A single-degree rise in temperature can increase atmospheric moisture content by 6–7%, potentially triggering extreme rainfall events and more severe wet extremes in regions that already receive high rainfall2,3. In India, the frequency of extreme wet years could increase eightfold between 2050 and 2100, compared to an average of five wet years between 1965 and 2015, with expected rainfall increases of 9.7 to 24.3%4. Additionally, seasonal mean precipitation is predicted to rise from 6.2 to 7.3 mm/day, reflecting an intensification of rainfall by 14 to 35% by the end of 2050 and 2100, respectively1,4. Currently, 16% of global cultivated areas are prone to waterlogging, resulting in approximately $1.5 billion in economic losses annually. In the future, high-intensity rainfall is likely to expand the areas affected by waterlogging, including irrigated regions in India, causing significant economic losses for farmers5,6.

Waterlogging occurs when the soil remains saturated for an extended period, exceeding 20% of its field capacity7. This condition significantly hinders oxygen diffusion resulting in rapid decreases in soil redox potential8,9,10. In such energy-deficient conditions, plants temporarily rely on glycolysis and ethanol fermentation for energy11. However, prolonged waterlogging can lead to the accumulation of toxic compounds such as lactate, which severely impairs respiration12. Additionally, soil carbon can be lost via ethanol production and the generation of reactive oxygen species (ROS)11,12. This energy deficit reduces the phosphorylation of aquaporins, thereby lowering the hydraulic conductivity of plant roots and hindering water and nutrient absorption13. The impaired root function triggers partial stomatal closure, reduced CO₂ influx and photosynthesis results in reduced plant growth, ultimately causing significant yield losses14,15. Some plants have developed adaptive mechanisms to cope with waterlogging, such as forming adventitious roots and aerenchyma cells, which facilitate oxygen and nutrient transport to aerial parts with wider variations among species or varieties5,16. However, waterlogging-induced imbalances in assimilate accumulation and translocation can reduce shoot and root dry matter by 75–77% and 64–75%, respectively, in soybeans15. Therefore, studying the impact of waterlogging on major crops is crucial to understanding their responses to changing weather patterns and developing strategies to mitigate yield losses under such conditions.

Soybean, a vital crop, known for its diverse uses in food, feed, industrial applications, and as a future bio-diesel crop17,18. It accounts for approximately half of the global edible oil production and about two-thirds of the world’s protein concentrate for livestock feed19. Globally, soybean is cultivated on 133.8 million hectares, with an annual production of 348.9 mt20. In India, soybean is primarily grown in the central and western states on vertisols, which are prone to waterlogging due to excessive rainfall and low infiltration rates, significantly impacting yields18. Studies reveal that waterlogging is the second most damaging abiotic stress affecting soybean after drought, with potential yield losses of up to 80%, depending on the crop’s phenology19,21,22. Yield reductions are typically ranges from 17 to 43% during the vegetative stage and 50–56% at the reproductive stage depending on the duration and timing of waterlogging in soybean23. Additionally, one of the key factors contributing to yield reduction is a decrease in the number of pods per plant, which can drop by as much as 37% under waterlogged conditions24.

Several studies have explored the response of soybean to waterlogging by examining fixed durations during different crop growth stages or vice versa24,25,26,27. However, there is limited information on the combined impact of variable waterlogging durations across different growth stages in soybean. Additionally, research has shown that a combination of elevated temperature and waterlogging can enhance photosynthesis and dry matter allocation in crops like tomato compared to ambient temperature conditions28. Similarly, elevated temperatures in conjunction with waterlogging have been reported to alleviate adverse effects following stress relief in rice29. Similarly, the adverse impact of elevated temperature, growth stages and waterlogging on maize has been reported by Shao et al.30. Despite these findings in tomato, rice and maize, studies investigating the combined influence of elevated temperature and waterlogging is lacking in one of the important crops like soybean. Moreover, the response of soybean to variable waterlogging durations at different growth stages, particularly under diverse climate conditions, is poorly understood. These interactions encompassing climate regimes, growth stages, and waterlogging durations play a critical role in determining root morphology, growth dynamics, and ultimately crop yield. Therefore, detailed investigations on the combined effects of these factors to mitigate stress impacts and address knowledge gaps for both current and future scenarios are required. In this context, we hypothesize that the impact of waterlogging on soybean is differentially regulated by varying climate regimes, which include changes in temperature, relative humidity, and sunshine hours. To test this hypothesis, the present study was designed for two growing seasons in a pot culture experiment with the following objectives: (i) To investigate the combined effects of climate regimes, growth stages, and waterlogging durations on growth, dry matter partitioning, and root surface architecture and (ii) To estimate yield losses and evaluate stress tolerance indices resulting from the interactions between climate regimes, growth stages, and waterlogging durations in soybean (Fig. 1).

Fig. 1
figure 1

Response of soybean to waterlogging stress at different growth stages under climatic regimes.

Full size image

Material and methods

Study location

The study was conducted at the ICAR-National Institute of Abiotic Stress Management, Baramati, Maharashtra, India (18° 09′ 30.62″ N; 74° 30′ 03.08″ E; mean sea level: 570 m). This location experiences semi-arid climatic conditions typical of the Deccan Plateau region, with an average annual rainfall of 560 mm. The majority of rainfall (70%) occurs during the southwest monsoon season (June–September), followed by 21% during the post-monsoon period (October–December)31. Weather parameters recorded at fortnightly intervals during the crop growth period are presented in Table S1. The average maximum and minimum temperatures during the summer season (CS) were 34.9 °C and 18.5 °C, respectively, whereas during the rainy season (CR), the average maximum and minimum temperatures were 31.2 °C and 20.5 °C, respectively. Bright sunshine hours were also higher during CS (8.3 h day−1) compared to CR (4.6 h day−1). Conversely, CR recorded greater cumulative rainfall (311 mm) and higher average relative humidity, with maximum and minimum values of 86.4% and 53.0%, respectively, compared to CS.

Experimental details and crop management

The study was conducted through two separate pot experiments during 2023: one in the summer season (CS, February to June) and the other in the rainy season (CR, July to November). The experiments were designed to analyze soybean responses to varied durations of waterlogging across different crop growth stages under distinct climatic regimes. The experiment followed a factorial randomized complete block design (FRCBD) with three factors, each replicated seven times (three replications for growth and yield observations, two for destructive sampling viz., dry matter accumulation, root attributes and biochemical and another two as non-treated control; one for destructive sampling and another for recording growth and yield observations). Thus, each replication consists of 21 pots with two plants in each pot (total 42 plants per replication). The first factor comprised two seasons with differing climatic regimes (CS and CR), the second factor included four growth stages (S15: 15 Days After Emergence (DAE), S30: 30 DAE, S45: 45 DAE, and S60: 60 DAE), and the third factor involved five waterlogging durations (D2: 2 days, D4: 4 days, D6: 6 days, D8: 8 days, and D10: 10 days), compared against a control treatment with no waterlogging. The soybean variety NRC-136 (procured from the Indian Institute of Soybean Research, Indore-452 001, Madhya Pradesh) which can withstand higher temperatures and susceptible to waterlogging conditions, suited to both rainy and summer conditions. The seeds were manually sown in pots of 8452 cm3 capacity (calculated using the hollow cone formula; Eq. 1). Each pot was filled with 14 kg of soil, up to 90% of its capacity, and four seeds were dibbled per pot. Regular irrigation at 60% of field capacity was provided to ensure better germination and seedling establishment16. In case of excessive rainfall, the extra water from the pots was immediately drained after cessation of rainfall. The recommended nutrient dose (40:80:25 kg N: P2O5: K2O ha−1) was applied to each pot based on weight. The nutrients were supplied through urea CO(NH2)2, diammonium phosphate (NH4)2HPO4, and potassium chloride (KCl) and thoroughly mixed with the soil before sowing. Thinning was performed 7 DAE, and two healthy plants were retained per pot. To address iron deficiency, a foliar spray of 0.1% chelated iron (12% Fe as Fe–EDTA) was applied at 12 DAE.

$${\text{Soil }}\;{\text{volume}}\;{\text{ of }}\;{\text{pot }}\left( {{\text{cm}}^{3} } \right) = \frac{1}{3}{ } \times \pi{\text{ h }}\left[ {{\text{R}}^{2} + {\text{Rr}} + {\text{h}}^{2} } \right]$$
(1)

where ‘R’ is the radius of topsoil surface, ‘r’ is the radius of a bottom surface, ‘h’ is the height of the filled soil in the pot.

Imposition of waterlogging

Waterlogging was imposed on the plants by placing the filled pots in a constructed concrete tank with a capacity of 38.08 m3 (5.43 m × 5.48 m × 1.28 m) for the specified durations as per the experimental treatments. The water level in the tank was maintained at approximately 2.5 cm above the soil surface. After the designated waterlogging period, the pots were removed from the tank to record various morphological observations at each growth stage.

Root sampling

Destructive root sampling was carried out at each growth stage to study root morphology during both CS and CR. Pots were carefully depotted, and the roots were thoroughly washed to remove adhered soil particles as depicted in Fig. S1. Immediately after root removal, the number and length of aerial roots were recorded. The roots were then cut at the crown portion and stored in polythene bags at − 20 °C in a deep freezer for further analysis. Subsequently, the roots were analyzed using a WinRHIZO root scanner (Regent–STD 1600 + WINRHIZOTM 2013, Regent Instruments, Canada, Quebec). Before scanning, the roots were defrosted and stained with a pinch of potassium permanganate (KMnO4) for 30 min. The stained roots were spread uniformly in a tray filled with clear water to prevent overlapping. Various root morphological traits were measured by analysing the root images, as described by Halli et al.32,33. After scanning, the roots were shade-dried and then oven-dried at 65 °C in a hot air oven to determine their dry weight.

Growth, physiology and yield parameters

Growth parameters, such as plant height, number of trifoliate leaves, leaf area, number of branches, and normalized difference vegetation index (NDVI), were recorded within 30 min of removing the pots from waterlogging stress to minimize the influence of microclimatic changes on the observations. The NDVI was measured 0.5 m above the plant canopy using a handheld GreenSeeker® device (Trimble, USA) as described by Verhulst and Govaerts34 in Eq. (2). Similarly, photosystem-II activity during the post-stress period (10 days after relieving stress) was measured using a high-sensitivity charge-coupled device (CCD) camera, as described by Basavaraj et al.16. The images were analyzed using the FluoroCam software package (FluoroCam 7), and PSII efficiency was calculated using Eq. (3).

$$\text{NDVI }= \frac{(\text{NIR}-\text{IR})}{(\text{NIR}+\text{IR})}$$
(2)

NIR and IR indicate the light reflections at near infra-red and red spectrum regions, respectively.

$$\text{Qmax}=\frac{(\text{Fm}-\text{F}0)}{\text{Fm}}$$
(3)

where Qmax is quantum efficiency of PSII, F0 and Fm are minimum and maximum fluorescence of dark adapted leaves. Fm–F0 indicates the variable fluorescence.

Leaf area and dry matter partitioning were recorded treatment wise from destructive sampling. The leaves were separated from the shoots, and the leaf area was measured using leaf area meter (a LI-COR Li3100C). Subsequently, the separated leaves and stem parts were oven-dried at 65 °C for 72 h to determine the dry weight. Matured pods were harvested by plucking 10 days after physiological maturity, and the yield per plant was recorded. The general growth of soybean plants at 90 DAS is presented in Fig. S2.

Stress indices

To determine the impact of climate regimes and waterlogging stress at different growth stages of soybean, few selected stress indices were calculated such as, Yield Stability Index; (YSI)35, Stress Susceptibility Index (SSI)36, and Tolerance Index (TOL)37. These indices were calculated using the equations listed below.

$$\text{YSI}=\frac{{\text{Y}}_{\text{si}}}{{\text{Y}}_{\text{ci}}}$$
(4)
$$\text{SSI}=\frac{{1- (\text{Y}}_{\text{si}-}{\text{Y}}_{\text{ci}})}{1-(\text{Ys}-\text{Yc})}\times 100$$
(5)
$$\text{TOL }=\text{ Yci}-\text{Ysi}$$
(6)

where Ysi and Yci are the yield of ith treatment under stress and control conditions, respectively. The Ys and Yc are the average yields of all treatments under stress and control conditions respectively.

Data analysis

The data recorded on various parameters of soybean, such as growth, root morphology, yield, and stress indices, were subjected to a normality test. Since the dataset was found to be normally distributed, no transformation was necessary. The data were then analyzed using analysis of variance (ANOVA), considering climate regime or growing season, growth stages, and durations of waterlogging as fixed effects, and replications as random effects in a factorial randomized complete block design (FRCBD) with three replications, using the F-test. Significant differences between treatments were tested using the Duncan Multiple Range Test (α = 0.05). Furthermore, Principal Component Analysis (PCA) and biplots were analyzed using an open access software “GRAPES” to interpret the relationships between key traits, such as root morphology, grain yield, and stress indices, across different treatments to discriminate the associated variability38.

Results

Modifications in root morphology due to waterlogging stress under different climates

The root morphological features of soybean were significantly affected by transient waterlogging stress under both climatic conditions (Tables 1, S2 and Fig. 2). Regardless of crop stages (S) and durations (D), the reduction in root parameters was greater during the summer (CS) compared to the rainy season (CR). The reduction in root length ranged from 5.0–32.6% in summer to 8.9–24.3% in the rainy season. Similarly, reductions in root surface area (2.7–44.0% in CS & 8.4–28.5% in CR), average root diameter (0.7–20.8% in CS & 0.3–23.0% in CR), root volume (2.2–69.7% in CS & 1.8–33.4% in CR), and root dry weight (5.6–57.5% in CS & 0–51.8% in CR) were observed. Furthermore, the soybean plants exhibited greater reductions in root morphological features when subjected to waterlogging at 60 DAE (S60) compared to other stages in both climates (Table 1 and Fig. 2a). Similarly, longer waterlogging durations (D2 to D10) consistently reduced root growth in both CS and CR (Table 1 and Fig. 2b). The maximum reduction in root length was recorded after 10 days of waterlogging (D10) under CS climate (22.9%), compared to CR (20.2%). Additionally, the highest reductions in root surface area (34.0%), root volume (38.0%), root tips (36.3%), and root dry weight (47.7%) occurred in CS. Regarding the interaction effect, growth stages and duration of waterlogging (S × D) significantly influenced root growth in both climates (Tables 1 and S2). The treatment combination S60 × D10 resulted in the maximum reduction in root surface area (44% in CS and 27.4% in CR), root tips (42% in CS and 27.2% in CR), root forks (60.9% in CS and 41.3% in CR), and root volume (69.7% in CS and 33.4% in CR). Meanwhile, the greatest reduction in average root diameter was observed in the combination S60 × D8–10.

Table 1 Percent reduction in root morphological traits of soybean to transient waterlogging under different climatic regimes.
Full size table
Fig. 2
figure 2

Percent reductions in root morphological features under waterlogging (a) at different crop stages, (b) for different durations of waterlogging. *S15: 15 DAE; Days after emergence, S30: 30 DAE, S45: 45 DAE, S60: 60 DAE, D2: 2 DWL; Days of waterlogging, D4: 4 DWL, D6: 6 DWL, D8: 8 DWL, D10: 10 DWL.

Full size image

Soybean plants formed adventitious or aerial roots as an adaptive strategy to waterlogging stress (Table 2). Regardless of the stages and durations of waterlogging, plants grown in summer produced a higher number of longer aerial roots (2.38 and 0.98 cm for the number and length of aerial roots, respectively) compared to those grown in rainy season (CR). Waterlogging at 15 DAE (S15) resulted in a higher number of adventitious roots (6.2 in CS and 5.3 in CR) with greater length (2.5 cm in CS and 1.9 cm in CR), followed by waterlogging at 30 DAE (S30). However, no adventitious roots were formed when plants were exposed to waterlogging at 45 (S45) and 60 DAE (S60) (Table 2). Regarding waterlogging duration, significantly more (5.1 in CS and 4.3 in CR) and longer aerial roots (2.1 cm during.

Table 2 Number of adventitious roots and adventitious root length of soybean to transient waterlogging stress under different climatic regimes.
Full size table

CS and 1.5 cm during CR) were formed at 10 days (D10) in both CS and CR. Furthermore, the interaction between the stage and duration (S × D) influenced adventitious root formation in soybean. The combination of S15 × D10 during the summer (CS) produced the highest number of roots with the greatest length compared to other treatments.

Waterlogging induced changes in plant growth attributes

The soybean growth attributes were significantly influenced by climate regimes, growth stages, durations, and their interactions (Table 3). Across growth stages and durations, the reduction in growth parameters compared to the respective control plants was greatest in CS (summer) compared to CR. Specifically, plant height was reduced by 3.13 cm, the number of trifoliate leaves by 5.30, leaf area by 73.31 cm2 pl–1, the number of branches by 1.44, and NDVI values by 0.12. Among the growth stages, S60 recorded the greatest reduction in the number of trifoliate leaves (12.33 in CS and 7.0 in CR), leaf area (113.20 cm2 pl–1 in CS and 108.9 cm2 pl–1 in CR), and the number of branches (2.8 in CS and 1.5 in CR), followed by 45 DAE (S45). However, the highest reduction in plant height (5.61 cm in CS and 2.78 cm in CR) and NDVI (0.17 in CS and 0.14 in CR) was observed at 45 DAE (S45). The least effect of waterlogging on growth attributes was observed at the initial stage (S15). Similarly, the greatest reduction in growth parameters occurred at D10, while the least reduction was observed at D2. Furthermore, the interaction of S45 × D10 significantly decreased plant height and NDVI (Table 3), whereas the reduction in the number of leaves and branches was greatest with S60 × D10.

Table 3 Reduction in growth attributes of soybean to transient waterlogging under different climatic regimes.
Full size table

Dry matter accumulation in waterlogging stress under different climatic conditions

Waterlogging stress significantly reduced the dry matter accumulation in various parts of the soybean plants in both climates (Table 4). The maximum reduction in leaf weight (0.76 g pl−1), stem weight (0.34 g pl−1), pod weight (0.54 g pl−1), and total dry matter (1.68 g pl−1) immediately after relieving waterlogging stress was observed during summer (CS), compared to rainy season (CR), over their respective controls. The reduction in dry matter accumulation progressively increased up to S45, with the maximum reduction in leaf weight, stem weight, and total dry matter. A declining trend was observed at S60. Meanwhile, waterlogging at later growth stages drastically reduced pod weight compared to the control. Similarly, longer waterlogging durations intensified the negative impact on total dry matter production, due to cumulative reductions in leaf and stem weight, with the highest reduction observed at D10 (Table 4). The interaction of S × D had a significant negative effect on the allocation of dry matter to different parts of the plant. The greatest reductions in leaf weight (3.51 g pl−1 in CS and 1.53 g pl−1 in CR), stem weight (1.39 g pl−1 in CS and 0.44 g pl−1 in CR), and total dry matter (6.37 g pl−1 in CS and 2.22 g pl−1 in CR) occurred at S45 × D10. Meanwhile, the highest reduction in pod weight (1.22 g pl−1) was observed at S60 × D2.

Table 4 Effect of transient waterlogging on dry matter accumulation of soybean under different climatic regimes.
Full size table

Similarly, dry matter partitioning in different parts of the soybean plant at the harvest stage was significantly influenced by waterlogging stress (Figs. 3a and 4b, and Table 5). Despite a greater reduction in growth attributes, higher dry matter accumulation at the harvest stage was observed in CS compared to CR across all growth stages and durations (Table 5). Regarding growth stages, stem dry weight, pod weight, and total dry matter accumulation were highest in S15 compared to other growth stages across both climate regimes (Figs. 3a and 4a). Similarly, increasing waterlogging durations consistently decreased dry matter accumulation in the leaf, stem, pod, and total dry matter of the plant (Figs. 3b and 4b). Waterlogging for 10 days (D10) resulted in significantly the lowest dry matter in all parts of the plant, while the least reduction in dry matter accumulation was observed with D2 under both CS and CR. Furthermore, the interaction of S15 × D2 resulted in higher total dry matter accumulation in CS, while S60 × D2 was superior for total dry matter accumulation in CR (Table 5).

Fig. 3
figure 3

(a) Soybean dry matter accumulation at different growth stages and (b) soybean dry matter accumulation at different durations of waterlogging at harvest during summer. *S15: 15 DAE; Days after emergence, S30: 30 DAE, S45: 45 DAE, S60: 60 DAE, D2: 2 DWL; Days of waterlogging, D4: 4 DWL, D6: 6 DWL, D8: 8 DWL, D10: 10 DWL, LW: Leaf weight (g pl−1), SW: Stem weight (g pl−1), PW; Pod weight (g pl−1), TDA; Total drymatter accumulation (g pl−1).

Full size image
Fig. 4
figure 4

(a) Soybean dry matter accumulation at different crop growth stages and (b) Soybean dry matter accumulation at harvest under different durations of waterlogging, in rainy season. *S15: 15 DAE; Days after emergence, S30: 30 DAE, S45: 45 DAE, S60: 60 DAE, D2: 2 DWL; Days of waterlogging, D4: 4 DWL, D6: 6 DWL, D8: 8 DWL, D10: 10 DWL, LW: Leaf weight (g plant−1), SW: Stem weight (g plant−1), PW; Pod weight (g plant−1), TDA; Total drymatter accumulation (g plant−1).

Full size image
Table 5 Interaction effect of stages and durations of waterlogging on soybean dry matter accumulation at harvest under different climatic conditions.
Full size table

Combined effect of climate, growing stages, and waterlogging on yield and tolerance indices of soybean

Climatic regimes influenced the yield attributes and stress indices of soybean across growth stages and waterlogging durations (Tables 6, S3 and Figs. 5a,b). The reduction in yield ranged from 13 to 56% during summer (CS) compared to 19–91% in the rainy season (CR) (Table 6). The crop was found to be more sensitive to waterlogging stress in the CR, with lower average seed yield (35.71%) and fewer pods per plant (5.95%) compared to the control. As a result, the crop showed a lower yield stability index (YSI) (0.41), higher stress susceptibility index (SSI) (1.00), and tolerance index (TOL) (3.92) compared to CS (Table S3). The least influence of waterlogging on 100-seed weight and seed yield was recorded at S15 in CS, while the maximum yield reduction was observed at S45. Among the waterlogging durations, an increase in duration (D10) exacerbated soybean yield loss, with a maximum reduction of 68.18% in CR and 47.61% in CS. The least reduction in yield and yield-contributing traits was observed at D2 (Table 6). A similar trend was observed for yield attributes, YSI, and stress tolerance indices (SSI and TOL), indicating that prolonged waterlogging durations negatively impacted soybean performance (Table 6 and Fig. 5b). Regarding the interaction effect, S45 × D10 caused the maximum reduction in grain yield (90.90% in CR and 55.22% in CS), number of pods per plant (88.30% in CR and 64.89% in CS), and number of seeds per pod (35.0% in CR and 22.22% in CS).

Table 6 Transient waterlogging effect on soybean yield under different climatic regimes.
Full size table
Fig. 5
figure 5

Different climatic regimes influenced changes in yield stability, susceptibility and tolerance indices of soybean to waterlogging (a) at different crop growth stages, (b) under varied durations of waterlogging. *S15: 15 DAE; Days after emergence, S30: 30 DAE, S45: 45 DAE, S60: 60 DAE, D2: 2 DWL; Days of waterlogging, D4: 4 DWL, D6: 6 DWL, D8: 8 DWL, D10: 10 DWL.

Full size image

Association between root, growth and yield attributes

The relationship between the reduction in root and growth attributes with yield and stress indices under different factors (C, S, and D) was analyzed using Pearson correlation coefficient analysis. The results revealed that the reduction in root attributes was negatively correlated with seed yield and the yield stability index (YSI) in both climates (data not presented). Additionally, Principal Component Analysis (PCA) showed that among the 10 principal components (PCs), PC1 (50.5%), PC2 (26.9%), and PC3 (10.9%) cumulatively accounted for 88.3% of the total variability (Fig. 6a). The PCA biplot indicated that total dry matter (TDA), number of pods (NP), stress susceptibility index (SSI), and leaf area (LA) made the greatest contribution to PC1, as evidenced by the angle and length of the vector. Meanwhile, the highest contribution to the variability in PC2 came from the reduction in root attributes, including root length, root volume, and root dry weight. The contributions of seed yield (SY) and yield stability index (YSI) to PC1 and PC2 were nearly equal, as displayed by the ~ 45° angle between their vectors (Fig. 6b–c). Notably, SY, NP, and YSI were closely associated with combinations of summer climate during later growth stages and shorter durations of waterlogging (CSS60D2–4). In contrast, reductions in root length, root volume, and root dry weight were strongly associated with summer during later growth stages and longer durations of waterlogging (CSS60D8–10). Leaf area reduction, however, was associated with CSS45D8-10 (Fig. 6c). SSI had a strong negative relationship with the number of pods, seed yield, and YSI, as indicated by the angle between the vectors. Additionally, reductions in root length, root volume, and root dry weight showed a moderate negative association with the production of aerial roots and yield attributes (Fig. 6b–c).

Fig. 6
figure 6

Principle component analysis of root, shoot and yield attributes of waterlogged plants indicated in (a) Scree plot, (b) variables vs pc and (c) biplot in summer season. *SY; Seed yield, NP; Number of pods plant−1, TDM; Total dry matter, YSI; Yield stability index, SSI; Stress susceptibility index, LA; Leaf area, RL; Root length, RV; Root volume, RDW; Root dry weight, AR; Adventitious roots.

Full size image

Discussion

Waterlogging is the second most important abiotic stress affecting soybean, causing yield losses of 40–80%. Furthermore, approximately 1700 mha of agricultural land are affected by waterlogging due to excessive rainfall and poor soil drainage. This intensive rainfall events is expected to increase by 35% by 2050, driven by rising global temperatures39,40. In this context, a well-planned, visionary study was conducted on the globally important soybean crop to understand the combined impact of climate regimes, growth stages, and durations of waterlogging on root morphology, stress recovery, and yield attributes in order to plan mitigation measures.

Root morphology of soybean vs waterlogging stress

Roots are the first plant organs that come into contact with edaphic stress factors and respond by modifying their surface architecture, including length, surface area, volume, number of forks, average diameter, and dry matter. In our study, the combined effect of CS × S60 × D8-10 drastically reduced root growth attributes (Tables 1 and S2). The faster depletion of O2 in warm water (~ average temperature 34.9 °C) due to its higher solubility compared to cold water (~ temperature 31.2 °C) during the summer climate regime (CS) possibly reduced root the morphological parameters over CR (Tables 1 and S2). As a result, oxygen-deficit conditions may have led to a higher accumulation of ethylene and a reduced biosynthesis of auxin (ANTHRANILATE SYNTHASE α1; ASA1, ANTHRANILATE SYNTHASE β1; ABS) and transport (PIN1, PIN2, PIN4, AUX1) during CS. This likely limited the primary root growth, possibly through jasmonic acid signaling, and diverted more energy toward the formation of aerial roots as a survival mechanism41,42,43. The present findings highlight the augmenting effect of higher temperatures combined with waterlogging in reducing root growth attributes. Our results align with those of Zhen et al.29, who found that higher temperatures (35–38 °C) reduced root diameter and outer root thickness in waterlogged conditions compared to moderate temperatures (30–34 °C) in rice.

The response of roots to hypoxic conditions varies with factors such as stress intensity, crop growth stage, and internal adjustments44. In our study, the greatest reductions in root attributes at S60 > S45 (Tables 1 and Fig. 2a) were related to the plant’s prioritization of completing its life cycle, thereby redirecting more resources toward reproductive parts, such as flowers and seeds, at the expense of root growth. This is supported by the reduced number of days to harvest at S60 (94 days during CS and 104 days in CR) and lower stem weight (Table 6, Figs. 3a and 4a). Consequently, no aerial roots were produced during later growth stages of soybean, which may have reduced the O2 supply, increased ROS activity, and resulted in greater root damage. In contrast, the lower reductions in root attributes during the earlier stages (S15 and S30) are likely due to the formation of adventitious roots, which may have supplied necessary O2 and facilitated the absorption of water and nutrients for maintaining growth (Table 2). These results align with the findings of Basavaraj et al.16, who indicated that varieties producing a higher number of aerial roots experienced less reduction in root attributes in cowpea.

The progressive reduction in root morphological features with increased waterlogging durations from 2 to 10 days was attributed to the reduced diffusion of oxygen under hypoxic conditions. Waterlogging for 5 days was found to create anoxic (absence of O2) conditions, leading to the formation of aerial roots to meet the plant’s oxygen demands for survival45,46. Further, hypoxia-induced ROS production and autophagy in root cells may result in accelerated root death and decay. Similar findings in soybean showed increased reductions in primary root length, diameter, volume, and surface area with longer waterlogging durations47,48. Additionally, a reduction in the quantum efficiency of PSII likely lowered photosynthesis and subsequent growth while increasing the activity of energy-driven processes, which might have negatively affected root attributes. As a result, soybean produced a greater number of longer adventitious roots (Table 2), likely due to the expression of 1-aminocyclopropane-1-carboxylic acid49,50. Despite the formation of adventitious roots, the reduction in primary root growth was due to the adventitious roots’ inability to supply sufficient oxygen for the plant’s shoot and root growth. Therefore, root morphology in soybean was found to be more sensitive to increased waterlogging durations, particularly at later growth stages, especially under summer climate conditions.

Growth attributes v/s waterlogging

The amplified negative effect of the summer climate (CS) on the growth attributes of soybean (Table 3) was associated with the combined stresses of waterlogging and higher temperatures. These stresses likely affected lignin biosynthesis, leading to reduced cell elongation and cell wall formation, which in turn hindered plant growth, as observed in maize30. Additionally, higher canopy temperatures during CS may have led to the generation of ROS, causing lipid peroxidation of chloroplasts, which subsequently damaged PSII and reduced NDVI values (0.12). Moreover, poor root development due to waterlogging combined with higher temperatures impaired the transport of endogenous hormones (auxin and cytokinin) to the shoot system and the absorption of nutrients, especially nitrogen, which collectively resulted in poor soybean growth. As a result, the accumulation of dry matter in the leaf, stem, pod and total dry matter immediately after relief from stress was significantly lower (Table 4). It is further speculated that the increased canopy temperature, due to stomatal closure, might have induced ethanol fermentation and accelerated carbohydrate depletion, leading to higher respiration and reduced dry matter accumulation51,52. In this context, similar results have been found in maize, where elevated temperatures combined with waterlogging led to greater reductions in plant height, stem girth, and dry matter accumulation30,53.

Across seasons, the considerable reduction in plant height, NDVI, and other growth attributes due to waterlogging at S45 (Table 3) was primarily attributed to poor root growth, reduced photosynthesis, and nutrient deficiencies, particularly nitrogen (N)19,54. Moreover, S45 coincides with the peak growth stages of soybean, specifically the early reproductive stage (flower development and pod initiation stages) which is an energy-intensive process. Therefore, waterlogging stress severely affected dry matter accumulation and growth attributes at this stage5,26,46. Hypoxic conditions may exacerbate nitrogen loss and reduce its uptake, leading to leaf chlorosis and a decline in overall plant growth, as reflected in the reduction of NDVI and dry matter accumulation in leaves and stems (Tables 3 and 4). In contrast, the least impact of waterlogging on growth and dry matter accumulation at the initial stage (S15) may be due to the lag phase (slow growth) and the formation of aerial roots, which likely helped supply oxygen, nitrogen, and water to the growing plants, in comparison to the later stages of S45 and S60 (Tables 3 and 4).

Increased reductions in growth and dry matter accumulation due to waterlogging durations ranging from 2 to 10 days at different growth stages and climatic regimes in soybean are primarily attributed to reduced root morphological attributes (Tables 1 and Fig. 2b). The poor root characteristics, such as reduced length, area, volume, and dry weight, were unable to meet the increasing water and nutrient demands of the crop, leading to poor dry matter accumulation. It is believed that prolonged stress (8–10 days) increased ROS activity due to higher canopy temperatures, which further exacerbated leaf chlorosis, as evidenced by the reduced NDVI values (Table 3). In addition, the reduced PS II activity (data not presented) and increased energy consumption for growth maintenance collectively decreased the accumulation of photosynthates in the leaf (68.15% during CS and 30.56% during CR) and stem (68.81% during CS and 22.45% during CR) (Table 4). Previous studies have shown that reductions in photosynthesis, stomatal conductance, and chlorophyll content hindered overall growth in soybean23,49. Similar reductions in plant height, number of trifoliate leaves, and shoot dry mass were observed in mung bean with increased waterlogging durations45. Therefore, the maximum reductions in soybean growth attributes were observed with 8 to 10 days of waterlogging at S45 under the summer climate (CS × S45 × D8-10).

Yield and dry matter at harvest versus waterlogging

The crop’s growth during stress and the subsequent post-stress recovery phase collectively determine the final yield. The present study demonstrates the contrasting reductions in yield parameters and seed yield of soybean under the rainy climate regime (CR). Despite less severe reductions in root and shoot growth attributes under CR, seed yield was significantly lower in this climate regime (35.7% lower than under the summer climate, CS), which may be due to slow post-stress recovery. The relatively poor growth recovery during the rainy climate was attributed to fewer bright sunshine hours (BSH; 4.6 h), which affected PSII activity, crop growth recovery, and yield contributing parameters (Tables 6, S1 and Fig. S3a). As a result, plants allocated more dry matter to vegetative parts, such as the stem (15.71% more than in CS), leading to a lower yield (Table 5). The plants were also found to be more susceptible to stress, as indicated by the higher stress susceptibility index (SSI; 1.00) and tolerance index (TOL; 3.92) (Table S3). Similarly, Henshaw et al.47 reported reduced soybean yield due to lower BSH under both waterlogged and control plants. In contrast, the higher number of BSH (8.3 h) and higher temperature (35.3 °C) during the post-stress period in summer likely influenced crop recovery by accelerating growth (105 days) and remobilization of photosynthates to the grains (27.4% higher in CS compared to CR) rather than to vegetative parts. Consequently, the higher evapotranspiration demand, coupled with lower relative humidity, might have enhanced stomatal conductance and photosynthesis, thus improving recovery during the post-stress period. This is evident from greater PSII activity and dry matter accumulation during summer which facilitated better seed filling during summer (Tables 5, S1, 5, and Fig. S3a). Therefore, plants exhibited higher test weight (10.67 g per 100 seeds) and seed yield (4.2 g per plant) under CS (Table 6). A previous study by Matsunami et al.55 emphasized the importance of crop growth rate during the post-stress period in determining soybean yield under waterlogging.

Regarding growth stages, the greater seed yield observed with waterlogging during the post-reproductive phase (S60) was attributed to the effective remobilization of accumulated photosynthates from leaves and stems to the pods, rather than spending extra energy on repairing waterlogging damage. Thus, soybean plants quickly redirected stored photosynthates toward grain filling, as evidenced by the higher reduction in root attributes (Table 1 and Fig. 2a). Meanwhile, the higher yield at the initial stage (S15) is due to the formation of aerial roots, which helped the crop survive during waterlogging. Additionally, the longer recovery period allowed for better formation and filling of reproductive parts (Table 6). Therefore, both the S60 and S15 stages showed a close association with seed yield and its stability index across both climate regimes (Fig. 6c). Notably, the lower seed yields observed during waterlogging at S30 can be attributed to the coincidence of flower initiation, which resulted in higher flower drop and poor pollination, thereby reducing the number of pods per plant. Similarly, the lower yield observed during waterlogging at S45 was likely due to the overlap with the peak flowering to pod formation stage, which reduced pod formation (Table 6). Furthermore, the importance of the number of pods per plant in determining final yield and its stability is reflected in the Principal Component Analysis (PCA), as depicted in Fig. 6c. Additionally, the higher biomass accumulation in vegetative parts, even at harvest, evidently led to lower yields (Figs. 3a and 4a). These results are consistent with the findings of Linkemar et al.24 and report of Ploschuk et al.19.

Regardless of climate regimes and growth stages, increased waterlogging intensity (10 days) exacerbated yield reductions by negatively affecting root architecture, crop physiology, and growth attributes. As a result, plants allocated more energy toward forming aerial roots and combating ROS through antioxidant production (data not presented). Consequently, lower leaf PSII activity (0.082 during CS and 0.085 during CR) was observed even during the post-stress period (Fig. S3c), leading to reduced dry matter accumulation and a lower number of seeds in soybean across both climates (Table 6 and Figs. 3b, 4b). As a result, plants exhibited lower YSI and tolerance across waterlogging durations (D2 to D10) (Fig. 5b). Our findings align with the results of Ploschuk et al.19, who reported that increased waterlogging for 13 days caused yield reductions of 37% to 47% compared to the control. Overall, the interaction of CR × S45 × D8-10 amplified the adverse effects of waterlogging, resulting in greater reductions in yield and attributes in soybean.

Conclusion

The present study highlights the sensitivity of soybean growth stages to waterlogging intensity under two different climate regimes. Flower initiation to pod development stages (S30 and S45) exhibited higher sensitivity to waterlogging. Meanwhile, the formation of aerial roots and the reallocation of reserved dry matter toward grain production resulted in better yields when waterlogging occurred at the early (S15) and later stages (S60), respectively. Additionally, increased waterlogging intensity beyond 6 days adversely reduced grain yield by 50%, due to poor below-ground and above-ground growth of soybean. Notably, the combination of a summer climate regime and waterlogging (above 6 days) amplified reductions in root morphology and growth in soybean, but favoured post-stress recovery and yield due to higher sunshine hours and temperatures compared to the rainy climate. The present findings are in compliance with current climatic conditions as well as projected future conditions (~ 4 °C increase in temperature by 2080–2100). This information could be useful for planning proactive mitigation strategies (e.g., developing climate-smart varieties and agronomic management practices) to counter the detrimental effects of waterlogging. Additionally, it could help farm managers effectively manage waterlogging and assist policymakers in estimating yield losses under increasing temperatures. However, authors suggest that further intensive studies should focus on understanding the post-stress recovery of the crop, as well as management strategies for faster recovery from waterlogging.