Study Area
With a large biogeographic domain, the valley of Kashmir is situated in the north-western Himalayas [62]. The region comes under coordinates of (33o22' and 34o50' N latitudes and 73o55 'and 73o33' E longitudes), covering an area of about 16,000 km2, with about 64 % of the total area is comprised of mountains [63]. The valley is a deep elliptical bowl-shaped surrounded by a girdling chain of high mountain ranges, namely the Pir Panjal in the south and south-west, and the Zanskar, the Greater Himalayan range in the north and north-east.
Also, the valley of Kashmir has a network of numerous glaciated streams, lakes, springs, rivers as well as alpine, sub-alpine, and freshwater ecosystems along an altitudinal gradient, mainly due to its unique edaphic characteristics, eco-climatic conditions, and geographical location, together with its undulating topography and lofty snowcapped mountains and hills.
Data collection
We targeted the aquatic flora of Kashmir Himalaya with special reference to lakes and wetlands for the present study. In the beginning, information about the occurrence, distributional range, invasiveness, clonality, origin, growth form, life form, and stages of the invasion were obtained through extensive screening of available literature such as, specimens of the targeted flora that was stored in the Kashmir University Herbarium (KASH), published research articles, weed floras, etc. The most valuable source of information about the presence of aquatic invasive species among different habitats were perused [25, 64-75].
Pursuant to useful information gained through the study of previous literature, we followed up with thorough field surveys undertaken for a period of 3 years (2014-2017), in different aquatic habitats of the Kashmir valley. Species were grouped in invasion stages based on the extent of spatial spread in the Kashmir Himalayan region, measured in terms of the frequency percentage across sites using standard quadrat sizes of 0.5, 1, and 5m2 for free-floating submerged, and emergent macrophytes, respectively, in the framework of the CM model.
The field data were collected during seasonal surveys inthe target water bodies. During the survey, a total of 125 species were recorded from 9 lakes, 5 wetlands, 7 streams, and 2 rivers. Out of them, 102 species were included in the present study, while the remaining 23 species were excluded because of insufficient information (Appendix 1). A total of 2300 (1 m2) quadrats were laid randomly (100 in each water body) across different zones (littoral, marsh, open waters) to take into account maximum macrophytic diversity for the present analysis.
All the species were uprooted to ascertain whether it is clonal or non-clonal. All the clonal species were evaluated for their clonal organs (Appendix1). We evaluated the clonal organs after the flowering and fruiting stage, and at the end of a season which was the best time for evaluation because some of the clonal traits develop at the later stages. For clonal organ evaluation, two to ten individual plants of each species were excavated with below-ground organs. For conformity whether we are in correct identification of the clonal growth organs and life form for each species from the wetland species pool,we used the CLO-PLA 3 (CLOnal PLAnts, version 3) of the CLOPLA database (http://clopla.butbn.cas.cz/). This online data-base of clonal growth of plants, which contains clonal traits and vegetative regeneration of about 186,157 records of 2,923 species for the European temperate flora. This data-base is freely available and can serve as a guide for clonal trait sampling in different parts of the globe, together with specific detailed information on how to use it and the nature of clonal traits [50, 76] and PLADIAS (Plant Diversity Analysis and Synthesis, 2014–2018) of the Pladias-database of the Czech Flora and Vegetation (www.pladias.cz) of clonal growth in plants. It includes more than 13 million records of almost 5 thousand taxa (species, subspecies, varieties and hybrids), which came from seven regional projects, five large national databases and records collected within the PLADIAS project. The Pladiasdatabasecovers largest set of data on vascular plants of the Czech Republic [77]. The present study covered almost all the major aquatic habitats, including ninelakes (Anchar, Dal, Hokarsar, Wular, Mansbal, Ahansar, Narangbagh, Nilnag, and Waskar), five wetlands (Shallabugh, Tulmula, Malangpora, and Kranchu), seven streams (Achabal, Bal-Kol, Irrigation canal and spring stream of Sundoo, Nambal rivulet, Nagrad stream, Aarpath rivulet, and Spring stream) and two rivers (Jehlum and Sindh) and their tributaries (Table 1).
River Jehlum is the main river of the Kashmir valley. The major tributaries of the Jehlum river includes; Sind (Ganderbal/ Srinagar), Lidder (Anantnag), Rambiara (Pulwama/ Kulgam) and Pohru (Kupwara).The minor tributaries of the Jehlum includes, ArapatKol, Vishav, Sandran, Bringi, and Romushali (Anantnag, Kulgam); Arapal (Pulwama); Harwan (Srinagar); Rambiara (Shopian); Vij-Dakil, Erin, Madumati, and Ningal (Bandipora, Bramulla) and Sukhnag-Firozepora, Dudhganga- Shaliganga, (Budgam, Baramulla). In addition to these, there are many ponds, marshes and irrigation channels which support various species of aquatic plants.
Data on geographic origins, degree of invasiveness, growth form, life form and stage of invasion of 102 invasive alien plant species in Kashmir Himalaya were obtained from personal field surveys and other published sources [73, 75] (Appendix I and Table 8) and the majority (72.89%) of invasive alien species are of European origin (Fig. 4).
Data analysis
Data analysis was done in R Studio (version 1.2.1335; R version 3.6.2). The association of clonality and invasion stage was discerned by performing the simple Pearson’s correlation. Additionally, we attempted to consolidate the individual contribution of different clonal growth organs towards clonality and how closely associated the latter is to different life forms by performing principal component analysis (PCA) using the ggbiplot package in R studio [78]. By doing this we transformed a large number of possibly correlated variables into an even smaller number of uncorrelated variables called principal components.