Introduction

Ongoing climate changes, most importantly, the change in global temperature and precipitation pattern (IPCC 2018), are likely to affect the structure and functioning of all ecosystems including tropical grasslands (Grimm et al. 2013; Campo 2016; Allen et al. 2017). Climate change is expected to intensify the summer monsoon in tropics (IPCC 2018), particularly leading to the occurrence of extreme events (Greve et al. 2014; Chadwick et al. 2016). Climate change also affects the spread of invasive species in new tropical areas (Thapa et al. 2018), raising enormous concern on the fate of invaded ecosystems (Neena and Joshi 2013; Thapa et al. 2018). Therefore, an understanding of the behaviour of invasive species will help in the framing of new conservation strategies in the case likely future vigorous expansion of the species due to climate change (Hannah et al. 2002).

Species living in a particular environment has well adapted ecophysiological traits (Ackerly et al. 2000), and hence affecting ecosystem processes accordingly. Studies also confirm that the ecosystem processes are closely regulated by ecophysiological traits of the species (Díaz et al. 2004; Singh et al. 2015). Thus, a species needs to have favourable traits to survive in the changing climatic condition. Additionally, it is also well documented that the invasive species perform better in their non-native ranges than native ranges (Crawley 1986; Daehler 2003). Besides, climate change would further affect the adaptive range and resilience power of the invasive species. So, the study of ecophysiological traits in Hyptis suaveolens in the present study is important to reveal whether or not it has the ability to expand in other parts of the world in changing climate condition.

Invasive plants are known to alter the ecosystem properties and processes (Vitousek et al. 2017; Afreen et al. 2017). These effects can result from specific traits which may be new to the recipient community (Strickland et al. 2010); for instance, in ecosystems which lack native nitrogen (N)-fixers, invasion by species having N-fixing ability may alter soil fertility and decomposition rates (Allison and Vitousek 2004). Likewise, the establishment of invasive grass species can enhance fire frequency and accelerate C-cycling rates (Strickland et al. 2010). Invaders may contribute nitrogen-rich litter with higher decomposition rates, increase soil nutrients and change soil pH (Ehrenfeld 2003). Such alterations show noticeable effects on soil nutrient status and consequently on the plant growth. Further, studies reported that invasive species alter the soil water budget due to high water use and growth rate (Cavaleri and Sack 2010). Such alteration in water budget could have negative effects on the native community, leading to an alteration in vital ecosystem processes, such as C, N and water cycling (Obrist et al. 2003; Prater et al. 2006).

Plant invasions in many parts of the dry tropical ecosystem of India have resulted in the replacement of native vegetation with Hyptis suaveolens, Parthenium hysterophorus, Ageratum conyzoides and Lantana camara (Sharma et al. 2005; Raizada et al. 2009; Singh et al. 2011). H. suaveolens is one of the invaders having great consequences, next only to Lantana camara in the Vindhyan forest of India (Sharma et al. 2005; Sharma and Raghubanshi 2007). Therefore, a mechanistic understanding of how precipitation variability affects H. suaveolens physiology and growth is important to adequately understand ecosystem response to climate change. It is well documented that H. suaveolens affects soil property and make it suitable for its growth by feedback mechanisms and this leads to biodiversity loss (Raizada 2006; Afreen et al. 2018). Therefore, a key emerging question is how the changes in precipitation regime and plant invasion will interact in tropical grasslands.

In the past few years, although some precipitation manipulation experiments have been performed globally, the understanding of how the invasive plant will perform in changing climate, particularly in tropical monsoon climate, is limited. Understanding the implications of precipitation variability related to invasion for the monsoonal climate is important since these systems are commonly water-limited, with concentrated seasonal rainfall (80% of the annual rainfall occurring in rainy season) and extended dry periods.

In addition, how H. suaveolens will perform in different precipitation condition is not known. This study addresses the above question with the help of some measurable ecophysiological traits. We also hypothesise that the shift in N-mineralisation activity would help in H. suaveolens vigorous growth in high precipitation condition. Thus, the objective of the study is to see the effect of the change in precipitation magnitude on growth and physiology of the invasive species, H. suaveolens, in dry tropical grassland, and also to investigate the effect of precipitation and invasion on selected soil properties.

Material and methods

Study area

The study was performed in the constructed rainout shelter of the Botanical Garden of the Banaras Hindu University (25° 16′ 3.3″ N and 82° 59′ 22.7″ E), Varanasi, Uttar Pradesh, India. The region experiences tropical monsoonal climate having three seasons, viz., winter (November–February), summer (April–June) and rainy (July–September). Two months, i.e., October and March, are considered transitional between the rainy and winter seasons, and winter and summer seasons, respectively. The annual average rainfall of the region is 1100 mm. The mean monthly temperature ranges from 13 to 35 °C during winter and summer season. The soil is categorized as inceptisol, having deep, pale brown in colour and silty loam texture.

Experimental design

The experimental plots were established in the Botanical Garden of BHU, both in rainout shelter and unsheltered plots (open C) in 2014. The designed rainout shelter is of the static type which is a well-ventilated greenhouse. Three rainout shelters R1, R2 and R3 were established and were treated with different precipitation rates: 1600 mm (60% more rainfall than ambient), 1100 mm (average) and 800 mm (20% lower rainfall than ambient), respectively. In each rainout sheltered and unsheltered plot, six replicate plots of 1 × 1 m were established randomly having (i) three indigenous grassland without invasive species H. suaveolens (NIG) and (ii) three indigenous grassland with invasive species H. suaveolens (IG). In each replicated IG and NIG plot, indigenous grassland patches were transplanted from the adjacent native grassland. H. suaveolens was established in IG plots along with native grassland patch by transplanting 15 seedlings per square meter that is higher than its natural density (8 per m2) (Afreen and Singh 2019). The number of transplanted seedlings was kept higher than that found in the field to permit destructive sampling.

In each rainout shelter, a fixed amount of water was sprinkled throughout the year on a fixed day. Thus, the altered precipitation would mimic the change imposed by climate on grassland and invasion and will also lead to the insight due to climate change on the ecophysiology of invasive plant H. suaveolens. The observations (vegetation and soil sampling) were taken after 1 year of the establishment of the experimental plots in 2015 and 2016.

Rain dose selection

Daily observed past meteorological data (max. and min. rainfall, relative humidity, temperature, and solar radiation) were collected from the Indian Metrological Department (IMD), Department of Agronomy, Institute of Agriculture Sciences, BHU. From these 12 years of rainfall data (2001–2012), maximum (1336.5 mm), average (1100 mm), minimum (670.8 mm) rainfall and distribution of rainfall in each month were calculated. Accordingly, we provided 800 mm, 20% less rainfall than ambient in one rainout shelter as the decrease in rainfall over India was by 20% up to 2010 as per Indian Metrological Department (Rathore et al. 2013). In another rainout shelter, we provided 1600 mm rain, i.e., 60% above the average, as IPCC (2013) report stated that monsoonal rain will increase in the near future over India. These calculations of rainfall were used to prepare 1 year calendar to sprinkle a fixed amount of water in experimental plots on a daily basis. Without disturbing the rainfall pattern, i.e., 80% of rain in four months and the rest of it in the remaining eight months, we manipulated the amount of rainfall as calculated earlier in each experimental plot. Summary of rainfall amount and events in both ambient and rain manipulated study are given in Table 1. Detail methodology and experimental design are given in Afreen and Singh (2019), as the present ms is the part of our extensive work.

Table 1 Summary of rainfall quantity and events occurred in the ambient control condition along with the quantity of rain doses given in rainout sheltered treatment plots (2014–2016)
Full size table

Soil sampling

Soil sampling was done in winter, summer and rainy season of the year 2015 and 2016. For the characterization of soil, samples were taken from each IG and NIG plots in triplicate. Samples were collected from the 0–10 cm depth by using soil corer. To examine the effect of invasion and soil moisture on soil properties, analyses of various parameters related to soil nutrient pool and processes were done. Soil samples were sieved through a 2-mm mesh screen after collection for further analysis.

Chemical analysis of soil

Soil moisture was assessed by the gravimetric method. Water holding capacity and bulk density were measured by the method described by Piper (1944). The pH of the soil was determined by digital pH meter (model 702 SM Titrino, Metrohm ion analysis, Metrohm Ltd., Switzerland). Organic C was estimated by using the method described by Nelson and Sommers (1982). For total soil nitrogen, the Kjeldahl method was adopted (Jackson 1958). Nitrate was estimated by the phenoldisulphonic acid method (Jackson 1958), ammonium by the phenate method (APHA 1985) and microbial biomass measurements by the fumigation-extraction method (Brookes et al. 1985; Vance et al. 1987). N-mineralisation was estimated by buried bag technique (Eno 1960).

Plant analysis

Three individuals of Hyptis were randomly selected from each replicate plot (IG) for the study of physiological and growth parameters in the rainy season of the year 2015 and 2016.

Gas exchange measurement

The measurement was done on the fully grown young healthy uppermost leaf from each individual on a sunny day between 08:00 to 11:00 h. Photosynthetic rate (Aarea), stomatal conductance (gs) and transpirational rate (E) of the plant were evaluated by the LI-6400 gas exchange system (LICOR, Lincoln, NE, USA). The flow rate was kept at 500 μ mol s−1and CO2 concentration of 390 ± 5 μ mol CO2 mol−1. Water use efficiency (WUE) was calculated as the ratio of photosynthetic and transpiration rate (A/E).

Leaf attributes

Following gas exchange measurement, leaves were collected and placed in a plastic bag for further analysis. Leaf area (LA) was determined using a leaf area meter (SYSTRONICS, Leaf Area Meter-211); the leaves were oven-dried at 70 °C for 48 h to obtain the dry mass. Specific leaf area (SLA) (cm2 g−1) was measured per unit mass (Singh et al. 2011; Singh and Singh 2012), and leaf mass per unit area (LMA) was calculated as 1/SLA (Singh et al 2011; Singh and Singh 2012).

Growth measurement

Height (H) of the plant was measured from ground to inflorescence. Whole plants selected for the physiological parameters were harvested and brought to the laboratory. Root length (RL) was measured from the ground to the tip of the taproot. In the laboratory, different parts such as stem, leaves and root were separated and dry weight was recorded.

Biomass partitioning

Leaf area ratio (LAR) was calculated as the ratio of leaf area to plant weight. Leaf mass fraction (LMF) was calculated by dividing dry leaf weight to plant weight. Stem mass fraction (SMF) was calculated as the ratio of dry stem weight to plant weight. The root-shoot ratio was calculated by dividing root weight by the dry shoot (stem + leaves) weight.

Statistical analysis

Multivariate analysis of variance (MANOVA) was used to determine the effect of precipitation on the ecophysiology of Hyptis and soil properties in invaded and uninvaded grassland plots. Post hoc (Tukey) test was applied to differentiate between means. Pearson correlation and regression analysis were done to establish the relationship between variables. Statistical analysis was done with the help of SPSS (SPSS Inc., version 16).

Results

Experimental manipulation of precipitation resulted in a substantial change in the soil moisture content of respective plots (Table 1). Among the soil properties, soil pH and N-mineralisation (N-MIN) differed significantly with the precipitation (Table 1). Other soil physiochemical properties showed no significant difference (Supplementary Table 1). The effects of invasion on the soil moisture and other related properties are easily visible (Fig. 1). Soil moisture, microbial biomass carbon (MBC) and microbial biomass nitrogen (MBN) decreased in invaded plots (IG) compared with uninvaded plots (NIG) (Fig. 1). However, soil pH, total inorganic-N and N-mineralisation showed a significant increase in invaded plots (IG) (Fig. 1).

Fig. 1
figure 1

Soil properties in uninvaded (NIG) and invaded (IG) grassland (A) soil moisture (B) soil pH, (C) total inorganic-N, (D) N-mineralisation, (E) MBC and (F) MBN (data of open C, 8T, 11T and 16T were pooled). Soil moisture and pH differ significantly at P < 0.01; total inorganic-N and N-mineralisation differ significantly at P < 0.05; MBC and MBN showed non-significant difference

Full size image

In this study, most of the parameters tested to see the performance of H. suaveolens under different precipitation regimes in IG plots did not differ significantly (Supplementary Table 2). Among the physiological properties and growth measurements, only photosynthetic rate (Aarea), height (H), diameter (D) and biomass (B) differed significantly with precipitation (Fig. 2). Also, aboveground biomass (AG) of H. suaveolens was found to be more sensitive towards the change in precipitation condition than belowground biomass (BG) as AG increases significantly with the increase in precipitation (Fig. 3). On an average, H. suaveolens BG only increased by 8% in 8T plots compared with 11T and 16T precipitation plots showing its inability to grow under drier condition (Fig. 3).

Fig. 2
figure 2

Effect of precipitation regime on invasive Hyptis suaveolens physiology, biomass and leaf attributes (A) photosynthetic rate (Aarea), (B) plant height, (C) plant diameter, (D) plant biomass, (E) specific leaf mass per unit area (LMA) and (F) leaf area ratio (LAR) (data of 2015–2016 were pooled). Bars affixed with different combinations of letters are significantly different from each other (P < 0.05). Histogram represents mean ± 1SE

Full size image
Fig. 3
figure 3

Proportional biomass (aboveground (AG), belowground (BG) of Hyptis suaveolens in different precipitation regime)

Full size image

Within leaf attributes and biomass partitioning parameters, only LMA and LAR varied significantly with the precipitation, showing the general trend as 8T > 11T > 16T (Fig. 2). In open C plots, H. suaveolens partitioned major biomass in the stem, however, in sheltered 11T plots in leaves (Supplementary Table 2). In 8T and 16T plots, biomass was equally distributed in leaves and stem (Supplementary Table 2).

Correlation among the plant physiological and growth parameters showed that Aarea was significantly and positively related with WUE (r = 0.76, P < 0.01) and SLA (r = 0.41, P < 0.05). H. suaveolens biomass was significantly and positively related with H (r = 0.74, P < 0.01), D (r = 0.82, P < 0.01), RL (r = 0.52, P < 0.01), root weight (RW) (r = 0.51, P < 0.05) and LA (r = 0.67, P < 0.01), and significantly and negatively related with LAR (r = − 0.81, P < 0.01) and LMF (r = − 0.48, P < 0.05).

A significant positive correlation of soil moisture was found with H, D, RL, RW, LA, SLA and B (Supplementary Table 3), and a significant negative correlation was observed with LAR and LMA (Supplementary Table 3). Soil pH was significantly and positively related to Aarea, WUE and B (Supplementary Table 3). TIN was significantly positively related to B (Supplementary Table 3). MBN was significantly positively related to gs and E, and significantly negatively related to WUE (Supplementary Table 3).

In the present study, we found that an increase in the precipitation affects soil moisture of the area which appears to be the prime regulator for the growth of H. suaveolens (Fig. 4; Tables 2 and 3). H. suaveolens biomass was linearly related to soil moisture (Fig. 4). Moreover, when the data of SM, soil pH, inorganic-N pool (NO3 − N + NH4 − N), N-MIN, MBC and MBN were regressed to predict the aboveground H. suaveolens biomass, we found that the growth of the H. suaveolens was primarily regulated by soil moisture (R2 = 0.73), then by pH as the linear combination of SM and soil pH increased the R2 value by 19% (Table 3).

Fig. 4
figure 4

Relationship between soil moisture and Hyptis biomass (n = 4)

Full size image
Table 2 Effect of precipitation and Hyptis suaveolens invasion on soil properties (data for the year 2015–2016 were pooled)
Full size table
Table 3 Regression equations based on inorganic nutrient pools (NO3 − N + NH4 − N), N-mineralisation (N-MIN), microbial biomass (MBC, MBN) and soil moisture (SM) for the predicting Hyptis biomass in invaded grassland
Full size table

Discussion

Climate change causes unusual changes in the pattern of precipitation and temperature which influence the success of the invasion. Studies showed that the change in precipitation is one of the major drivers to trigger invasion providing new niche opportunities (Gill et al. 2018). These factors combine to affect soil processes and properties and alter ecological processes. If high temperature and low moisture condition prevail in the growing season, it may alter the circadian rhythm of plant and microorganisms, leading to further alteration of soil processes and properties (Kim et al. 2009; Xu 2016).

In the present study, the overall decrease in photosynthesis rate, height, diameter and biomass of H. suaveolens was observed in low precipitation condition. This may be because in low moisture condition, plants divert metabolic pathway towards the synthesis of heat shock proteins (HSPs) to check the damage in the photosynthesis apparatus. HSPs are synthesized to overcome the adverse effect of high temperature on vegetative and reproductive growth stages (Wang et al. 2004). Gas exchange was strongly affected by low moisture condition. The photosynthesis rate continuously decreased with increased severity of water stress (Zouaoui et al. 2013). In drought-tolerant plants, WUE increases as reported by Zouaoui et al. (2013) in Ziziphus lotus (L.) growing in arid condition. However, in the present case, WUE, stomatal conductance and transpiration rate were not affected.

Biomass allocation helps plants to adjust to the changing environment. However, in this study, no significant change in biomass partitioning (LMF, SMF and root:shoot) was found in different precipitation regimes. Most of the biomass was allocated in the leaves and stem irrespective of the precipitation quantity received. Moreover, LMA and LAR of H. suaveolens increased in low precipitation condition despite low photosynthesis rate. Higher LMA or low SLA indicates that plants invest more in compounds related to leaf defences (structural component) (Cornelissen et al. 2003) to cope with the adverse condition. High LAR, i.e., leaf area per unit of plant mass in low precipitation, also makes H. suaveolens inefficient to cope with the low soil moisture condition. Increase in LAR was caused by low plant biomass under reducing precipitation.

Larger and faster-growing species such as H. suaveolens have a huge demand for nutrients and water. Low precipitation has an adverse effect on morphology and physiology of H. suaveolens, which showed that the population of H. suaveolens had not evolved drought tolerance to succeed drier conditions. The study showed that plants growing in adverse condition reduce their biomass more strongly (Fernández et al. 2002). Lack of water-conserving traits such as high LMF and thicker leaves, i.e., small LAR, SLA and WUE, in H. suaveolens suggest that it is highly unlikely for invasive H. suaveolens to expand to the drier parts of the landscape. Padalia et al. (2014) in their simulated study have also predicted that H. suaveolens prefers wet and warm areas for growth and spread.

Similar to the present study, others have also reported that invasive plants which almost displaced the native flora and disturbed the nutrient cycling of the region are unable to perform well in low soil moisture condition (Kane et al. 2011; te Beest et al. 2013). Many invasive plant species have rapid growth rates, high photosynthetic rate, leaf areas and water use which may be at a disadvantage if drought conditions recur or persist (Cavaleri and Sack 2010; Diez et al. 2012). This suggests that the species is currently invading habitats that are at the limits of its climatic tolerance, as determined by moisture and temperature (Peterson 2003; Kriticos et al. 2005).

We found that in H. suaveolens, aboveground biomass responds more than the belowground biomass to change in precipitation. Many studies showed that plants increase their root length and biomass if they had to survive low moisture condition (Rolli et al. 2015; Paez-Garcia et al. 2015). The root system architecture is affected by external environmental factors such as soil moisture, soil pH, nutrient content and soil temperature (Paez-Garcia et al. 2015). We found no significant change in the root system architecture or root length of H. suaveolens which shows its sensitivity towards low precipitation, particularly it is an inability to adapt to low precipitation.

Previous studies related to invasion stated that invasive plant affects the biodiversity of an area by modulating inorganic-N pool and N-mineralisation rate (Afreen et al. 2018; Vitousek et al. 2017). In the present study, we also found the same result (Table 2), a significant difference in N-mineralisation rate between low (8T) and high precipitation (11T and 16T) (Table 2). However, there is a complete lack of studies related to the invasiveness of the H. suaveolens due to the changing precipitation regime in natural as well as in the experimental condition. Besides, increase in the biomass of H. suaveolens with the increase in precipitation (Fig. 4) is in agreement with the modelling study performed by Padalia et al. (2014, 2015).

As it is the first experimental study under rainout shelter in a tropical environment, further studies on invasive species are suggested using various combinations of frequency, and rain doses should be performed to validate the present finding and to be more precise.

Conclusion

High precipitation or soil moisture condition favours the growth of invasive plant H. suaveolens. H. suaveolens mediates soil properties for faster growth in favourable condition, especially related to N-mineralisation, to maintain a constant supply of nutrient. Selected physiological activities of H. suaveolens, especially photosynthetic rate, height, diameter and biomass, increase with the increase in precipitation. In addition, no change in biomass partitioning parameters was found in H. suaveolens experiencing different precipitation conditions. Its below growth biomass increased only by 8% in low precipitation from high, confirming H. suaveolens inability to perform in low precipitation. Thus, the study indicates that the population of H. suaveolens lacks drought tolerance, and it is likely that it might not spread in the areas which are drier either naturally or become dry in the future due to climate change. However, further studies on the species with a different combination of rain doses are required to ascertain this hypothesis.