Journal of Threatened Taxa | www.threatenedtaxa.org | 26 December 2022 | 14(12): 22221–22231

 

 

 

ISSN 0974-7907 (Online) | ISSN 0974-7893 (Print) 

https://doi.org/10.11609/jott.7953.14.12.22221-22231

#7953 | Received 05 April 2022 | Final received 14 November 2022 | Finally accepted 06 December 2022

 

 

 

Species distribution modeling of a cucurbit Herpetospermum darjeelingense in Darjeeling Himalaya, India

 

Debasruti Boral 1  & Saurav Moktan 2

 

1,2 Department of Botany, University of Calcutta, 35, B.C. Road, Kolkata, West Bengal 700019, India.

1 debasruti.boral@gmail.com, 2 smbot@caluniv.ac.in (corresponding author)

 

 

 

Editor: Afroz Alam, Banasthali Vidyapith, Rajasthan, India.         Date of publication: 26 December 2022 (online & print)

 

Citation: Boral, D. & S. Moktan (2022). Species distribution modeling of a cucurbit Herpetospermum darjeelingense in Darjeeling Himalaya, India. Journal of Threatened Taxa 14(12): 22221–22231. https://doi.org/10.11609/jott.7953.14.12.22221-22231

 

Copyright: © Boral & Moktan 2022. Creative Commons Attribution 4.0 International License.  JoTT allows unrestricted use, reproduction, and distribution of this article in any medium by providing adequate credit to the author(s) and the source of publication.

 

Funding: DB is thankful to University of Calcutta for the University Research Fellowship.

 

Competing interests: The authors declare no competing interests.

 

Author details: D. Boral is a research scholar under the guidance of SM. Her research work focuses on the taxonomy and distribution modeling of rare and threatened plants from Darjeeling Himalaya. Dr. S. Moktan is a faculty in the Centre of Advanced Study, Department of Botany, University of Calcutta. Areas of interest in research include taxonomy, ecology, biodiversity and conservation.

 

Author contributions: DB—field data collection, data analysis, writing original draft; SM—conceptualization, reviewing, editing and supervision.

 

Acknowledgements: The first author is thankful to University of Calcutta for financial assistance. The authors are also thankful to the Department of Forests, Government of West Bengal, India for all the necessary permissions.

 

 

Abstract: Herpetospermum darjeelingense (C.B.Clarke) H. Schaef. & S.S. Renner is a rare cucurbit found in Darjeeling, Himalaya. It is known for its use as food and medicine with possible pharmaceutical applications. Here we assess the current and future habitat suitability of H. darjeelingense in the study area using MaxEnt modeling. In order to obtain accurate results for future models, the ensemble method was used. The current suitable habitat covers only 13% of the study area, while the future models for 2050 and 2070 show zero habitat suitability for the species. This strongly indicates a possible local extinction of the species indicating a need for rapid and decisive conservation efforts.

 

Keywords: BioClim, climate change, ecology, elevation, ensemble, habitat suitability, MaxEnt, population, taxonomy, vulnerable.

 

Abbreviations: AUC—Area Under the Curve | CCSM4—Community Climate System Model 4 | CMIP5—Coupled Model Intercomparison Project 5 | GCM—General Circulation Model | GFDL-CM3—Geophysical Fluid Dynamics Laboratory- Climate Model 3 | IPCC—International Panel on Climate Change | LPT—Least Presence Threshold | MIROC5—Model for Interdisciplinary Research on Climate 5 | RCP—Representative Concentration Pathways | ROC—Receiver Operating Curve | SDM—Species Distribution Modeling | SEM—Scanning Electron Microscopy | TSS—True Skill Statistic.

 

 

 

 

INTRODUCTION

 

The Himalaya biodiversity hotspot is one of the 36 currently recognised by CEPF (2021). The eastern region of the hotspot stands out in its global significance as it contains several centres of plant diversity (CEPF 2005). The complex landscape of the region has contributed to its floristic diversity, which includes several threatened plants (Kandel et al. 2019). In particular, the political boundary of India harbours an estimated 5,800 species of plants from the eastern Himalaya (Pande & Arora 2014).

The Darjeeling Himalaya is a part of the extension with its characteristic vegetation & landscape (CEPF 2005). Some of the major threats to this region include rapid urbanisation and climate change (Pandit et al. 2014). The impact of climate change on plants results in changes in phenology (Hart et al. 2014) and geographic ranges (Gómez-Ruiz & Lacher Jr. 2019). A distinctive pattern of upward altitudinal shift is also observed in mountainous regions (Dullinger et al. 2012). Another impact of climate change includes invasion by alien species which are hardier and more competitive (Pandit et al. 2014).

As the effects of climate change become more drastic, there is an urgent need to study consequences for significant species such as H. darjeelingense, which have vulnerable status. SDM functions on the principle of comparing the environmental conditions of the known location of the species to novel climatic conditions (Pearson 2007). Several different algorithms have been developed to model species distribution, such as MaxEnt (Elith et al. 2011), BIOCLIM (Beaumont et al. 2005), and GARP (Peterson et al. 2007). The accuracy of each modeling system is dependent on the sampling size and ecology of the species. Ultimately, species distribution models are an effective tool that can provide focus to possible practical applications (Hernandez et al. 2006). Among these tools, MaxEnt has been used widely for many different species such as Picrorhiza kurroa Royle ex Benth. (Chandra et al. 2021), Podophyllum hexandrum (Royle) T.S. Ying (Banerjee et al. 2017), Rhododendron niveum Hook.f. (Chhetri & Badola 2017) and including vulnerable species such as Ornduffia calthifolia (F.Muell.) Tippery & Les, O. marchantii (Ornduff) Tippery & Les (Ball et al. 2020), and Lavatera acerifolia Cav. (Villa-Machío et al. 2020). MaxEnt uses presence-only data to create a probability map predicting the distribution of a species across a spatial dimension (Elith et al. 2011). Thus, the objectives of the present study were to: i. characterize the taxonomy and habitat ecology of the taxa in Darjeeling Himalaya and, ii. identify current and future potential habitat and environmental variables determining distribution.

 

 

MATERIALS AND METHODS

 

Study Area

The study encompassed the Darjeeling Himalayan region that extends between 27°13’10”–26°27’05’’ N & 88°53’–87° 30’’ E covering an altitudinal range between 130–3,636 m in the lap of the eastern Himalaya hotspot. The region is bordered by Bangladesh to the south-east, Nepal to the west, and Bhutan to the east. The region is also flanked by the state of Sikkim (Figure 1).

As an extended part of the Himalayan hotspot, the region boasts several types of vegetation ranging from tropical to sub-alpine (Das 1995). A combination of topography & climate along with its location makes the region floristically diverse. The region harbours vegetation of Indo-Chinese, Indo-Malaysian, and western Himalayan origin including rare species such as Gastrochilus corymbosus A.P. Das & Chanda, Liparis tigerhillensis A.P. Das & Chanda, Globba teesta S. Nirola & A.P. Das to mention a few (Nirola & Das 2017).

 

The Species

The present study uses MaxEnt to explore the distribution of Herpetospermum darjeelingense (C.B.Clarke) H. Schaef. & S.S. Renner, a member of the family Cucurbitaceae in Darjeeling Himalaya (Image 1). The genus Herpetospermum comprises of four known species found restricted in the Himalaya and southeastern Asia (POWO 2021), of which three are found in the Darjeeling Himalayan region (Renner & Pandey 2013). H. darjeelingense (syn. Edgaria darjeelingensis C.B. Clarke) is one of the species found in the eastern Himalaya (Renner & Pandey 2013). The presence of this species has been recorded in Bhutan (Grierson & Long 1991), southern China, and Nepal (Renner & Pandey 2013). In India, the species is distributed sparsely in the states of Sikkim and Arunachal Pradesh. Threat search classified the species as Vulnerable in 2017 (BGCI 2021).

 

Species Occurrence Data

The occurrence points were gathered through a field study conducted during 2019–2020 within the Darjeeling Himalaya. The coordinate points in the locations were recorded using Garmin eTrex H. The collected coordinates were first converted to decimal degrees and then thinned using spThin package in R in order to remove duplicates and to remove any coordinates with a distance of less than 1 km between them. The resulting 21 coordinates were used for modeling suitable habitat. The taxonomy of the species was studied through the collection of voucher specimens. Pollen grains were collected from the partially opened bud, and the process of acetolysis was followed (Erdtman 1960) and thereby, SEM observations were made. The population of the species was assessed along with its habitat ecology and the associated species.

 

Environmental variables

Elevation data were sourced at 30-arc second (~1 km2) resolution from WorldClim 2.1 (Fick & Hijmans 2017). From this, slope and aspect data were generated using QGIS 3.4 Madeira software in ASCII format. The elevation, slope, and aspect constituted the three topographic predictors used in this paper. The current bioclimatic variables were obtained from WorldClim 2.1 at 30-arc second (~1 km2) resolution (Fick & Hijmans 2017). The future bioclimatic variables were based on CMIP5, obtained from WorldClim 1.4 (Hijmans et al. 2005). The selected dataset were the GCMs (General Circulation Models) GFDL-CM3 (Griffies et al. 2011; Chaturvedi et al. 2012), CCSM4 (Meehl et al. 2012; Purohit & Rawat 2021) and MIROC5 (Watanabe et al. 2010) for years 2050 & 2070 for three different Representative Concentration Pathways (RCPs), RCP 2.6, RCP 4.5, and RCP 8.5. The RCP 2.6, RCP 4.5, and RCP 8.5 represent three different carbon emission levels (IPCC, 2014). All data were trimmed to the appropriate size and converted to ASCII format using QGIS 3.4 Madeira.

 

Modeling Procedure

First, highly correlated variables (variables with Pearson’s coefficient r value > 0.9) were identified and removed using ENM Tools 1.3 (Warren et al. 2010) (Figure 2). The remaining list of environmental variables is given in Table 1. Overall, seven bioclimatic variables and three topographic variables, i.e., elevation, slope, and aspect, were used for modeling. Models were run on MaxEnt ver.3.4.1 (Phillips et al. 2006). As there were merely 21 occurrence points, only linear and quadratic features were applied. Five replicated models were run using the random test percentage of 25% (Srivastava et al. 2018; Qin et al. 2020). For predictions based on future climate, current occurrence data was projected onto future climactic variables. These were from the datasets GFDL-CM3, CCSM4, and MIROC5 for years 2050 & 2070; for RCP 8.5, 4.5 & 2.6. This resulted in 18 different future models to consider. An ensemble approach was applied wherein; the three different models from each GCM for each RCP of a particular year were combined (Araújo & New 2007; Khanum et al. 2013).

 

Model Validation

The area under the curve (AUC) values were used to assess individual models. Along with AUC, models were also appraised by true skill statistic (TSS) values (Allouche et al. 2006). TSS values were calculated for each model iteration with the lowest presence threshold (LPT). The value of LPT is equal to the lowest probability at a species occurrence point. LPT thus excluedes all areas that are at least not as suitable as locations where the species occurred (Pearson et al. 2007).

 

 

RESULTS

 

Taxonomy and Ecology of H. darjeelingense

H. darjeelingense is described as being an annual with a climbing habit, bifid tendrils, deeply cordate-ovate, and unlobed leaves. The leaves were pubescent with undulate and denticulate margin. The plant is dioecious with male flowers being paired. Bracts are absent or inconspicuous. Both male and female flowers have elongated calyx tube, teeth subulate; corolla is rotate, bright yellow, with deep lobes. Male flowers carry three stamens, anthers connate, single-celled. Female flowers are solitary, with ellipsoid ovary, three stigmas. Fruits are broadly fusiform, carrying about three-six seeds. SEM analysis of the pollen grains revealed that they are spherical, triporate, with distinctly spinous exine (Image 1).

Ecologically, the species is found to grow on roadsides, hilly slopes, stream banks, jhoras, and scrubs within an elevation range of around 1,400–2,600 m. The associated species in the niche includes major trees like Magnolia cathcartii (Hook.f. & Thom.) Noot., Symplocos glomerata King ex Clarke, Alnus nepalensis D. Don, and Cryptomeria japonica (Thunb. ex L.f.) D. Don. The associated undershrubs are Tetrastigma serrulatum (Roxb.) Planch., Aconogonon molle (D. Don) Hara, Boehmeria macrophylla Hornem., Yushania maling (Gamble) Majumdar & Karth., Ageratina adenophora (Spreng.) King & Rob, Girardinia diversifolia (Link) Friis, while the ground covers include Galium elegans Wall. ex Roxb., Strobilanthes divaricata (Nees) T. Anders., Persicaria chinensis (L.) H. Gross, Drymaria cordata (L.) Willd. ex Schult., Pouzolzia hirta Blume ex Hassk., Lecanthus peduncularis (Wall. ex Royle) Wedd., and species of Pilea. It is difficult to tally number of individuals of H. darjeelingense as it has climbing/creeping habit and thus in some cases forms dense sprawling clumps. The site characteristics revealed 48% of the population was distributed towards south-east, followed by south-west with 28% and north-east with 24% aspect location. Majority of the populations was distributed on the hilly slope with around 15°─30° inclination followed by roadside while only few populations were distributed at steep habitat.

Reportedly, H. darjeelingense is used both as food (Mueller-Boeker 1993) and as medicine to treat cattle (Shrestha & Khadgi 2019), traditionally among different communities from the Himalayan belt. A recent study also reports the presence of 13 antioxidants from leaf material, indicating the pharmaceutical potential of the species (Chakraborty et al. 2021). The species is classified as Vulnerable (BGCI 2021) regionally in China. However, information regarding its current status in the study area is scant.

 

Habitat Suitability for Present Day

The different variables used for predicting suitable habitat for H. darjeelingense included temperature,precipitation data, altitude, slope, and aspect. The present-day model with the predicted suitable habitat is shown in Figure 3 along with the ROC curve and the jackknife in Figure 4. The current model performed very robustly with the AUC value at 0.986 and the TSS value 0.948. The potential distribution of H. darjeelingense was stretched over an area of 416.25 km2 (13.21%) after application of LPT. The percentage of contribution is highest for the bioclimatic variable mean temperature in the coldest quarter (BIO11) at 61.2 %, followed by precipitation of seasonality (BIO15) at 24.5%, mean diurnal range (BIO02) at 4.4% and precipitation of warmest quarter (BIO18) at 4.4%. The jackknife also reveals that BIO11 is the most important environmental variable while the other influential variable according to the jackkife is precipitation of seasonality (BIO15) (Figure 4b).

 

Response to Variables

The species response curve of H. darjeelingense to each variable is depicted in Figure 5.  The probability of the presence of the species increases with ALT sharply peaking at 2,000 m (Figure 5a) with the range 1,500─3,000 m. The altitude of almost all sample points fell within this range. For aspect, the response increases with an increase in degree (Figure 5b). For BIO02, BIO03, BIO18, response decreases with increase in variable while, the response increases as BIO15 increases. For BIO11, suitable habitat requires a mean temperature ranging from 5°C─12°C in the coldest quarter. For BIO19, suitable habitat required mean precipitation between 40─90 mm for the coldest quarter.

 

Habitat Suitability for Future Models

The six future ensemble models have an AUC value ranging from 0.99─0.985. The TSS value ranges from 0.903─0.944. The highest percentage of contribution is mean temperature in the coldest quarter (BIO11) for all six ensemble models. Similarly the altitude (ALT) has the highest permutation of importance for both the current and future models. The jackknife shows some difference in the results for the future models where ALT has the highest training gain when used in isolation in some models while mean temperature in the coldest quarter (BIO11) has the highest training gain when used in isolation in other models. The prediction accuracy details of the individual models, along with the ensemble models, are given in Table 3. After the LPT value (0.49) was applied for all future models, probable spatial distribution was 0 km2 for all.

 

 

DISCUSSION

 

The present study explores the ecological status and assesses the habitat distribution of H. darjeelingense in current and future climate scenarios. Previous studies on other species have been conducted using MaxEnt, such as  Angelica glauca Kitam. (Singh et al. 2020), Rosa arabica (Crép. ex Boiss.) Déségl. (Abdelaal et al. 2019), Ixora sp. (Banag et al. 2015), Berkheya cuneata (Thunb.) Willd. (Potts et al. 2013), Acer cappadocicum subsp. lobelia (Ten.) A.E. Murray (Sumarga 2011), Pterocarpus santalinus L.f. (Babar et al. 2012), Aglaia bourdillonii Gamble (Irfan-Ullah et al. 2006). MaxEnt has also been used to explore the distribution of endangered species such as Dioscorea sp. (Hills et al. 2019). MaxEnt is one of several modeling algorithms available for species distribution modeling. MaxEnt predicts probable distribution using presence-only data and a set of climatic grids generating output where each grid cell has a value ranging from 0 (least suitable) to 1 (most suitable) (Phillips et al. 2017). MaxEnt is also effective even with small sample sizes making it suitable for studying endangered species (Pearson et al. 2007). Concerning the performance of MaxEnt models, both AUC and TSS values were used. Swets (1988) classified model performance into failing (0.5–0.6), bad (0.6–0.7), reasonable (0.7–0.8), good (0.8–0.9), or great (0.9–1) based on AUC value. Like AUC, TSS also ranges from 0─1, with a higher value indicating a better-performing model (Allouche et al. 2006). The LPT was also used to prevent an over-fitted model. In the current study, only about 13.21% of the total study area was determined to be suitable habitat for H. darjeelingense. The current model was well-performing, with high AUC (0.986) & TSS (0.948) values. 

The IPCC 5th assessment report (IPCC 2014) presents the projected climate in the future driven by anthropogenic carbon emissions. The report highlights the projected scenarios based on the mitigation strategy applied. The RCPs 2.6, 4.5, and 8.5 represents scenarios where either stringent, intermediate or poor implementation of climate strategy occurred. As each GCM is published by separate research groups, it can make modeling future climate change tricky. Hence, the ensemble method as per Khanum et al. (2013) was applied which reduces the ambiguity of using a single GCM. Overall, all future models created using the ensemble method, which combines three different GCMs, show the probable complete disappearance of H. darjeelingense. Hence, no matter the climate change mitigation strategy, it is quite possible that the species under study might disappear from the study area by 2050. In the case of endangered species, a complete disappearance from the local environment can indicate further downstream effects on other plants. It should be noted that the results species distribution models, are based on extrapolation from available data and methods (Elith & Leathwick 2009). However, these models can provide valuable awareness of urgent future steps to be taken for the preservation of the species under study.

 

 

CONCLUSION

 

The present study highlights the probable suitable habitat of the cucurbit Herpetospermum darjeelingense in the future as well as the present day. The taxon that is often found along roadsides and hilly slopes make its current population vulnerable to habitat destruction due to anthropogenic pressure as well as natural catastrophes. This along with climate change can result in the complete disappearance of the species. MaxEnt modeling of the present-day scenario exhibits a narrow habitat range. Furthermore, future models show that regardless of the climate mitigation strategy, the species faces local extinction. Keeping in mind the availability of limited data on distribution coordinates and population status of the taxa including the rarity of the species in the present study, the taxa should be immediately assigned to Endangered in the IUCN Red List. Furthermore, an urgent requirement to investigate active in situ and ex situ conservation strategies through botanical gardens and local nurseries is of the utmost priority at this juncture because the taxon has both traditional and pharmaceutical potential. One possible method can include the collection of seeds for storage and germination.

 

 

Table 1. Site characteristics of Herpetospermum darjeelingense in different habitats.

Latitude (N)

Longitude (E)

Altitude (m)

Aspect

Slope (◦)

Habitat

Population

26.99395

88.28557

2449

SW

0–15

Hilly slope

5

27.00408

88.22867

2176

SE

15–30

Roadside

1

27.05187

88.27033

1830

NE

15–30

Roadside

5

26.98553

88.1428

2246

SE

15–30

Roadside

1

26.99013

88.1141

2170

SW

15–30

Hilly slope

3

26.9908

88.15693

2197

SE

0–15

Hilly slope

3

27.00958

88.17978

2151

SE

15–30

Hilly slope

8

27.01318

88.19185

2188

SE

15–30

Hilly Sslope

23

26.87283

88.28515

1457

SW

15–30

Jhora

1

27.01334

88.29806

2134

SE

15–30

Roadside

2

27.02173

88.31473

1951

SE

15–30

Roadside

4

27.03048

88.3302

1799

SE

30–45

Stream bank, Hilly slope

8

27.0599

88.3569

1628

NE

15–30

Hilly slope

7

27.0764

88.62195

1847

NE

30–45

Hilly slope

2

27.08777

88.64798

2079

SE

15–30

Hilly slope

3

27.08153

88.67965

2219

SE

0–15

Hilly slope

5

27.09168

88.69067

1910

SW

15–30

Hilly slope

1

27.09618

88.6527

2150

NE

15–30

Hilly slope

4

27.10963

88.65315

1938

NE

15–30

Scrub

1

27.07688

88.66888

2090

SW

15–30

Scrub

1

27.06005

88.67223

1772

SW

15–30

Hilly slope

4

 

 

Table 2. Variables used for species distribution modeling in MaxEnt.

Variable abbreviation

Variable Name

Units

BIO02

Mean diurnal range

°C

BIO03

Isothermality

%

BIO07

Temperature annual range

°C

BIO11

Mean temperature of coldest quarter

°C

BIO15

Precipitation seasonality

mm

BIO18

Precipitation of warmest quarter

mm

BIO19

Precipitation of coldest quarter

mm

ALT

Altitude

m

ASPECT

Aspect

NA

SLOPE

Slope

(°)

 

 

 

Table 3. Prediction accuracy with important variables of Herpetospermum darjeelingense models.

 

AUC

TSS

Percentage contribution

Permutation importance

Jackknife training gain

Variable

Value

Variable

Value

In isolation

In absence

Current

0.986

0.949

BIO11

61.2

ALT

59.5

BIO11

BIO15

2050

RCP 2.6

CCSM4

0.988

0.946

BIO11

59.9

ALT

66.2

ALT

BIO15

GFDL-CM3

0.987

0.942

BIO11

63

ALT

46.2

BIO11

BIO15

MIROC5

0.987

0.933

BIO11

60.3

ALT

61.2

BIO11

BIO15

Ensemble

0.987

0.94

BIO11

61.1

ALT

57.9

-

-

RCP 4.5

CCSM4

0.985

0.957

BIO11

63.6

ALT

46.7

BIO11

BIO15

GFDL-CM3

0.986

0.93

BIO11

63.6

ALT

56.8

BIO11

BIO15

MIROC5

0.99

0.93

BIO11

57.9

ALT

43.7

BIO11

BIO15

Ensemble

0.987

0.939

BIO11

61.7

ALT

49.1

-

-

RCP 8.5

CCSM4

0.989

0.966

BIO11

60.3

ALT

57.8

BIO11

BIO15

GFDL-CM3

0.989

0.946

BIO11

58.7

ALT

63.1

BIO11

BIO15

MIROC5

0.985

0.921

BIO11

60.5

ALT

45.5

BIO11

BIO15

Ensemble

0.987

0.944

BIO11

59.8

ALT

55.5

-

-

2070

RCP 2.6

CCSM4

0.988

0.954

BIO11

61.7

ALT

48.4

BIO11

BIO15

GFDL-CM3

0.989

0.94

BIO11

56.3

ALT

60.7

BIO11

BIO15

MIROC5

0.988

0.906

BIO11

61.5

ALT

52.7

ALT

BIO15

Ensemble

0.988

0.933

BIO11

59.8

ALT

53.9

-

-

RCP 4.5

CCSM4

0.988

0.926

BIO11

60.9

ALT

32

BIO11

BIO15

GFDL-CM3

0.989

0.891

BIO11

62.5

ALT

48.9

BIO11

BIO15

MIROC5

0.986

0.892

BIO11

61

ALT

54.8

BIO11

BIO15

Ensemble

0.987

0.903

BIO11

61.4

ALT

45.2

-

-

RCP 8.5

CCSM4

0.985

0.949

BIO11

59.4

ALT

61.6

BIO11

BIO15

GFDL-CM3

0.988

0.955

BIO11

60.9

ALT

47.4

BIO11

BIO15

MIROC5

0.988

0.918

BIO11

60.4

ALT

58.6

BIO11

BIO15

Ensemble

0.987

0.941

BIO11

60.2

ALT

55.9

-

-

 

 

 

For figures & images - - click here for full PDF

 

 

REFERENCES

 

Abdelaal, M., M. Fois, G. Fenu & G. Bacchetta (2019). Using MaxEnt modelling to predict the potential distribution of the endemic plant Rosa Arabica Crép. in Egypt. Ecological Informatics 50: 68–75. https://doi.org/10.1016/j.ecoinf.2019.01.003

Allouche, O., A. Tsoar & R. Kadmon (2006). Assessing the accuracy of species distribution models: prevalence, kappa and the true skill statistic (TSS). Journal of Applied Ecology 43(6): 1223–1232. https://doi.org/10.1111/j.1365-2664.2006.01214.x

Araújo, M.B. & M. New (2007). Ensemble forecasting of species distributions. Trends in ecology & evolution 22(1): 42–47. https://doi.org/10.1016/j.tree.2006.09.010

Babar, S., G. Amarnath, C.S. Reddy, A. Jentsch & S. Sudhakar (2012). Species distribution models: ecological explanation and prediction of an endemic and endangered plant species (Pterocarpus santalinus L. f). Current Science 102(8): 1157–1165.

Ball, J.W., T.P. Robinson, G.W. Wardell-Johnson, J. Bovill, M. Byrne & P.G. Nevill (2020). Fine-scale species distribution modelling and genotyping by sequencing to examine hybridisation between two narrow endemic plant species. Scientific reports 10(1): 1–12. https://doi.org/10.1038/s41598-020-58525-2

Banag, C., T. Thrippleton, G.J. Alejandro, B. Reineking & S. Liede-Schumann (2015). Bioclimatic niches of selected endemic Ixora species on the Philippines: predicting habitat suitability due to climate change. Plant Ecology 216(9): 1325–1340. https://doi.org/10.1007/s11258-015-0512-6

Banerjee, A., M. Devi, A. Nag, R.K. Sharma & A. Kumar (2017). Modelling Probable Distribution of Podophyllum hexandrum in North-Western Himalaya. Indian Forester 143(12): 1255–1259.

Beaumont, L.J., L. Hughes & M. Poulsen (2005). Predicting species distributions: use of climatic parameters in BIOCLIM and its impact on predictions of species’ current and future distributions. Ecological Modelling 186(2): 251–270. https://doi.org/10.1016/j.ecolmodel.2005.01.030

BGCI (2021). Threat Search online database. Botanic Gardens Conservation International. Richmond, UK. https://www.bgci.org/threat_search.php. Electronic version accessed on 05 May 2021.

CEPF (2005). Ecosystem Profile: eastern Himalayas region. In: Critical Ecosystem Partnership Fund.  https://www.cepf.net/sites/default/files/final.ehimalayas.ep_.pdf. Downloaded on 05 May 2021.

CEPF (2021). Critical Ecosystem Partnership Fund. https://www.cepf.net/our-work/biodiversity-hotspots/himalaya. Electronic version accessed on 04 April 2021.

Chakraborty, S., S. Majumder, A. Ghosh, S. Saha & M. Bhattacharya (2021). Metabolomics of potential contenders conferring antioxidant property to varied polar and non-polar solvent extracts of Edgaria darjeelingensis CB Clarke. Bulletin of the National Research Centre 45(1): 1–12. https://doi.org/10.1186/s42269-021-00503-3

Chandra, N., G. Singh, S. Lingwal, M.P.S. Bisht & L.M. Tiwari (2021). Population assessment and habitat distribution modelling of the threatened medicinal plant Picrorhiza kurroa Royle ex Benth. in the Kumaun Himalaya, India. Journal of Threatened Taxa 13(7): 18868–18877. https://doi.org/10.11609/jott.5603.13.7.18868-18877

Chaturvedi, R.K., J. Joshi, M. Jayaraman, G. Bala & N.H. Ravindranath (2012). Multi-model climate change projections for India under representative concentration pathways. Current Science 103(7): 791–802.

Chhetri, B. & H.K. Badola (2017). Predicting suitable habitats for the vulnerable species, Rhododendron niveum Hook. f. (Ericaceae) in eastern Himalaya. NeBIO 8(3): 139–146.

Das, A. (1995). Diversity of angiospermic flora of Darjeeling hills, pp 118–127. In: Pandey, A. (eds.) Taxonomy and Biodiversity CBS Publishers & Distributors, India.

Dullinger, S., A. Gattringer, W. Thuiller, D. Moser, N.E. Zimmermann, A. Guisan, W. Willner, C. Plutzar, M. Leitner,T. Mang, M. Caccianiga, T. Dirnböck, S. Ertl, A. Fischer, J. Lenoir, J. Svenning, A. Psomas, D. Schmatz, U. Silc, P. Vittoz & K. Hülber (2012). Extinction debt of high-mountain plants under twenty-first-century climate change. Nature Climate Change 2: 619–622. https://doi.org/10.1038/nclimate1514

Elith, J. & J.R. Leathwick (2009). Species distribution models: ecological explanation and prediction across space and time. Annual review of ecology, evolution, and systematics 40: 677–697. https://doi.org/10.1146/annurev.ecolsys.110308.120159

Elith, J., S.J. Phillips, T. Hastie, M. Dudík, Y.E. Chee & C.J. Yates (2011). A statistical explanation of MaxEnt for ecologists. Diversity and Distributions 17(1): 43–57. https://doi.org/10.1111/j.1472-4642.2010.00725.x

Erdtman, G. (1960). The acetolysis method: a revised description. Svensk Botanisk Tidskrift 54: 561–564.

Fick, S.E. & R.J. Hijmans (2017). WorldClim 2: new 1km spatial resolution climate surfaces for global land areas. International Journal of Climatology 37(12): 4302–4315. https://doi.org/10.1002/joc.5086

Gómez-Ruiz, E.P. & T.E. Lacher Jr. (2019). Climate change, range shifts, and the disruption of a pollinator-plant complex. Scientific Reports 9(1): 1–10. https://doi.org/10.1038/s41598-019-50059-6

Grierson, A.J.C & D.G. Long (eds.) (1991). Flora of Bhutan: Volume II Part 1. The Royal Botanic Garden of Edinburgh, United Kingdom.

Griffies, S.M., M. Winton, L.J. Donner, L.W. Horowitz, S.M. Downes, R. Farneti, A. Gnanadesikan, W.J. Hurlin, H. Lee, Z. Liang, J.B. Palter, B.L. Samuels, A.T. Wittenberg, B.L. Wyman, J. Yin & N. Zadeh (2011). The GFDL CM3 coupled climate model: characteristics of the ocean and sea ice simulations. Journal of Climate 24(13): 3520–3544. https://doi.org/10.1175/2011JCLI3964.1

Hart, R., J. Salick, S. Ranjitkar & J. Xu (2014). Herbarium specimens show contrasting phenological responses to Himalayan climate. Proceedings of the National Academy of Sciences 111(29): 10615–10619. https://doi.org/10.1073/pnas.1403376111

Hernandez, P.A., C.H. Graham, L.L. Master & D.L. Albert (2006). The effect of sample size and species characteristics on performance of different species distribution modelling methods. Ecography 29(5): 773–785. https://doi.org/10.1111/j.0906-7590.2006.04700.x

Hijmans, R.J., S.E. Cameron, J.L. Parra, P.G. Jones & A. Jarvis (2005). Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25: 1965–1978. https://doi.org/10.1002/joc.1276

Hills, R., S. Bachman, F. Forest, J. Moat & P. Wilkin (2019). Incorporating evolutionary history into conservation assessments of a highly threatened group of species, South African Dioscorea (Dioscoreaceae). South African Journal of Botany 123: 296–307. https://doi.org/10.1016/j.sajb.2019.03.032

IPCC (2014). Climate Change 2014: synthesis report. In: Pachauri RK, Meyer LA (eds) Contribution of working groups I, II and III to the Fifth Assessment Report of the Intergovernmental Panel on Climate Change Core Writing Team. IPCC, Switzerland.

Irfan-Ullah, M., G. Amarnath, M.S.R. Murthy & A.T. Peterson (2006). Mapping the geographic distribution of Aglaia bourdillonii Gamble (Meliaceae), an endemic and threatened plant, using ecological niche modelling, pp. 343–351. In: Hawksworth, D.L. & A.T. Bull (eds.). Plant Conservation and Biodiversity. Springer, Dordrecht, 421 pp.

Kandel, P., N. Chettri, R.P. Chaudhary, H.K. Badola, K.S. Gaira, S. Wangchuk, N. Bidha, Y. Uprety & E. Sharma (2019). Plant diversity of the Kangchenjunga Landscape, Eastern Himalayas. Plant diversity 41:  153–165. https://doi.org/10.1016/j.pld.2019.04.006

Khanum, R., A.S. Mumtaz & S. Kumar (2013). Predicting impacts of climate change on medicinal asclepiads of Pakistan using Maxent modelling. Acta Oecologica 49: 23–31. https://doi.org/10.1016/j.actao.2013.02.007

Meehl, G.A., W.M. Washington, J.M. Arblaster, A. Hu, H. Teng, C. Tebaldi, B.N. Sanderson, J. Lamarque, A. Conley, W.G. Strand & J.B. White III (2012). Climate system response to external forcings and climate change projections in CCSM4. Journal of Climate 25(11): 3661–3683. https://doi.org/10.1175/JCLI-D-11-00240.1

Mueller-Boeker, U. (1993). Ethnobotanical studies among the Chitawan Tharus. Journal of the Nepal Research Center 9: 17–56. https://doi.org/10.5167/uzh-33899

Nirola, S. & A.P. Das (2017). Endemic monocot flora of Darjeeling Himalaya, West Bengal, India. Pleione 11(1): 116–124.

Pande, H.K. & S. Arora (2014). India’s Fifth National Report to the Convention on Biological Diversity. Ministry of Environment and Forestry, Government of India, India.

Pandit, M.K., K. Manish & L.P. Koh (2014). Dancing on the roof of the world: ecological transformation of the Himalayan landscape. BioScience 64(11): 980–992. https://doi.org/10.1093/biosci/biu152

Pearson, R.G. (2007). Species distribution modelling for conservation educators and practitioners. Synthesis: American Museum of Natural History 50: 54–89.

Pearson, R.G., C.J. Raxworthy, M. Nakamura & A.T. Peterson (2007). Predicting species distributions from small numbers of occurrence records: a test case using cryptic geckos in Madagascar. Journal of Biogeography 34(1): 102–117. https://doi.org/10.1111/j.1365-2699.2006.01594.x

Peterson, A.T., M. Papeş & M. Eaton (2007). Transferability and model evaluation in ecological niche modelling: a comparison of GARP and Maxent. Ecography 30(4): 550–560. https://doi.org/10.1111/j.0906-7590.2007.05102.x

Phillips, S.J., R.P. Anderson & R.E. Schapire (2006). Maximum entropy modelling of species geographic distributions. Ecological Modelling 190(3-4): 231–259. https://doi.org/10.1016/j.ecolmodel.2005.03.026

Phillips, S.J., R.P. Anderson, M. Dudík, R.E. Schapire & M.E. Blair (2017). Opening the black box: An open-source release of Maxent. Ecography 40(7): 887–893. https://doi: 10.1111/ecog.03049

Potts, A.J., T.A. Hedderson, J.H. Vlok & R.M. Cowling (2013). Pleistocene range dynamics in the eastern Greater Cape Floristic Region: a case study of the Little Karoo endemic Berkheya cuneata (Asteraceae). South African Journal of Botany 88: 401–413. https://doi.org/10.1016/j.sajb.2013.08.009

POWO (2021). Herpetospermum darjeelingense (C.B.Clarke) H.Schaef. & S.S.Renner | Plants of the World Online | Kew Science. http://powo.science.kew.org/taxon/urn:lsid:ipni.org:names:77110566-1. Electronic version accessed 01 May 2021.

Purohit, S. & N. Rawat (2021). MaxEnt modeling to predict the current and future distribution of Clerodendrum infortunatum L. under climate change scenarios in Dehradun district, India. Modeling Earth Systems and Environment 8: 2051–2063. https://doi.org/10.1007/s40808-021-01205-5

Qin, A., K. Jin, M.E. Batsaikhan, J. Nyamjav, G. Li, J. Li, Y. Xue, G. Sun, L. Wu, T. Indree, Z. Shi & W. Xiao (2020). Predicting the current and future suitable habitats of the main dietary plants of the Gobi Bear using MaxEnt modelling. Global Ecology and Conservation 22: e01032. https://doi.org/10.1016/j.gecco.2020.e01032

Renner, S.S. & A.K. Pandey (2013). The Cucurbitaceae of India: Accepted names, synonyms, geographic distribution, and information on images and DNA sequences. PhytoKeys 20: 53–118. https://doi.org/10.3897/phytokeys.20.3948

Shrestha, I. & P. Khadgi (2019). Herbal veterinary practices by Tamang community in central Nepal. NUTA Journal 6(1-2): 5–11.

Singh, G., N. Chandra, S. Lingwal, M.P.S Bisht & L.M. Tiwari (2020). Distribution and threat assessment of an endemic and endangered species Angelica glauca in high ranges of western Himalaya. Journal of Herbs, Spices and Medicinal Plants 26: 394–404. https://doi.org/10.1080/10496475.2020.1748783

Srivastava, V., V.C. Griess & H. Padalia (2018). Mapping invasion potential using ensemble modelling. A case study on Yushania maling in the Darjeeling Himalayas. Ecological modelling 385: 35–44. https://doi.org/10.1016/j.ecolmodel.2018.07.001

Sumarga, E. (2011). The comparison of logistic regression, geostatistics and Maxent for distribution modeling of a forest endemic: a pilot study on lobels maple at MT Pizzalto, Italy. MSc Thesis. Faculty of Geo-information Science and Earth Observation, University of Twente, vii+63 pp.

Swets, J.A. (1988). Measuring the accuracy of diagnostic systems. Science 240(4857): 1285–1293. https://doi.org/10.1126/science.3287615

Villa-Machío I., A.G.F. de Castro, J.F. Aguilar & G.N. Feliner (2020). Colonization history of the Canary Islands endemic Lavatera acerifolia, (Malvaceae) unveiled with genotyping by sequencing data and niche modelling. Journal of Biogeography 47(4): 993–1005. https://doi.org/10.1111/jbi.13808

Warren, D.L., R.E. Glor & M. Turelli (2010). Enmtools: a toolbox for comparative studies of environmental niche models, evaluation. Ecography 31: 161–175. https://doi.org/10.1111/j.1600-0587.2009.06142.x

Watanabe, M., T. Suzuki, R. O’ishi, Y. Komuro, S. Watanabe, S. Emori, T. Takemura, M. Chikira, T. Ogura, M. Sekiguchi, K. Takata, D. Yamazaki, T. Yokohata, T. Nozawa, H. Hasumi, H. Tatebe & M. Kimoto (2010). Improved climate simulation by MIROC5: Mean states, variability, and climate sensitivity. Journal of Climate 23(23): 6312–6335. https://doi.org/10.1175/2010JCLI3679.1