Our N Distribution Pattern and Spec Nepal H Our Nature Bution Pattern
Total Page:16
File Type:pdf, Size:1020Kb
Rana et al. / Our Nature (2016), 14 (1): 22-29 Our Nature│December 2016│14 (1): 22-29 ISSN: 1991-2951 (Print) Our Nature ISSN: 2091-2781 (Online) Journal homepage: http://nepjol.info/index.php/ON Distribution pattern and species richness of Liliaceae in the Nepal Himalaya Hum Kala Rana1, Santosh Kumar Rana1,2,3* and Suresh Kumar Ghimire1 1Central Department of Botany, Tribhuvan University, Kirtipur, Kathmandu, Nepal 2Key Laboratory of Plant Diversity and Biogeography of east Asia, Kunming Institute of Botany, Chinese Academy of Sciences, Kunming, Yunnan 650204, China 3University of Chinese Academy of Sciences, Beijing 100049, China *E-mail: [email protected] Abstract The most important aspect of plant conservation is to predict the potential distribution and its richness in response to climate change. Contributing to the management pro- gram, this study aimed to predict the distribution and richness pattern of Liliaceae in Nepal. The BIOCLIM in DIVA GIS 7.5 model based on distribution records of 19 spe- cies belonging to three subfamilies of Liliaceae (Lilioideae, Streptopoideae and Calo- chortoideae) and 19 climatic variables (derived from Worldclim), revealed that Lilioi- deae and Streptopoideae are potentially distributed in most of the hilly and mountainous regions of Nepal; whereas Calochortoideae mostly in Eastern and very scanty in Central Nepal. Lilioideae is projected to have high species richness in Central and Western Nepal as compared to other subfamilies. Key words: Conservation, DIVA GIS, Lily, Modelling DOI: http://dx.doi.org/10.3126/on.v14i1.16437 Manuscript details: Received: 27.07.2016 / Accepted: 28.11.2016 Citation: Rana, H.K., S.K. Rana and S.K. Ghimire 2016. Distribution pattern and species richness of Liliaceae in the Nepal Himalaya. Our Nature 14(1):22-29. DOI: http://dx.doi.org/10.3126/on.v14i1.16437 Copyright: © Rana et al. 2016. Creative Commons Attribution-NonCommercial 4.0 International License. Introduction geophytes (but also possess other forms of The family is named after Lilium, the type enlarged underground stem). This family genus (Simpson, 2006). Lilium is Latin for contains many members with economic ‘lily’ and which came from the Greek word importance. Notable among them are spe- leírion (lily flower “a classic symbol of cies of Fritillaria, Lilium, Tulipa and Tri- purity”). Liliaceae are generally bulbous cyrtis. Particularly, Fritillaria cirrhosa, 22 Rana et al. / Our Nature (2016), 14 (1): 22-29 Lilium nepalense and L. wallichianum are el of DIVA-GIS to predict the species cur- also valued in Nepal as medicine and food rent distribution (Guisan and Zimmerman, (Ghimire et al., 2008; Kunwar et al., 2006). 2000; Hijmans and Graham, 2006; Partha- F. cirrhosa and L. nepalense are traded sarathy et al., 2006; Rajbhandary et al., from Nepal to overseas herbal market. Con- 2010; Barman et al., 2011; Babar et al., sidering their increasing exploitation for 2012), taking one or few genera and even a trade, Conservation Assessment Manage- whole family. But, no work has focused on ment Prioritization Workshop (CAMP) held the family Liliaceae. However, very less in Nepal has listed F. cirrhosa and L. nepa- work has been done on species distribution lense as threatened medicinal plants under modeling for Nepal. Although distribution threat categories ‘Vulnerable’ and ‘Data modeling cannot be treated as substitute for Deficient’, respectively (Tandon et al. detailed empirical field data, including data 2001). Both of these species are also in- on demography, abundance and species- cluded in National Register of Medicinal environment interactions, its prediction ac- Plants (IUCN, 2004) and these medicinal curacy can be judged based on available plants are important components in the flo- information. In case of Liliaceae, the col- ra of Nepal (Rana et al., 2015). lection of occurrence records is limited The members of the family are widespread mostly to the sites where field excursions throughout the world, but most common in have been made in the past, which may re- temperate to subtropical regions, especially sult in the perception that they grow under East Asia and North America in the North- certain conditions when other areas were ern hemisphere. They occur mostly in the simply not examined. Selected parameters steppes and mountain meadows. The centre may not be sufficient to describe the spe- of diversity is considered as S.W. Asia to cies habitat requirements. Certain species China (Simpson, 2006). have specific ecological limitations govern- It is a well known fact that plant spe- ing where they are known to occur, which cies are not homogenously distributed, each may result in over-prediction of suitable species depends on the existence of a spe- habitat by the model. Distribution model cific set of environmental conditions for its (BIOCLIM) predicts suitability areas only long-term survival (Gaston and Blackburn, in the neighbourhood of occurrence 2000). Detailed knowledge of species eco- records. Thus, the present work also aims, logical and geographic distribution is fun- 1) to predict the current potential geograph- damental for conservation planning and ic distribution of three subfamilies of Lilia- forecasting (Ferrier, 2002; Funk and Rich- ceae: Lilioideae, Streptopoideae and Calo- ardson, 2002; Rushton et al., 2004), and for chortoideae; 2) to generate the overall spe- understanding ecological and evolutionary cies richness pattern for the family Lilia- determinants of spatial patterns of biodiver- ceae in Nepal. sity (Rosenzweig, 1995; Brown and Lomo- lino, 1998; Ricklefs, 2004). Materials and methods To predict species potential geo- Climatic and species data graphic distribution and its richness, a Locality records (in total, 218) for 19 enu- range of models have been developed. merated species (classified into 3 subfami- Many scientists have used BIOCLIM mod- lies viz., Lilioideae, Streptopoideae, Calo- 23 Rana et al. / Our Nature (2016), 14 (1): 22-29 chortoideae) of family Liliaceae were used Data cleaning for distribution modeling. These locality The initial record of presence points were records were obtained from the herbarium 355, after removing the multiple collection specimens housed at different herbaria Na- records. The presence points were georefe- tional Herbarium and Plant Laboratories renced using ArcGIS 9.3 and duplicate (KATH), Tribhuvan University Central records were removed within each 2 min Herbarium (TUCH), University of grid cell (separately for species of 3 subfa- Tokyo (TI) and Royal Botanic Garden milies) to obtain least spatial autocorrela- tion. Finally, 164 collection points for sub- Edinburgh (RBGE). A set of 19 climatic family Lilioideae (16 species), 42 for Strep- variables for Nepal (Tab. 1) were extracted topoideae (2 species) and 12 for Calochor- from WORLDCLIM (http//www.worldclim toideae (1 species) were obtained. .org) (Hijmans et al., 2005). WORLDCLIM contains climatic data at a spatial resolution Distribution modeling of 2.5 arc sec (~5 Km2) obtained by inter- In ecological and biogeographical studies, polation of climatic station records from the present distribution of organism is ana- 1950-2000. The variables were derived lyzed with regard to the characteristics of from monthly precipitation and maximum the physical environment, involving mod- and minimum temperature. eling. Prediction of distributions have Table 1. List of 19 bioclimatic variables used for predicting been carried out for many types of organ- the species distribution. BIO1 Annual Mean Temperature ism, especially with the application of DI- Mean Diurnal Range (Mean of monthly (max VA-GIS (Guisan and Zimmerman, 2000; BIO2 temp –min temp)) Delanoy and Damme, 2006; Hijmans and BIO3 Isothermality (BIO2/BIO7) (* 100) Graham, 2006; Parthasarathy et al., 2006; Temperature Seasonality (standard deviation Zafra-Calvo et al., 2010; Hou et al., 2010; BIO4 *100) Rajbhandary et al., 2010; Barman et al., BIO5 Max Temperature of Warmest Month 2011; Babar et al., 2012). The environmen- BIO6 Min Temperature of Coldest Month tal suitability for each species was ranked BIO7 Temperature Annual Range (BIO5–BIO6) under five classes (Tab. 2). The places not BIO8 Mean Temperature of Wettest Quarter suitable for the species were displayed in BIO9 Mean Temperature of Driest Quarter white. Areas with various classes of suita- BIO10 Mean Temperature of Warmest Quarter temperature min and from max Derived bility for the species were shown from dark BIO11 Mean Temperature of Coldest Quarter green to red color. The dark green color BIO12 Annual Precipitation means lower suitability (0-2.5 percentile) BIO13 Precipitation of Wettest Month while red (20-30 percentile) indicates place BIO14 Precipitation of Driest Month with higher suitability for species present Precipitation Seasonality (Coefficient of BIO15 occurrence. For each grid cell of the map, Variation) percentile value is assigned, which range BIO16 Precipitation of Wettest Quarter from zero (low) to the theoretical maximum BIO17 Precipitation of Driest Quarter of 50 (very high). Locations where the val- BIO18 Precipitation of Warmest Quarter Derived from precipitation from Derived ues lie within the 2.5-30 percentile of the BIO19 Precipitation of Coldest Quarter climate envelope are traditionally classified as ‘core’ regions of species occurrence. Ex- 24 Rana et al. / Our Nature (2016), 14 (1): 22-29 cellent distribution of species belonging to family Calochortoideae, mostly the North- subfamilies is shown by the region with 20-