Plantations on Forest Gaps and Edges Disturbs the Ecology of Threatened Understory Flora: A Case of Critically Endangered Gentiana Kurroo

Background: Within biodiversity hotspots, forest habitats have been disproportionately reduced and conservation efforts are insucient, making restoration ventures extremely important. However, the impacts of restoration efforts on native threatened understory biodiversity residing in forest gaps and open edges have been seldom investigated. The current study attempts to ll this knowledge gap by investigating the ecology of a critically endangered understory plant namely Gentiana kurroo, across different successional stages of forest plantations including, open grassland (OG), 3-5-year-old plantation (YP), 5-15-year-old plantation (OP) and a mature forest (MF). Results: The highest population of the species was observed in OG, followed by YP, OP and MF. The regeneration was blocked by the plantation as evidenced by a ‘poor’ status in MF and OP, ‘fair’ at the YP and was ‘good’ only at OG. With an increase in canopy openness, the number of regenerating individuals increased, indicating the negative effect of the closed canopy on regeneration. The ordination projections show that the species density responds negatively with the stand basal area of the woody layer, litter depth, and tree density, but responds positively with soil organic carbon, available phosphorus, and reduction in the plantation. Conclusion: The present study suggests that plantations are more likely to benet when established on degraded land rather than a replacement of the natural ecosystems, though those systems were initially forested or not. The results have signicant implications in designing forest restoration programs, particularly on sites with a high diversity of threatened understory species. Thus, we propose that the forest restoration projects should be designed in a way that favours sensitive understory species without compromising the supply of timber and other non-timber forest products.


Introduction
Keeping in view the need for a major reduction in global deforestation, to mitigate climate change and reduce biodiversity loss, ecological restoration has been placed at the forefront by declaring the decade of 2021-2030 as 'Decade of Ecosystem Restoration' under the Bonn Challenge declaration (Stanturf and Mansourian 2020). However, as plantations are becoming an abundant land use, their effects on biodiversity are poorly understood and heavily debated among conservationists as to whether these plantations are bene cial or harmful for native ora and fauna (Verheyen et al. 2016;Lindenmayer 2020).
To recognize the effects of plantations on biodiversity, studies have tested different parameters like landuse transition, tree species being planted, effect on nearby vegetation, seed dispersal, livelihood patterns and soil properties (Sandewall et al. 2010;Coll et al. 2011;Ali et al. 2019). Some studies have found that plantations play a vital role in ameliorating forest fragmentation (Silva et al. 2019), maintenance of soil carbon (Gibson 2009), conservation of biodiversity by providing critical habitat for threatened species (Brockerhoff et al. 2008) and enhancing biodiversity at sites where natural forests are rare (Upton et al. 2014). While many others have found plantations causing adverse changes in species diversity, shading regime, micro-climate, productivity, water balance and nutrient cycling (Ouyang et al. 2016;Chazdon et al. 2016Chazdon et al. , 2020. Though plantations established on ecologically degraded land may offer considerable conservation opportunities, there is a concern regarding the effects on the native threatened understory plants dwelling in natural forest gaps, open forest edges and grasslands (Brockerhoff et al. 2008). Frequent plantations within the gaps and open edges have been observed in many Asian countries (Cao et al. 2011;FAO and UNEP 2020). This is especially common in India, where massive plantation has been started under Compensatory Afforestation Fund Management and Planning Authority (CAMPA). The plantations are often being done without scienti c assessments and site validation, making the threatened ground layer plants prone to extinction. Though the ground layer species play an important role in nutrient supply, carbon stock, species diversity, primary production and serve as a habitat and food for other organisms, their existence is outwardly neglected during the plantation programs (Ali et al. 2019). This warrants an in-depth study to look at the effects of these plantations on understory vegetation. However, there is hardly any evidence to document the effects of plantation on threatened understory ora especially in forest gaps and open edges. Against this backdrop, the current study was conducted to evaluate the effects of plantations in forest gaps and edges on the ecology of native underground threatened ora by taking a critically endangered herb namely Gentiana kurroo as a model species.

Study location and species
The study was conducted in Panchari Forest Range in Kashmir India, a part of Northwestern Himalaya.
The region experiences a moist-temperate climate, divided into four seasons including winter (December to February), spring (March to May), summer (June to August) and autumn (September to November). Gentiana kurroo Royle is native to Northwestern Himalaya and has been classi ed as critically endangered by the IUCN (Fig. 1a) (Ved et al. 2015). The species grows in sub-alpine to alpine regions from an elevation of 1500-3500 m asl. It owers from September to November and fruits from October to January.

Sampling design and data collection
To evaluate the effect of plantation on the model species, sites were selected in such a way that comprise all the successional stages of the plantation. This includes a 3-5 year young plantation site (hereafter YP), 10-15 year old plantation site (hereafter OP) and mature forest (hereafter MF) (Fig. 1b-e). A control open site (hereafter OG) was also selected which had no plantation and was purely natural. To have uniform ecological conditions, sites were selected in such a manner that were very close to each other, hence have similar environmental conditions like altitude, temperature and precipitation.
To study community characteristics, at each site a transect of 50 m wide and 50 m long was laid. All woody species having ≥ 1 cm diameter at breast height (dbh) were enumerated and their height and diameter at breast height was measured. In each plot, the number of individuals of G. kurroo was also counted. Community parameters such as frequency, density, importance value index and the basal cover of the species were computed following Misra (1968 Composite soil samples were collected from each site and were subjected to analysis. pH was determined by a digital pH meter in a 1:2.5 w/v suspension of soil in distilled water (Anderson and Ingram 1993). Soil organic carbon (SOC) was determined by the colorimetric method (Anderson and Ingram 1993). A gram of soil was swirled in potassium dichromate and concentrated H 2 SO 4 , followed by cooling and immediate addition of Barium Chloride. The optical density of the supernatant was measured at 600 nm with a spectrophotometer to get organic Carbon. The total Nitrogen was determined by the Kjeldahl method (Allen et al. 1974) by digesting the soil at 350℃ in presence of concentrated H 2 SO 4 and catalyst mixture of K2SO4 and CuSO4, followed by distillation in presence of NaOH and titration with M/140 N HCl using Boric acid indicator. Available phosphorus was determined by the Ammonium-Molybdate blue method by subjecting the soil to 0.5 M Sodium bicarbonate and Ammonium Molybdate solution followed by measuring the optical density of ltrate at 700 nm (Allen et al. 1974).
To analyze the effect of the plantation at different sites, stand data on basal area, density, cover and height were compared through analysis of variance (ANOVA). Linear and polynomial regression models were used to analyze how the population of the species changes at different sites with respect to change in percent canopy openness, woody species density and litter depth. Canonical Correspondence Analysis (CCA) was used to identify the relationships between sites and various community characteristics. The density data of the species was used for the ordination analysis as it captures changes in the number of stems effectively. Soil physicochemical properties, canopy cover, litter depth, basal area and tree height were used as environmental variables to run the analysis.

Stand characteristics
Stand characteristics of four sites showed a notable variation (Table 1). A total of 10 woody species (≥ 1cm dbh) were recorded. The woody species richness ranged from 03 at OG to 07 at the MF site. 40% species were common between MF and OP, 30% between MF and YP, 20% between MF and OG, 40% between OP and YP, 10% between MP and OG and 30% between YP and OG (Fig. 2). The total density of woody species (≥ 1 cm dbh) ranged from 156 individuals ha − 1 at OG to 852 individuals ha − 1 at MF.
Based on density, Cedrus deodara at MF (density = 416 individual's ha − 1 ), Pinus wallichiana at OP and YP (296 and 84 individual's ha − 1 respectively) and Berberis lyceum at OG (96 individual's ha − 1 ) were the dominant species. While moving from the sites with no plantation through young plantations towards the mature forest, an increase in the basal area was observed. The total stand basal area ranged from 0.12 m 2 ha 1 at OG to 76.28 m 2 ha 1 at MF (Table 1). A signi cant difference (p < 0.05) was observed between the sites with respect to both density and basal area.

Soil properties
Soil properties varied between the sites (Fig. 3). The pH ranged from 5.01 at MF to 6.05 at OG and varied signi cantly (p < 0.05) across the sites. OG had the highest content of organic carbon (3.35%) and available phosphorus (5.48 µg g − 1 ) whereas MF had the highest concentration of total nitrogen (0.36%).
Soil organic carbon ranged from 2.23 to 3.52%, total nitrogen from 0.21 to 0.38% and available phosphorus from 3.22 to 5.56 (µg g − 1 ) across all sites. Based on organic carbon and total nitrogen there was no signi cant difference (p > 0.05) between sites, however, the sites differed signi cantly (p < 0.05) with each other based on values of available phosphorus.

Community characteristics of G. kurroo
The community characteristics of G. kurroo are shown in Table 2. The density of the species ranged from 320 individuals ha − 1 at MF to 8532 individuals ha − 1 at OG. The species was frequently and abundantly found in OG while it was very rare in MF (Table 2). A high importance value index was recorded at OG followed by YP, OP and MF. The species covered the largest area (41.6%) in OG, which was lowest in MF (1.22%) ( Table 2). Based on the community characteristics, all the sites showed a signi cant difference from each other. The individual pattern of height class distributions of the species at different sites showed a varied trend (Fig. 4). At all the sites (except MF) an inverse J-shaped curve was observed having many small individuals but a smaller number of big individuals. Of all the sites, OG had the highest proportion of bigger height class individuals (16-20 cm) followed by YP and OP. The individuals in the lowest (< 4 cm) and highest (16-20 cm) classes were completely absent in MF (Fig. 4).  (Fig. 5). Based on the number of seedlings, saplings and adults, the regeneration of G. kurroo was poor at MF and MP, fair at YP and good at OG (Fig. 5). There was a signi cant difference between the sites based on the number of regenerating individuals (seedlings and saplings).
The regeneration status of G. kurroo was negatively correlated with tree density (r = − 0.73) and litter depth (r = − 0.78). With an increase in canopy openness, the number of regenerating individuals increased indicating a negative effect of canopy cover on regeneration (Fig. 5). There was no linear correlation between canopy openness and the number of regenerating individuals. Conversely, a quadratic relationship between the two was apparent (Fig. 6), where test for polynomial trends showed a signi cant quadratic relationship for OG (F = 15.24, P = 0.02), YP (F = 19.61, P = 0.03), OP (F = 60.25, P = 0.01) and MF (F = 1.087, P = 0.001) (Fig. 6).
Woody species composition seems to have exerted a considerable in uence over the population of G. kurroo. Signi cant (p = 0.042) and a negative relation (r= -0.737) was found between the density of G. kurroo with woody species density (Fig. 7). Similar, but the non-signi cant trend was observed between G. kurroo density with litter depth (r= -0.782, p = 0.71) (Fig. 7).
Ordination analysis between different sites depicted the levels of in uence by various variables on the density of G. kurroo. In ordination space, the overlap of the studied species in different plots showed that the density of G. kurroo responds negatively with the stand basal area, litter depth, tree density, canopy cover, tree height and total nitrogen in the soil, but responds positively with soil organic carbon, available phosphorus, and reduction in plantation intensity (Fig. 8). The rst axis of canonical correspondence analysis (CCA) explained 44.26% of data variance, while the second axis explained 36.99%. At 999 permutation value, the Eigenvalue for both the axis was 0.3 but the p-value was higher (0.41) in rst and lower (0.16) in second. The rst axis was mainly related to soil properties, while the second axis was mostly determined by a combination of community characteristics and total nitrogen (Fig. 8).

Discussion
Though plantations might be helpful in ecosystem restoration and biodiversity conservation in certain cases, our study found that the effect of plantations on ecology and habitat suitably of threatened understory plants should not be underestimated. It seems that the forest managers consider forest gaps, open edges, and grasslands as treeless, degraded vegetation stages, which they feel can improve by afforestation programs. The reforestation and/or afforestation programs are generally focused on the economy and restoration of the upper canopy, while the understory diversity is neglected. The results of the present study reveal that the change in land use signi cantly affects the ecology and population biology of the studied species. Signi cant reduction in the density of G. kurroo from OG to MF indicates a negative effect of the upper-story diversity on the species. Plantation overstory structure has been reported to exert a signi cant in uence on understory colonization rates by limiting their seedling density and hampering regeneration (Ali et al. 2019). This suggests that the conversion of natural forest gaps and edges into forested ecosystems would be detrimental for these understory plant species in a long run.
The The total absence of individuals in the highest height classes in MF is a concern because once the mature individuals disappear, it will not only disturb the reproductive cycle of the species but also wipe out the species in a long run. The progressive increase in plantation height and reduction in canopy cover with intensi cation in plantations would not only reduce the light availability for understory species but also change the structure and composition of the understory vegetation. After the plantation establishment, canopy strati cation leads to niche separation of local ora by creating different gradients in the light, which hampers the growth and establishment of shade-intolerant species (Ou et al. 2020).
The present study revealed that increased litter accumulation affects the density and acts as a barrier to the regeneration of the studied species. Studies have established that the presence of a litter layer tends to reduce species diversity of understory vegetation by many mechanisms like the release of allelopathic  (Loydi et al. 2013). Change in the soil properties affects the population and survival of the studied species as evidenced by low population in acidic and nitrogen rich soil and high in carbon rich soil. Studies have found that conifer species especially Pinus and Cedrus result in soil acidi cation, lower soil organic carbon, and decrease nutrient availability, hence affecting the population of ground species (Barbier et al. 2008).

Conclusion
The present study showed that plantations can deteriorate the habitat suitability and ecology of threatened ground ora. Overlooking the remarkable value of understory diversity and community dynamics needs to be reconsidered and a rethinking process should be started by forest managers and policymakers while planning a forest restoration venture. Our ndings propose that forest restoration projects should be designed in a way that favor sensitive species without affecting the supply of timber and other non-timber forest products. Appropriate diversity assessments and site validations are required before the execution of any plantation program. Perhaps the most puzzling question facing large scale plantation is, how to maintain the native understory diversity without compromise on the degree of plantations. Future research is required to understand the mechanisms behind the patterns and dynamics of local understory ora to realize the consequences of the plantation before its execution along with bene ts from the perspective of ecosystem services rather than purely economic.
3 . Wang X, Wang J, Zhang R, Huang Y, Feng S, Ma X, Zhang Y, Sikdar A, Roy R (2018) Allelopathic effects of aqueous leaf extracts from four shrub species on seed germination and initial growth of Amygdalus pedunculata Pall. Forests 9: 711 https://doi:10.3390/f9110711. Figure 1 Photos showing a) G. kurroo, b) open grassland, c) young plantation, d) old plantation and e) mature forest