Armine Abrahamyan Distribution Modeling and Population Ecology Of

DAUGAVPILS UNIVERSITY Institute of Life Sciences and Technology

ARMINE ABRAHAMYAN

DISTRIBUTION MODELING AND POPULATION ECOLOGY OF WILD MELISSA OFFICINALIS L. (LAMIACEAE) AND ORIGANUM VULGARE L. (LAMIACEAE) IN ARMENIA

DOCTORAL DISSERTATION IN BIOLOGY FOR THE SCIENTIFIC DEGREE (BRANCH: ECOLOGY)

SCIENTIFIC ADVISORS: DR. AGR. ANDREAS MELIKYAN DR. BIOL. ARVIDS BARSEVSKIS

Daugavpils 2015

Institute of Life Sciences and Technology, Daugavpils University, Latvia.

Type of work: Doctoral Dissertation in the branch of Biology, sub-branch of Ecology.

The Doctoral Dissertation was prepared at Armenian National Agrarian University in 2007-2011 and at Daugavpils University in 2012-2014.

The work was supported by EU European Social Fund, within the Project “Support for the implementation of doctoral studies at Daugavpils University” Agreement Nr. 2012/0004/1DP/1.1.2.1.2/11/IPIA/VIAA/011.

Scientific advisors:

 Dr. sc. agr., Prof. Andreas Melikyan (Armenian National Agrarian University, Yerevan, Armenia)  Dr. biol., Prof. Arvīds Barševskis (Daugavpils University, Daugavpils, Latvia)

Reviewers:

 Artūrs Škute, Dr. biol., prof., Daugavpils University, Latvia  Ingrida Šaulienė, PhD., prof., Siauliai University, Lithuania  Glen Clark Shinn, PhD., senior scientist, Texas A&M University, Norman Borlaug Institute for International Agriculture, ASV

The Chairman of the Promotion Council: Dr. biol., Prof. Arvīds Barševskis

Commencement: Room 115, Parādes 1, Daugavpils University, Daugavpils, at 12:00 pm on May 12, 2015.

Secretary of the Promotion Council: Dr. biol. Jana Paidere, researcher at the Institute of Life Sciences and Technology of Daugavpils University.

The Doctoral Thesis and it's summary are available at the library of Daugavpils University, Vienības street 13, Daugavpils and: http://www.du.lv/lv/zinatne/promocija/aizstavesanai_iesniegtie_promocijas_darbi.

TABLE OF THE CONTENTS Table of the contents...... 3 List of original papers...... 4 List of thesis ...... 5 Approbation of the research results...... 6 Annotation...... 8

Chapter I: INTRODUCTION...... 9 1.1. Introduction and Theoretical Framework...... 9 1.2. Statement of the Problem...... 11 1.3. Novelty and Tasks of the Research...... 14 1.4. Hypothesis of the Research Work...... 16

Chapter II: REVIEW OF LITERATURE...... 17 2.1. Description of Definitions...... 17 2.2. Brief Description of Melissa officinalis L. (Lamiaceae)...... 18 2.3. Brief Description of Origanum vulgare L. (Lamiaceae)...... 21 2.4. Potential Distribution of Wild Medicinal plants in Armenia...... 24 2.5. Species Potential Distribution Modeling...... 28 2.6. The Applications of Species Potential Distribution Modeling...... 30 2.7. Algorithms for Modeling Plants’ Potential Distributions...... 35 2.8. Desktop GARP (Genetic algorithm)...... 37

Chapter III: RESEARCH METHODS...... 39 3.1. Introduction...... 39 3.2. Eco-geographic Survey and Field Data collection...... 40 3.3. Environmental Niche Modeling (ENM)...... 45 3.4. Methods of Statistic Analysis of Data Processing...... 57

Chapter IV: RESULTS AND DISCUSSIO ...... 58 4.1. Changes in Distribution of wild M. officinalis L. and O.vulgare L. Populations during the Last Decade in Armenia...... 58 4.2. M. officinalis L. and O. vulgare L. Population’s Ecology and Dynamics...... 65 4.3. Interdependent Effects of Habitat Factors and Climate on Population Dynamics of M. officinalis L. and O. vulgare L. Species in Armenia...... 75 4.4. Wild Melissa officinalis L. and Origanum vulgare L. Population’s Structure...... 91 4.5. Distribution Modeling of M. officinalis L. and O. vulgare L. in the RA...... 101

Chapter V: CONCLUSIONS...... 111 ACKNOLEDGMENT...... 115 REFERENCES...... 116

LIST OF ORIGINAL PAPERS

The results of this research work are published in the international scientific peer reviewed journals. The order of the authors’ names reflects their involvement in the paper.

I. Abrahamyan, A., Barsevskis, A. & Melikyan, A. 2014. Populations Dynamics in Sizes of Wild Melissa officinalis L. (Lamiaceae) During the Last Decade in Armenia. Journal of Medicinal Plants Studies Vol. 3(1): 21-26.

II. Abrahamyan, A., Barsevskis, A., Crockett, S., & Melikyan, A. 2014. Distribution of Origanum vulgare L. and Population Dynamics during the Last Decade in Armenia. Journal of Life Sciences Vol. 8(8): 690-698; doi: 10.17265/1934-7391/2014.08.007

III. Abrahamyan, A., & Barsevskis, A. 2013. Environmental Niche Modeling with Desktop GARP for Wild Origanum vulgare L. (Lamiaceae) in Armenia. Proceeding of the 9th International Scientific and Practical Conference “Environment, Technology, Resources”. Rezekne, Latvia, Vol. (III): 7-11.

IV. Abrahamyan, A. 2012. Wild Origanum vulgare L. (Lamiaceae) Populations’ Distribution Modeling in Armenia. Proceeding of International Young Scholars scientific-practical conference “A challenging opportunity and time: problems, solutions and perspectives”. Rezekne, Latvia, Vol. (I):144-146.

V. Abrahamyan, A. 2011. Changes in Distribution and Structure of Wild Melissa officinalis L. Populations during the Last Decade in Armenia and Implications for Conservation. Proceeding of the 8th International Scientific and Practical Conference “Environment, Technology, Resources”. Rezekne, Vol. (II): 321-324.

VI. Abrahamyan A. 2012. Wild Melissa officinalis L. Populations’ Distribution Modeling in Armenia. Proceeding of the 16th International Scientific and Practical Conference “Human. Environment. Technology”. Rezekne, Latvia, Vol. (I):11-16.

LIST OF THESIS

I. Abrahamyan1 A., Barsevskis2 A. & Melikyan3 A. 2013. Impact of Climate Change and Habitat Loss on Wild Origanum vulgare L. (Lamiaceae) Populations Extinction in the Republic of Armenia. „55th International Scientific Conference of Daugavpils University”, Daugavpils, Latvia. Abstract Book: pp6-7. II. Abrahamyan1 A., Barsevskis2 A. 2013. Potential Distribution of Wild Melissa officinalis L. (Lamiaceae) in Armenia for Conservation Practices. 1thMediterranean Symposium on Medicinal and Aromatic Plants, Gazimagosta, Turkish Republic of Northern Cyprus, Abstract Book: pp340. III. Abrahamyan1 A., Barsevskis1 A. & Melikyan2 A. 2013. Populations Dynamics in Sizes of Wild Melissa officinalis L. (Lamiaceae) for the Last Decade in the Republic of Armenia. “56th Symposium of the International Association for Vegetation Science”, Tartu, Estonia. Abstract Book: pp6. IV. Abrahamyan1 A., Teilans2 A. & Zorins3 A. 2011. Climate change impact on conservation status of wild Melissa officinalis L. (Lamiaceae) populations in Armenia.”59th International Congress and Annual Meeting of the Society for Medicinal Plant and Natural Product Research” Antalya, Turkey. PLANTA MED: pp33. V. Abrahamyan1, A; Crockett, S2 2010. Changes in distribution and structure of wild Origanum vulgare L. (Lamiaceae) populations during the last decade in Armenia and implications for conservation. ”58th International Congress and Annual Meeting of the Society for Medicinal Plant and Natural Product Research”, Berlin, Germany. PLANTA MED: pp12. VI. Abrahamyan, A. 2010. The Populations of Melissa Officinalis L. (Lemon Balm) in Armenia and its Conservation through Introduction into Agricultural Production System. “28th International Horticultural Congress” Lisbon, Portugal. Abstract Book: pp73.

APPROBATION OF THE RESEARCH RESULTS

The basic results of the research were approbated and presented in the 11 international scientific conferences and symposiums:

I. International Forum on Innovation and Research Development: Research, Knowledge, Innovations (Klaipeda, Lithuania, 19-20 th November, 2014): Current Status and Trends of Medicinal Plants Diversity in Armenia towards Sustainable Future. II. 9th International Scientific and Practical Conference: Environment, Technology, Resources. (RA, Rezekne, Latvia, 20-20th June 2013): Environmental Niche Modeling with Desktop GARP for Wild Origanum vulgare L. (Lamiaceae) in Armenia. III. 16th International Scientific and Practical Conference: Human. Environment. Technology. (RA, Rezekne, Latvia, 25th April, 2012): Wild Melissa officinalis L. Populations’ Distribution Modeling in Armenia. IV. 2th International Scientific-practical Conference of Young Scholars: A challenging opportunity and time: Problems, Solutions and Perspectives (Rezekne, Latvia, 15-17th May 2012): Wild Origanum vulgare L. (Lamiaceae) Populations’ Distribution Modeling in Armenia. V. 13th EMBL/EMBO Science and Society Conference: Biodiversity in the Balance-Causes and Consequences, (Heidelberg, Germany, 09-10th November2012). Wild Medicinal Plants Populations Sustainablality Assessment in Armenia. VI. 8th International Scientific and Practical Conference: Environment, Technology, Resources (RA, Rezekne, Latvia, 20-20th June 2011): Changes in Distribution and Structure of Wild Melissa officinalis L. Populations during the Last Decade in Armenia and Implications for Conservation. VII. 55th International Scientific Conference of Daugavpils University (Daugavpils, Latvia, 10-12th April 2013): Impact of Climate Change and Habitat Loss on Wild Origanum vulgare L. (Lamiaceae) Populations Extinction in the Republic of Armenia.

VIII. 1th Mediterranean Symposium on Medicinal and Aromatic Plants (Gazimagosta, Turkish Republic of Northern Cyprus, 17-20th April 2013): Potential Distribution of Wild Melissa officinalis L. (Lamiaceae) in Armenia for Conservation Practices. IX. 56th Symposium of the International Association for Vegetation Science (University of Tartu, Tartu, Estonia, 26–30th June, 2013): Populations Dynamics in Sizes of Wild Melissa officinalis L. (Lamiaceae) for the Last Decade in the Republic of Armenia. X. 59th International Congress and Annual Meeting of the Society for Medicinal Plant and Natural Product Research (Antalya, Turkey, 4th-9th September, 2011): Climate change impact on conservation status of wild Melissa officinalis L. (Lamiaceae) populations in Armenia. XI. 58th International Congress and Annual Meeting of the Society for Medicinal Plant and Natural Product Research (Freie University of Berlin, Germany, 29th August- 2th September 2010): Changes in Distribution and Structure of Wild Origanum vulgare L. (Lamiaceae) Populations during the Last Decade in Armenia and Implications for Conservation. XII. 28th International Horticultural Congress (Lisbon, Portugal, 22th - 27th August 2010): The Populations of Melissa Officinalis L. (Lemon Balm) in Armenia and Its Conservation through Introduction into Agricultural Production System.

ANNOTATION As one of the first countries to join the Convention on Biological Diversity (CBD), Armenia has a strong interest in examining the biodiversity of its native plant species, particularly those with potential or existing economic value (e.g., medicinal plants), and assessing their conservation status (IUCN, WHO, WWF (1993)). Only limited information, however, is at this time available on the genetic biodiversity, population location, structure and size, and conservation status of most of these species. Anthropogenic threats to this biodiversity such as overpopulation, deforestation and urbanization have simultaneously hindered research and increased the need for it. The study examines population ecology and distribution modeling of wild medicinal and culinary herbs native to Armenia. Researches of population ecology and distribution modeling might supply better predictions about future changes in population sizes/growth rates and possible shift in distribution across the country under the potential impacts of climate change (Costa et al. 2010). The structure of the work is composed of introduction, literature review, materials and methods, results and their analysis, conclusions, list of literature. The work has been prepared in respect with International Organization for Standartalization ISO/R 9:1968, sub-standard 2, ISO 9. The dissertation is comprised 128 pages and has illustrative 5 pictures, 9 tables and 34 figures. Used literature is 165, and more than 140 were cited in the dissertation. In addition field trips illustrative pictures mostly were presented in the appendix of the thesis.

This researches is significant for conservation planning in recent years such as assessing the impact of human activities on biodiversity, predicting the impacts of climate change on species’ distribution, preventing the spread of invasive species, searching rare or endangered species, planning new conservation areas, including identifying biological hotspots and setting conservation priorities.

This research provided baseline data that can be used for the development of further conservation strategies of these unique genotypes, as well as to assess the vulnerability of wild populations with regard to the IUCN Red Book Criteria, of this important medicinal and culinary species in Armenia.

Chapter I: Introduction 1.1. Introduction and Theoretical Framework The word is facing a biodiversity crisis of epic proportions. Species is occurring at unprecedented levels, estimated at up to 1,000 times the “back-ground” or natural extinction rate. The current global extinction rate, about 1% of species going extinct in 100 years, is predicted to increase to at least 10% by 2050. The causes are many, including habitat destruction, land conversion for agriculture and development, climate change, pollution, illegal wildlife trade, and the spread of invasive species. As one of the first countries to join the Convention on Biological Diversity (CBD), Armenia has a strong interest in examining the biodiversity of native plant species, particularly those with potential or existing economic value (e.g. medicinal plants), and assessing their conservation status (IUCN, WHO, WWF, 1993). Only limited information, however, is available at this time on the genetic biodiversity, population location, structure and size, and conservation status of most of these species. Armenia is the most mountainous of the Transcaucasia republics, with an average elevation of 1,800 meters above sea level. Forests and woodlands cover less than a tenth of Armenia, arid land nearly a half, and 41.8 percent is pastures and meadows (1244500 hectares). Only ten percent of the country lies below 1,000 m, and its highest point is the 4, 090 m Mt Aragats. The country has an area of some 30,000 sq km. The variation in altitudinal range, overlying four distinct geological regions, has resulted in a great diversity of climates and adapted habitats relative to the size of the country. As a result, Armenia hosts exceptionally rich and globally significant biodiversity. The country is situated at a biological crossroads, on the junction of Circumboreal, Irano-Turanian and Southern Caucasic floristic regions.

Landscape diversity of Armenia and its relief peculiarities are a decisive factor enhancing plant diversity. On a territory of about 30 000 km2. Armenian flora comprises about 3,600 species of vascular plants, which makes about half of entire Caucasian flora, with both Caucasian and Iranian elements, distributed across desert and semi-desert, steppe, forest and alpine landscapes. Of the ca. 2500 species in the Armenian flora with a record of medicinal and/or economic use, ca. 50 species are used in the folk medicine and include both wild-collected (Crataegus sp., Hypericum perforatum, Artemisia absinthium) and cultivated (Chamomilla recutita, Mentha piperita, Crocus sativus) species (Fayvush, G., Danielyan T. & Nalbandyan A., 2004; Ghandilyan, P. A., Barseghyan, A. M. 1999).

Most components of social and economic development in Armenia can be related, directly or indirectly, to biodiversity e.g. in agriculture, biodiversity has provided sources of food, fodder and grazing for livestock, genetic variation for selection, etc; biodiversity has provided important natural raw materials like leaves, fruits, and berries for the food industry; in medicine, some plants are extremely important sources of natural and commercial remedies; forest resources are widely used in industry and construction, landscapes have important aesthetic and recreational value and provide the basis for tourism etc.

Armenia is undergoing economic transition from a centralized economy to free market conditions, and at this time it is important to recognize the inter-relations between human society and the natural environment. The wealth of biodiversity in Armenia is widely used in different spheres, and provides an important contribution to the social and economic development of the country.

Also, throughout the world, millions of people depend partly or fully on booth wild and manned biological diversity to fulfill their basic subsistence requirements. Among these crucial resources are plants, which in developing countries are important in providing rural people with building materials, fuel, fiber, medicine and also income. Although, animal and mineral materials are used, medicinal plants play a central role in traditional healing practices.

Besides the consumptive benefits, plants as integral components of ecosystems, contribute to the provision of non-consumptive benefits that add to making human life both possible and worth living. Some of the ecosystems non-consumptive services include the regulation of extreme temperatures, floods, droughts, the forces of wind and the provision of recreational, inspirational and educational sites (Diaz, S., Fargione, J., Chapin, F. S. & Tilman, 2006, Millennium Ecosystem Assessment 2005). Some of these non-consumptive benefits enhance not only human well-being, but also contribute to improving their mental health.

Medicinal plants represents and important asset to the livelihoods of people from developing countries. This is the case of Armenia, where most of the rural and also urban communities rely on medicinal plants for their primary healthcare needs and income generation.

Harvesting for domestic usage is not generally determined to the wild populations of medicinal plants in Armenia. Some of the traded/used species are rare, vulnerable, endangered, critically endangered and are declining from the wild (National Report on the State of Plant Genetic Resources in Armenia, 2008). However, the shift from subsistence to commercial harvesting is 10

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. posing unprecedented extinction threat to the wild populations of medicinal plants. In addition, there are even more extinction threats for wild medicinal plants population in Armenia, such as their habitat loss and degradations, environmental deteriorations and climate change, other anthropogenic threats (e.g. overgrazing, wild medicinal plants harvesting at an inappropriate times).

That is why the extensive observations and research of wild medicinal plants populations’ biological, growing, environmental, habitat and other characteristics is very vital for investigating their distributional changes over the years and improving their conservation through contributing its implication based on received essential data.

The effective research of wild medicinal plants, especially wild Melissa officinalis L. and Origanum vulgare L. (for their economic value of having versatile or universal utility), its sustainable use and conservation must be a priority for Armenia as it intends to reinforce its economic power through the conservation and sustainable use of biodiversity.

1.2. Statement of the Problem

One of the most important problems that exists it is medicinal plant resources dwindling processes in worldwide. It is believed that habitat destruction and unsustainable harvesting practices are the main causes for the loss of medicinal plants ((Diaz, S., Fargione, J., Chapin, F. S. & Tilman, 2006, Millennium Ecosystem Assessment 2005). This is true in most developing countries where the shift from subsistence to income generation harvesting has escalated the threats. Indiscriminate sitting of agriculture and urban development, invasive alien plants, unsustainable and over-use of resources and climate change are direct factors threatening the biological diversity of Armenia (National Report, 2009 & 2011).

Environmental deterioration, habitat loss, anthropogenic threats, global climate changes are ongoing processes. Likewise, their negative impact would foster to the impoverishment of wild medicinal plants populations and increase the risk of their extinction (National Report, 2008).

Approximately the half to Armenian flora needs to be protected; however, only 387 species are included in the Red Book of the country. 50 of the plant species on the verge of extinction (rare, extinguishing, and diminishing) are crop wild relatives and wild medicinal plants. Those species have high socio-economic value (wild medicinal plants), of which the knowledge of indigenous

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. people is kept and used in the people’s selection. However, the existing programs and implementing measures do not ensure the conservation of 70% of plants diversity. These conditions suggest the urgent necessity for biodiversity conservation.

The key threatening processes for the biodiversity in the country include: habitat loss and modification, environment deterioration and climatic changes, decline of natural pastures and grasslands due to intensive over-use and exploitation of useful plant populations, untimely and unsustainable harvesting of wild edible and medicinal plants, which significantly affects their natural regeneration. Records indicate that approximately 200 tons of wild edible and/or medicinal plants are sold in Yerevan markets each year (Cunningham, A. B., 2001, National Report 2011). Wild O. vulgare and M. officinalis are used by local inhabitants primarily in the form of an infusion (tea) as to treat disorders of the nervous and digestive systems (Gabrielian E, Zohary D 2004). What is more, these plants collection in inappropriate time destroy their re-growing process and prevent proliferation, which in its turn reflects negatively on their abundance and distribution and foster their eradication processes.

So, in Armenia both habitats and species have suffered from unregulated use. Due to unsustainable harvesting and destruction of natural habitats medicinal plant’ resources are continually dwindling in Armenia. Among species most at risk are plants of edible, medicinal or decorative use, and over-collection of such species has affected the semi-deserts, steppes and meadows in which they occur. The effects of habitat loss or modification are also evident at a local scale, and a number of species/populations (including wild medicinal plants) have been affected by activities such as local deforestation, construction and road building, especially in the central and northern regions of Armenia (Fourth National Report to the Convention on Biological Diversity, Republic of Armenia, 2009).

Anthropogenic threats to the biodiversity, such as overpopulation, deforestation and urbanization have simultaneously hindered research and increased the need for it. Also, the factors influencing the state of wild plant genetic resources in the country are directly or indirectly conditioned by the human impact upon biodiversity.

The use of Armenia’s biodiversity has mainly advanced with little thought as to how the reproduction of these bio-reserves can be ensured. As a result, degradation and even the extinction of certain wild species have occurred, resulting in an overall impoverishment of biodiversity. This

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. took place most prominently in the last century due to increased environmental pollution (industrial, agricultural, transport, energy) and intensive exploitation of forests, pastures and other ecosystems.

Environment deterioration and climatic changes increases of 4-6 C are predicted, along with declines in rainfall, resulting in increased risks of desertification in Armenia, at the end of XXI century by PRECIS model. It is predicted that, as a result of global warming, the average temperature in Armenia will rise by 2-3ºC, and rainfall will decrease by 10-15%, within the next 50-100 years. This will have devastating affect both cultivated and wild plants recourses conservation (Climate Change Problems, GEF, 2012).

In fact, climate change and temperature may lead to long-term irregularities in interspecific interaction and may alter plant populations’ dynamics, its structure and ecosystem functioning in the region (Miles L., Grainger A. & Phillips O. , 2004, Hughes, L; 2000).

Climate change is important because it not only represents a significant additional threat to plants biodiversity; it also requires a substantially increased use of wild edible plants to maintain resilience and adaptability in agro-ecosystems. Recent studies have suggested that, without deliberate interventions, over 20% of wild important edible or medicinal species are at risk and the threats to the genetic diversity present within plants, medicinal species, livestock and pollinators may be even greater (Parmesan, C. 2006, Parmesan, C., 1996).

Likewise, studies on possible effects of climate change on medicinal plants biodiversity and conservation status are particularly significant due to their value within traditional systems of medicine and as economically useful plants. So, it is important to predict how changing climate conditions would affect on wild medicinal plants populations’ dispersal throughout the country.

The second problem is the concern that the threatening processes caused by biogen or abiogen factors are ongoing. Therefore, it is becoming more and more crucial to decrease the negative impact of these factors and contribute to the conservation implication of wild medicinal plants in the country. It is critical to investigate in that complex environment the “behavior” of wild medicinal plants, such as distributional changes over the years in different habitat/ regions of Armenia in order to use the received data for predicting the population’s potential distribution.

The final problem for this research is that, only limited information on the genetic biodiversity, population location, structure and size, and conservation status of most of wild medicinal species, at this time available. In fact, the lack of knowledge about wild medicinal plants

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. past and current conditions are limited the possibilities of scientist to assess and project the wild medicinal plants ‘populations future distributional condition and in case of need the implications for conservation.

So, the study of the potential distribution of wild species and the population ecology in Armenia is of great importance for developing conservation strategies and sustainable use of biodiversity. That is why, it is very important to realize the research of wild medicinal plant populations current condition and distributional changes over the years as well as to predict populations distributional changes under the global climate change.

1.3. Novelty and Tasks of the Research

The topical task of this research is to identify changes in distribution of wild M. officinalis L. ; O. vulgare L. (Lamiaceae) and predict their potential distribution in Armenia as well as to ascertain definite features of populations’ ecology. The scientific novelty of this research is based on insufficient research and limited information on population location, ecology and conservation status of M. officinalis L. and (Lamiaceae) species in Armenia. In this work new populations of O. vulgare L. and M. officinalis L. Were found for the first time since 1979 in Armenia, which enlarged the area of this species especially in the south part of the country. In this research significant changes in the distribution and structure of these species have been identified and illustrated for the first time. In this research populations ecological characteristics e.g. size, structure, abundance, density, grow pattern etc. were investigated and described for the first time in Armenia. In this research habitat types and ecological facts were described for the first time of O. vulgare L. and M. officinalis L. species as well as processed in the distribution modeling for the first time of these species in Armenia. Also, this was the first time to model the species distributions and to define the most crucial environmental variables to predicting habitat suitability of Origanum vulgare L. and Melissa officinalis L. and their future condition in Armenia. The basic object of research is Origanum vulgare L. and Melissa officinalis L. (Lamiaceae) population’ ecology, changes in distribution and modeling in the territory of Armenia. Distribution of species in Armenia, populations grow pattern, size, structure, abundance, density, populations’ habitats were researched practically.

The main tasks of the study are the following: I. To identify changes and clarify distribution of Melissa officinalis L. and Origanum vulgare L. in Armenia; in the process of eco-geographic survey has been gathered and synthesizing ecological, geographic and taxonomic information of this wild valuable species population in Armenia during 2007-2011. In this respect it has been 1) re-located native populations of the important medicinal and culinary herb, Origanum vulgare L.; Melissa officinalis L. 2) located new populations in the territory of Armenia. The selection of survey territories across the country has been conducted based on historical records (Taghtajyan A. L., 1987; herbarium labels and geo- botanical, vegetation studies) as well as taking the advantage of statistical probability existence of new populations. II. To investigate population growth pattern and dynamics in sizes and structure based on Maxted et al. (1995) methodology. Populations are dynamic, and their size is the result of a balance of their biotic potential, and their maximum growth rate assuming ideal conditions and the limits placed on it by the environment (i.e., food, space, etc.). There are a number of environmental factors which can be limit the size of a given population: availability of nutrients, light, and competition with other species. There also are a number of factors which are components of the population itself which can affect how the population will respond to its habitat, the age structure, size structure, density, frequency, cover, and periodicity. III. To investigate populations habitat types and features of wild of M. officinalis L. and O. vulgare L. species in Armenia. To examine how the interactions between temporal variation in climatic variables and local environmental conditions affect population dynamics are therefore of major importance to enhance predictions of future local population viability and growth rates. IV. To model species distribution and identify habitat suitability level across the country. The modeling of species distribution has been conducted in the Institute of Systematic Biology at DU during 2012-2013 based on GARP modeling algorithm. The model commonly utilizes associations between environmental variables and known species’ occurrence records to identify environmental conditions within which populations can be maintained. Predicting species’ distribution has become an important component of conservation planning in recent years for assessing the impact of human activities on biodiversity, predicting the impacts of climate change on species’ distribution, preventing the spread of invasive species, searching rare or endangered species, planning new conservation areas, including identifying biological hotspots and setting conservation priorities (Godown and Peterson, 2000; Chen and Townsend Peterson, 2002; Store and Jokimäki, 2003; Sánchez-Cordero et al., 2005; Parviainen et al., 2009; Trotta-Moreu and Lobo, 2010).

The goal of this study is to survey the current ecological status of wild Melissa officinalis L. and Origanum vulgare L. species, causes of the declines and where is disappearing most quickly in order to develop relevant conservation actions to protect and restore wild medicinal plants populations and their native habitats. The concluding hypothesis of this study is that wild medicinal plants diversity will only be maintained in the Republic of Armenia if their conservation is mainstreamed into supporting policies and an enabling system is created for agro-ecosystems that promote food security, sustainable livelihoods and sustainable economic development. Institutional policies and strategies must be modified and capacities and market incentives that encourage and support agricultural biodiversity promoted.

1.4. Hypothesis of the Research Work

1. The basic hypothesis of the research is the presumption that O. vulgare L. and M. officinalis L. (Lamiaceae) distribution is decreased from the northern and central regions of Armenia where they were historically recorded. Remaining population would face a greater risk of extinction and are more susceptible towards climate change impacts.

2. The second hypothesis of the research is the presumption that there are more than historically known populations of O. vulgare L. and M. officinalis L. and these new populations are mostly located in the south and south-east regions of Armenia; where the abundance and distributional range of these species is expanding.

3. The third hypothesis of the research is the presumption that populations from south regions have higher carrying capacity determined by populations’ growth and habitat peculiarities.

4. The fourth hypothesis of the research is that population dynamics is linked to differences in habitat quality and in among years climatic variation. And, that the effects of climate depend on local habitat quality.

Chapter II: Review of the Literature

2. 1. Description of Definitions

Medicinal plants: The term refers to plants (tree, shrub or herb, fresh or dried) utilized in any forms of medicinal use which are thought to, proven to, promote well being or produce cures. It also includes plants that have multiple uses, i.e. as a source of perfumery (aromatic plants) or as the constituents of herbal teas, shampoos, soaps; cosmetics etc. and includes cultivated and wild materials, or medicinal plant is a plant which at least one of its parts contains substances that can be used for therapeutic purposes. Population: The group of organisms of one species, that are able to interbreed and living in a certain area. Population Ecology: Ecologists study many different aspects of ecosystems. One aspect that is of particular importance is population ecology. This field of study is concerned the dynamics of species populations and how these populations interact with the environment Population density: It refers to the number of individuals in a given area. Population size: Simply the number of individuals in the population at any given time. It is sometimes called abundance. Species: The group of populations whose individuals have the ability to breed and produce fertile offspring. Ecological niche: In ecology, a niche is a term describing the relational position of a species or population in an ecosystem. A niche refers to the way in which an organism fits into an ecological community or ecosystem. Through the process of natural selection, a niche is the evolutionary result of a species’ morphological (morphology refers to an organism’s physical structure), physiological, and behavioral adaptations to its surroundings. Distribution model: The method that predict the distribution of species

2.2. Brief Description of Melissa officinalis L. (Lamiaceae)

Common name is Lemon Balm, Latin name is Melissa officinalis L. (Pathrindzh in Armenian) other names are Bee Herb, Balm Mint, Garden Balm, Honey Plant, Melissa, Sweet Balm. Belong to Lamiaceae Family.

The word 'melissa' is from the Greek, meaning 'bee' due to the busy activity of the creatures around these plants. It was known to the Greeks as melisophyllon and to the Romans as apiastrum. It was often mentioned in the Greek and Latin classics and steeped in wine by the Greeks for fevers. It is said by the Arabs to bring intelligence to all who feed on it, animals and man alike.

The word balm (at one time 'bamm') is a shortened form of 'balsam' (meaning sweet smelling oil). Paracelsus (1493-1541) held lemon balm in high regard. He made a preparation called 'primum ens melissa' which he and others believed renewed their youth. It has long been regarded as useful for all complaints of a nervous or melancholy disposition.

It was listed in the London Dispensary of 1696 and it was remarked "An essence of balm, given in Canary Wine, every morning will renew youth, strengthen the brain, relieve languishing nature and prevent baldness". Balm steeped in wine was said to "comfort the heart and drive away melancholy and sadness".

The leaves were steeped in wine, then the wine drunk. The leaves were applied externally as a cure for the bites and stings of venomous animals and insects. This belief may owe some semblance of fact to the balsamic oils of aromatic plants which made good surgical dressings in their day due to antiputrescent action. Lemon balm is sometimes mistaken for Lemon Catnip (Nepeta citridora) which is an emmenagogue.

Lemon balm is a perennial herb grown (both by planting seeds and by vegetative propagation) for its lemon-scented foliage, used to flavor food, for preparation of herbal tea, and for use in traditional medicine (Pistrick, K. 1987; Small, E. 1995)

Habitat: The plant is native to South Europe, Northern Africa, Caucus, South western Asia, Persia and Mediterranean regions. However, lemon balm is grown all over the world. A member of the mint (Labiatae) family, which is native to southern, eastern and central Europe, especially in the mountainous regions. It has been cultivated in the Mediterranean area for about 2000 years and

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. known to be cultivated for its fragrance since the early 1600's. It is believed it was brought to Spain by Arab traders, then brought to Germany by Benedictine monks.

In Armenia, wild forms of M. officinalis occur in Aragatz, Idjevan, Zangezur and Meghri floristic regions occupying rather wet, deep soils at edges of oak-hornbeam forests, and in natural Platanus orientalis and Juglans regia groves, at altitudes ranging from 600 to 1600 m (Taghtajyan A. L., 1987; Gabrielian E, Zohary D 2004).

It is naturalized in England and the United States and was an important part of the early Colonial garden. When lemons were scarce, the dried leaves were added to jams and jellies. It grows one to two feet high, the rootstock is short (roots do not creep like other members of the mint family), the stem is square and branching with broadly ovate, hairy, toothed leaves which appear in pairs at each stem joint.

When brushed, they emit a lemon fragrance, the lowest part of the plant being richest in essential oil. The taste is also that of lemon. The flowers are white in loose, small bunches in the upper leaf axils. The fruit is a smooth nettle and the seeds within it are tiny.

Botanical Description: Lemon Balm is a perennial plant, growing from 70 to 150 centimeters in height. It has an angular stem and oppositely arranged leaves, dark to yellow green in color. In the spring and summer, clusters of small, light yellow flowers grow where the leaves meet the stem (Picture 1).

Picture 1. Melissa officinalis L.

The leaves are very deeply wrinkled and range from dark green to yellowish green in colour, depending on the soil and climate. If you rub your fingers on these leaves, your fingers will smell tart and sweet, like lemons. The leaves are similar in shape to mint leaves, and come from the same plant family (Pistrick, K. 1987.) Parts used: Leaves, fresh or dried

Useful components: Essential oil, flavonoids, chlorogenic, ferulic and caffeic acid;triterpenes Medicinal use: Traditionally, lemon balm was believed to ensure longevity. It was used as a mild sedative, and as a remedy for mild gastrointestinal problems. It was used as far back as the Middle Ages to reduce stress and anxiety, promote sleep, improve appetite, and ease pain and discomfort from indigestion (including gas and bloating as well as colic). Even before the Middle Ages, lemon balm was steeped in wine to lift the spirits, help heal wounds, and treat venomous insect bites and stings. Nowadays, Lemon Balm is mainly used in combination with other calming herbs in treatments of anxiety and insomnia.

Lemon balm is an excellent herb for the treatment of fever. It is very safe and can be used for infants and small children. It is also effective for treatment of eruptive fevers such as measles or chicken pox. And its antiviral properties help the immune system to eliminate viral infections. This makes it even more effective for treatment of fevers of viral origin.

Lemon Balm is also considered to be carminative, and can be used in treatment of excessive gas and indigestion. Used topically, an ointment made from it can be an effective remedy in cases of cold sores caused by herpes.

Several studies have found that lemon balm combined with other calming herbs (such as valerian, hops, and chamomile) helps reduce anxiety and promote sleep.

Few studies have examined lemon balm by itself, except for topical use. For example, in one study of people with minor sleep problems, those who took an herbal combination of valerian and lemon balm reported sleeping much better than those who took placebo. But it's not clear from this and other studies whether it is lemon balm or valerian (or the combination) that were responsible for the result. The same is true of several studies for anxiety, which used a combination of herbs to reduce symptoms.

2. 3. Brief Description of Origanum vulgare L. (Lamiaceae)

Common name is Origanum, Latin name is Origanum vulgare L. (Khnkatsaghik in Armenian) other names are Spanish Thyme, Wild Marjoram. Belong to Lamiaceae family. The name Origanum is derived from two Greek words, oros (mountain) and ganos (joy), in allusion to the gay appearance these plants give to the hillsides on which they grow. The Ancient Greeks gave this cooking plant the name Oros Ganos or Joy of the Mountain. Used in bride's bouquets it was also grown on tombs to ensure peace for the spirit of the dead.

Origanum was eaten by turtles, according to Aristotle,to counter-attack the poison after having swallowed a snake. Origanum was used to perfume baths, as an oil for hair and head massage but the therapeutic value was noted in Ancient Egypt where it was a disinfectant and a conserving agent. Today it alleviates depression. Oregano is a perennial, diploid (2n=2x=30) herb, widely grown (mostly by vegetative propagation) for its aromatic leaves that serve to flavor foods, for medicinal purposes, and for extraction of essential oils (Taghtajyan A. L., 1987; Gabrielian E, Zohary D 2004).

Habitat and Certain Growing Characteristics: Origanum is a perennial herb that is native to Europe, the Mediterranean and Asia.Native of dry, infertile and usually calcareous soils. Habitats include grasslands, hedge banks, road verges and scrub but not pasture as it is vulnerable to grazing. Origanum has limited means of vegetative spread and is strongly reliant upon seed for reproduction; as a consequence it can be lost from non disturbed sites where taller, more vigorous species can grow. However, the plant produces large amounts of seed which remains dormant in the soil allowing it to reappear following fire or physical disturbance. It is also a ready coloniser of bare or sparsely vegetated ground such as quarries. The long roots with numerous root hairs allow Origanum plants to exploit subsoil water in periods of drought. The plant prefers light (sandy), medium (loamy) and heavy (clay) soils, requires well-drained soil and can grow in nutritionally poor soil. The plant prefers acid, neutral and basic (alkaline) soils. It can grow in semi-shade (light woodland) or no shade. It requires dry or moist soil. The plant can tolerate strong winds but not maritime exposure.

In Armenia wild forms of O. vulgare are widely distributed. They abound in the lower, middle and upper altitude zones (between 1200 and 2200 m) in mountain steppes, meadows, bush thickets and forests, practically all over the country. O. vulgare is commonly collected in the wild. Selected

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. clones are also planted in the country, particularly in family gardens (Gabrielian E, Zohary D, 2004, Ghandilyan, P. A., Barseghyan, A. M. 1999).

Botanical Description: As it mentioned before, Oregano is a perennial growing to 0.6 m (2ft) by 0.8 m (2ft 7in). Trunk is almost horizontal, woody base up to 80 cm. high, reddish green in color. Some have leaves, others flowers. Foliage is long, oval, pointed-tipped leaves, grey-green in color (Picture 2).

Picture 2. Origanum vulgare L.

Upper part is sometimes red. Small bifarious and glabrous stem. Flowers are pinky-violet or reddish-white flowers inside a cob. The flowers are hermaphrodite (have both male and female organs) and are pollinated by Bees, lepidoptera. It is noted for attracting wildlife. It is in flower from Jul to September, and the seeds ripen from Aug to October. Fruits are four cylindrical, smooth achene with a brown surface enclosed in a persistent calyx (Taghtajyan A. L., 1987). Part used: Leaves Useful components: Rosmarinic acid, linalool, thymol, carvacrol, tannins, flavonoids, triterpenes. Culinary and Beauty Usage: Typical Mediterranean aroma mixed with garlic to flavor pizzas, tomatoes, eggs and cheeses, roast meats and mixed with breadcrumbs for stuffing. An essential oil from the plant is used as a food flavoring, in soaps and perfumery. The herb contains 0.15 - 0.4% essential oil and makes good herbal pillows and baths. A relaxing, purifying and detergent bath 22

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. infusion. Recommended for hair care, coughs, stomach problems, stress, excellent for sea-sickness and as a cream for inflammation. Purifies mouth and throat.

Medicinal Usage: The leaves and flowering stems are strongly antiseptic, antispasmodic, carminative, cholagogue, diaphoretic, emmenagogue, expectorant, stimulant, stomachic and mildly tonic. The plant is taken internally in the treatment of colds, influenza, mild feverish illnesses, indigestion, stomach upsets and painful menstruation. It is strongly sedative and should not be taken in large doses, though mild teas have a soothing effect and aid restful sleep. It should not be prescribed for pregnant women. Externally, oregano is used to treat bronchitis, asthma, arthritis and muscular pain. The plant can be used fresh or dried - harvest the whole plant (but not the roots) in late summer to dry and store for winter use. Oregano is often used in the form of an essential oil that is distilled from the flowering plant. A few drops of the essential oil, put on cotton wool and placed in the hollow of an aching tooth, frequently relieves the pain of toothache. This plant is one of the best natural antiseptics because of its high thymol content. The essential oil is used in aromatherapy to treat the same kinds of complaints that the herb is used for. Due to its anti-oxidant functions, Oregano could become helpful agent in treatment of cancer, heart disease and high blood pressure. It is also useful as a digestive aid, since it promotes salivation. Used externally, Oregano is successful in treatments of rheumatism, muscle and joint pain, sores and swellings. Oregano oil can help combat toothache.

2.4. Potential Distribution of Wild Medicinal plants in Armenia

Knowledge of the potential distribution of species is of great importance for developing strategies for conservation, public health and sustainable development. Predicting species’ distributions has become an important component of conservation planning in recent years, and a wide variety of modeling techniques have been developed for this purpose (Guisan, A., and W. Thuiller. 2005). Environmental niche modeling, alternatively known as species distribution modeling, (ecological) niche modeling, predictive habitat distribution modeling, and climate envelope modeling refers to the process of using computer algorithms to predict the distribution of species in

geographic space on the basis of a mathematical representation of their known distribution in environmental space (= realized ecological niche). The environment is in most cases represented by climate data (such as temperature, and precipitation), but other variables such as soil type, water depth, and land cover can also be used. These models allow for interpolating between a limited number of species occurrence and they are used in several research areas in conservation biology, ecology and evolution. The extent to which such modeled data reflect real-world species distributions will depend on a number of factors, including the nature, complexity, and accuracy of the models used and the quality of the available environmental data layers; the availability of sufficient and reliable species distribution data as model input; and the influence of various factors such as barriers to dispersal, geological history, or biotic interactions, that increase the difference between the realized niche and the fundamental niche. These models commonly utilize associations between environmental variables and known wild plants’ occurrence records to identify environmental conditions within which populations can be maintained. The spatial distribution of environments that are suitable for the species can then be estimated across a study region. This approach has proven valuable for generating biogeographically information that can be applied across a broad range of fields, including conservation biology, ecology and evolutionary biology. Example applications include: searching for rare or endangered species, planning new conservation areas, assessing the impact of human activities on biodiversity, predicting the impacts of climate change on species’ distribution, preventing the spread of invasive species, identifying disease vectors, understanding the abiotic needs of species, increasing agricultural productivity, and others (Peterson A.T. & Robins C.R. 2003). We are used to thinking about the occurrence of species in geographical space; that is, the species’ distribution as plotted on a map. To understand species’ distribution models it is important to also think about species occurring in environmental space, which is a conceptual space defined by the environmental variables to which the species responds. Soberón & Peterson (2005) distinguish three broad categories of factors that determine the distributions of species: abiotic environmental factors, biotic factors concerning interactions among species, and factors that affect the ability of species to disperse to different areas. The concept of environmental space has its foundations in ecological niche theory. The term ‘niche’ has a long and varied history of use in ecology (Chase, J.M., and M.A. Leibold 2003). Joseph Grinnell (1917, 1924) is credited with first using the term ―niche‖ to describe the

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. environmental conditions within which a species can survive and reproduce; these could include abiotic factors, such as temperature or rainfall, or interactions with other species. Charles Elton (1927), on the other hand, saw a species’ niche as its place or role within the ecological community, placing less emphasis on the abiotic conditions and more on relationships with other species (Vos, C., P. Berry, P. Opdam, H. Baveco, B. Nijhof, J. O’Hanley, C. Bell, & H. Kuipers 2008) and the impact that species have on the environment. However, the definition proposed by Hutchinson (1957) is most useful in the current context. Hutchinson defined the fundamental niche of a species as the set of environmental conditions within which a species can survive and persist (Hutchinson GE, 1957). Visualizing a species’ distribution in both geographical and environmental space helps us to define some basic concepts that are crucial for species’ distribution modeling. Notice that the observed localities constitute all that is known about the species’ actual distribution; the species is likely to occur in other areas in which it has not yet been detected. If the actual distribution is plotted in environmental space then we identify that part of environmental space that is occupied by the species, which we can define as the occupied niche. The distinction between the occupied niche and the fundamental niche is similar, but not identical, to Hutchinson’s (1957) distinction between the realized niche and the fundamental niche. With reference to the case of two species utilizing a common resource, Hutchinson described the realized niche as comprising that portion of the fundamental niche from which a species is not excluded due to biotic competition. The following diagram shows a hypothetical situation where a species distribution is controlled by just two environmental variables: temperature and moisture

Figure 1: The green and yellow areas describe the combinations of temperature and moisture that the species requires for survival and reproduction in its habitat. This resource space is known as the fundamental niche. The green area describes the actual combinations of these two variables that the species utilizes in its habitat. This subset of the fundamental niche is known as the realized niche. 25

The definition of the occupied niche used in this synthesis broadens this concept to include geographical and historical constraints resulting from a species’ limited ability to reach or re- occupy all suitable areas, along with biotic interactions of all forms (competition, predation, symbiosis and parasitism). Thus, the occupied niche reflects all constraints imposed on the actual distribution, including spatial constraints due to limited dispersal ability, and multiple interactions with other organisms. If the environmental conditions encapsulated within the fundamental niche are plotted in geographical space then we have the potential distribution. Notice that some regions of the potential distribution may not be inhabited by the species, either because the species is excluded from the area by biotic interactions (e.g., presence of a competitor or absence of a food source), because the species has not dispersed into the area (e.g., there is a geographic barrier to dispersal, such as a mountain range, or there has been insufficient time for dispersal), or because the species has been extirpated from the area (e.g. due to human modification of the landscape).

While suitable environmental conditions determine a species’ fundamental niche, biological factors such as competition tend to reduce the fundamental niche into the realized niche (Huston, M.A., 2002). The potential distribution of a species can be seen as the geographical expression of its realized niche at a particular time (i.e., where there is a fulfillment of both abiotic and biotic requirements) (Peterson AT, Papes M., 2007). In fact, a niche refers to the way in which an organism fits into an ecological community or ecosystem.

Through the process of natural selection, a niche is the evolutionary result of a species’ morphological (morphology refers to an organism’s physical structure), physiological, and behavioural adaptations to its surroundings (Thuiller W, 2003, Vandermeer, J. H. 1972).

A habitat is the actual location in the environment where an organism lives and consists of all the physical and biological resources available to a species. The collection of all the habitat areas of a species constitutes its geographic range. Habitats have many features or factors that are important to the organisms living there. Conveniently, we can divide habitat factors into two major groupings, physical factors and biotic factors.

In terrestrial habitats some important physical factors are elevation, steepness, slope direction, soil type, and water availability. Elevation affects air temperature and rainfall—higher elevations

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. are cooler and moister than lower ones. The steepness of a slope will affect the kind of soil that can form there and the amount of water that can soak into the ground after it rains.

Slope aspect (the direction a slope faces) is particularly important. In the northern hemisphere, south-facing slopes get more sun, are warmer and dryer, and thus support different vegetation than north-facing slopes. In aquatic habitats such characteristics as pH, salinity, dissolved oxygen concentration, temperature, and flow rate are important physical factors.

Biotic factors include all the other species that occur in the habitat. For an herbivore such as the desert bighorn sheep many of the grass, shrub, and herb species of the desert mountains constitutes its food source.

Finally, physical and biotic factors may interact to determine the quality of the habitat for a given organism. For example, the nutritional quality of plants available as food for herbivores, such as deer, is determined in large part by the quality of the soils present.

Obviously a species cannot survive without its natural habitat, except perhaps in a zoo. It follows then that the fundamental unit in the conservation of biodiversity is not the species but the habitat. Wild Melissa officinalis L. and Origanum Vulgare L. distributions are influenced by many factors in the country (Massot, M., J. Clobert, & R. Ferrière. 2008, Mander, M., 1998). One factor that limits the distribution of a species is its dispersal ability, i.e., how well individuals or their offspring can move from place to place. Thus the availability of suitable habitats within a potentially traversable distance affects the distribution of a species.

Another factor that limits distributions is a species’ tolerance to different environmental conditions. Factors such as temperature, moisture, and light have profound effects on species’ distributions (Guisan, A., & W. Thuiller, 2005).

Biotic Interactions with other species can also limit a species’ distribution. The fact that competition, predation and symbiosis with other species influence a species’ distribution was recognized a long time ago. Biotic interactions are thus shown to have important impacts on species distributions. We are left with a view of the natural system as a complex web of interactions and feedbacks between species, whereby changes to the distribution of a single species could have significant knock-on impacts on the distributions of many other species.

A species invading a new environment will encounter other species with which it has never had contact. If one of these is a predator that uses unfamiliar tactics, the invading species is likely to 27

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. be eaten. Similarly, if one of the species in the new environment uses the same kinds of resources as the invading species and if it is better able to compete for those resources, then the invading species will have trouble gathering enough resources to meet the needs of survival and reproduction. Obviously in both of these cases the invading species will not be very successful in extending its range (Diaz, S., Fargione, J., Chapin III, F. S. & Tilman, D. 2006).

2.5. Species Potential Distribution Modelling

Species distribution modeling has its origin in the late 1970s when computing capacity was limited. Early work in the field concentrated mostly on the development of methods to model effectively the shape of a species’ response to environmental gradients (Anderson RP, Lew D, Peterson AT, 2003, Anderson R.P., Gómez-Laverde M. & Peterson A.T. 2002a). Enormous advancements have occurred over the last decade, with hundreds if not thousands of publications on species distribution model (SDM) methodologies and their application to a broad set of conservation, ecological and evolutionary questions.

It is a central premise of biogeography that climate exerts a dominant control over the natural distribution of especially, wild plants. Evidence from the fossil record (Thorn J.S., Nijman V., Smith D. & Nekaris K.A.I. 2009; Davis, M.B. & Shaw, R.G., 2001) and from recently observed trends (Walther, G.R., Post, E., Convey, P., Menze, 1, A., Parmesan, C., Beebee, T.J.C., Fromentin, J.M., Hoegh-Guldberg, O. & Bairlein, F. 2002) shows that changing climate has a profound influence on species’ range expansion and contraction. It is therefore expected that predicted future climate change will have a significant impact on the distribution of species (Kozak, J.H. and J.J. Wiens. 2006, Lawler, J. J., D. White, R.P. Neilson, and A.R. Blaustein, 2006). Climate change presents unprecedented challenges for biological conservation that we should take into account while predicting the potential distribution of wild medicinal plants in the country. In fact, in Armenia the impact of climate change on biodiversity remains unclear but a temperature rise of 4-6C is predicted, which would result in increased desertification, and possibly to species extinction. Also, an indirect impact of pollution on the natural environment comes from the predictions of global warming. This is likely to severely affect wetland habitats and associated species, while changes in the distribution of habitats may affect the range and viability of a number of wild plants (Thorn J.S., Nijman V., Smith D. & Nekaris K.A.I., 2009).

There has long been agreement that climate change will force some species to shift their geographic ranges, or face extinction (Vos, C., P. Berry, P. Opdam, H. Baveco, B. Nijhof, J. O’Hanley, C. Bell, and H. Kuipers 2008). Proactive responses to ameliorate such impacts have been proposed. These include species translocation and construction of dispersal routes to assist species to track shifting climate (Hoegh-Guldberg, O., L. Hughes, S. McIntyre, D. B. Lindenmayer, C. Parmesan, H. P. Possingham, C. D. Thomas, 2008). Agencies are increasingly looking to modeled projections of species’ distributions under future climates to inform management strategies. Ecological systems face significant threats from climate change, and the need for effective responses is becoming a public policy imperative in many jurisdictions (Hannah, L., G. F. Midgley, T. Lovejoy, W. J. Bonds, M. Bush, J. C. Lovett, D. Scott, & F. I. Woodward, 2002). Species distribution models under global climate changes are empirical models relating field observations to environmental predictor variables based on statistically or theoretically derived response surfaces (Guisan, A. & Zimmermann, N.E. 2000). Species data can be simple presence, presence–absence or abundance observations based on random or stratified field sampling, or observations obtained opportunistically, such as those in natural history collections. However, there are often insufficient biodiversity data to support these activities in respect with projection of wild plants population (Yesson C, Brewer PW, Sutton T, Caithness N, Pahwa JS, BurgessM, GrayWA,White RJ, Jones AC, Bisby FA, Culham A, 2007; Canhos VP, Souza S, De Giovanni R, 2004). In response to this problem, many modeling techniques have been used and developed in an attempt to calculate the potential distribution of species as a proxy for actual observations. Potential distribution modeling is the process of combining occurrence data (locations where the species has been identified as being present or absent) with ecological and environmental variables (such as temperature, precipitation, and vegetation) to create a model of the species’ requirements (Anderson RP, Lew D, Peterson AT, 2003). Modeling strategies for predicting the potential impacts of climate change on the natural distribution of species have often focused on the characterization of a species’ bioclimatic envelope. A number of recent critiques have questioned the validity of this approach by pointing to the many factors other than climate that play an important part in determining species distributions and the dynamics of distribution changes. Such factors include biotic interactions, evolutionary change and dispersal ability.

2.6. The Applications of Species Potential Distribution Modeling 2.6.1 Conservation of Species

One of the main potential applications of species distribution models is in making decisions regarding the conservation of particular, often threatened, species. One use of species distribution models, which can be of immediate benefit, is in guiding surveys for species. A number of studies have demonstrated the utility of species’ distribution modeling for guiding field surveys toward regions where there is an increased probability of finding new populations of a known species (Guisan, A., O. Broennimann, et al 2006, Soberón J, Peterson AT 2005).

For example, Guisan et al. (2006a) used distribution models for alpine sea holly (Eryngium alpinum) in Switzerland to guide field surveys, leading to the detection of seven new populations. This is probably one of the most powerful applications of species distribution models, driving an increase in our knowledge of species‘ranges, knowledge which can be used to guide conservation decisions. Data from the new surveys can be used to build more accurate distribution models, which can in turn be used to direct further surveys, and so on (Guisan, A., O. Broennimann et al 2006).

Accelerating the discovery of new populations in this way may prove extremely useful for conservation planning in Armenia, especially in poorly known and highly threatened landscapes. Models can also be used to identify potential areas for species reintroductions (Peterson A.T. & Cohoon K.P. 1999). For example, one study (Klar et al. 2008) modeled the 29 distribution of European wildcats (Felis silvestris) in Germany. It was suggested that a suitable, but unoccupied, area could be used for reintroductions of the species (Herborg L.M., Rudnick D.A., Siliang Y., Lodge D.M. & MacIsaac H.J. 2007). If distribution models are to be used in this way, it is crucial that the models are very accurate, since the outcome of potentially very expensive projects is at stake. Given that there are many uncertainties about the determinants of species’ distributions, and consequently in models based on only a subset of these determinants, it is probably too soon to base important decisions solely on the outcome of species distribution models. On the other hand, where knowledge of species’ ecologies and distributions is lacking, as is the case for the vast majority of taxa (especially invertebrates), models could provide a good

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. starting point. In the face of rapid habitat degradation, the conservation of species may depend on their inclusion in networks of protected areas. Many studies have used distribution models to assess the coverage of particular species by protected areas (Walther, G.R., Post, E., et al 2002, Schweiger O., Settele J., Kudrna O., Klotz S. & Kühn I. 2008). These studies have often found that coverage is poor; in this case, the models can be used to propose additions and extensions to existing protected areas networks (Walther, G.R., Post, E., 2002). Species distribution models can also be used to infer the causes for species decline. For example, Southgate et al. (2007) developed distribution models for the bilby (Macrotis lagotis) in Australia to assess different hypotheses for its declines (Schweiger O., Settele J., Kudrna O., Klotz S. & Kühn I. 2008).

2.6.2. Predicting Future Distributions

Species distribution models can be used to predict how the distributions of species will change in the future as a result of climate and land-use changes. A distribution model is built for the current time, using contemporary species occurrence and climate data. This model is then updated to reflect predicted changes in the climate or land use in the future. Many papers have used distribution models in this way, mostly at regional or global scales (Thomas C.D., Cameron A., Green R.E., Bakkenes M., et al 2004, Bakkenes M., Alkemade J.R.M., Ihle F., Leemans R. & Latour J.B. 2002). Most have considered only changes in the climate, but land-use changes will also have important effects on the distributions of species (Thuiller, W., D.M. Richardson, et al 2005). The impact of climate change on biodiversity is a crucial issue in conservation. Species distribution models are being used increasingly to predict how species ranges will shift with changing climate. Evaluating the accuracy of these predictions is difficult because the changes have not yet occurred.

One possible approach is to test the ability of models to predict changes that have happened in the past. Increasingly, species distribution models have been used to predict the impact that climate change will have on species ranges in the future, on the basis of projections of future environmental conditions (Thomas C.D., Cameron A., Green R.E., et al 2004, Miles L., Grainger A. & Phillips O. 2004).

The accuracy of model predictions is generally evaluated by comparing models with data on the current distribution of species, before projecting into the future. However, there are at least five kinds of uncertainties associated with the use of distribution models to project into the future (Pearson, R.G., and T.P. Dawson 2003) which are ignored when evaluated this way. First, predictions of changes in the environment are unlikely to be entirely accurate and there may be variability in the predictions made by different models of climate change (Reilly J., Stone P.H., Forest C.E., Webster M.D., Jacoby H.D. & Prinn R.G. 2001). Second, species distribution models are correlative; therefore the predictor variables used may not directly influence the distributions of species (Pearson, R.G., and T.P. Dawson 2003) and may not correlate with the distribution of the same species in the future. Third, the realized distributions of species may be determined to a large extent by interactions with other species (Hampe, A. 2004, Pearson, R.G., and T.P. Dawson 2003). Including variables describing interactions among species has been shown to alter dramatically predictions of future distributions made by correlative models (Schweiger O., Settele J., Kudrna O., Klotz S. & Kühn I. 2008). Furthermore, experimental tests have shown that interactions among species can have a large effect on how species respond to climate change (Harmon J.P., Moran N.A. & Ives A.R. 2009). Fourth, models often assume that species can disperse to new suitable habitat as fast as is necessary to keep up with changes in the environment (Hampe, A. 2004, Pearson, R.G., and T.P. Dawson 2003) although some studies do include different dispersal-ability scenarios (Miles L., Grainger A. & Phillips O. 2004, Thomas C.D., Cameron A., Green R.E., et al 2004) or use dispersal models to simulate how ranges might shift given dispersal limitation. Studies of both plants and animals have shown considerable differences among taxa in the extent to which they are at equilibrium with the climate. Dispersal limitation may lead to considerable differences between predicted and observed distributions after climate change (Mustin K., Benton T.G., Dytham C. & Travis J.M.J. 2009, Best A.S., Johst K., Münkemüller T. & Travis J.M.J. 2007). Fifth, species may adapt to the changing environment rather than shifting their distributions (Skelly D.K., Joseph L.N., Possingham H.P., Freidenburg L.K., Farrugia T.J., Kinnison M.T. & Hendry A.P. 2007, Hampe, A. 2004). So, projecting distribution models in space will incur some of the uncertainties associated with predicting distributions in different time periods, including dispersal limitation, adaptation and

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. changes in interactions among species (Pearson & Dawson 2003; Hampe 2004; Pearson & Dawson 2004; Skelly et al. 2007). Although some evidence suggests that niches are conserved over evolutionary time (Peterson A.T. & Cohoon K.P. 1999), other authors have suggested that species could adapt rapidly to changing climates (Knouft J.H., Losos J.B., Glor R.E. & Kolbe J.J. 2006). So, projecting distribution models in space will incur some of the uncertainties associated with predicting distributions in different time periods, including dispersal limitation, adaptation and changes in interactions among species.

2.6.3 Predicting the Extent of Species Invasions

Models can be projected in space as well as in time, to predict distributions outside the area for which they were developed. Such projections can be used, for example, to predict where invasive species will be able to establish and survive outside their native ranges. A number of studies have used distribution models in this way, often finding that known invasions are predicted very successfully (Herborg L.M., et al 2007, Thuiller, W., D.M. Richardson, et al 2005). The full extent of the potential distribution is not predicted, but the model can be useful for identifying sites that may be suitable for reintroduction of a species (Pearce, J., and D.B. Lindenmayer 1998) and to foresee the possible changes in wild plants distributions over the country, or sites where a species is most likely to become invasive (if it overcomes dispersal barriers and if biotic competition does not prevent establishment (Anderson RP, Lew D, Peterson AT, 2003) Model predictions of this type also have the potential to accelerate the discovery of previously unknown species that are closely related to the modeled species and that occupy similar environmental space but different geographical space (Raxworthy, C.J., C. Ingram, N. Rabibosa, & R.G. Pearson 2007). On the other hand, in some cases the distributions of species in their invaded ranges are predicted very poorly by distribution models based on data from their native ranges (Randin C.F. et al 2006). Model failure may be caused by differences in the fundamental or realized niches in the invaded range (Broennimann O., et al 2007). Differences in realized niches may result from species not yet having reached equilibrium with climate in the new range owing to dispersal limitation, from the species not having been in equilibrium with climate in its native range, or from changes in interactions among species Steiner F.M., et al 2008). In species invasions, suitability of climate is only one of several factors that determine invasion success. Propagule pressure, characteristics of 33

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. the invading species, species composition of the invaded area and human influence can also be important (Thuiller, W., et al 2005).

2.6.4 Addressing Ecological and Evolutionary Questions

In addition to more applied problems, species distribution models can also be used to tackle more fundamental ecological or evolutionary issues. For example, they have been used to assess the extent to which climate drives distribution patterns compared to other factors, such as interactions among species (Araújo M.B. & Luoto M. 2007), dispersal limitation or habitat. Other studies have used distribution models to test whether niches are evolutionarily conserved by comparing modelled niches among closely-related species (Eaton M.D., Soberón J. & Peterson A.T. 2008, Peterson A.T. & Cohoon K.P. 1999). Despite this, potential distribution models provide a powerful tool for predicting species distribution in different geographical and temporal contexts, as well as for studying other aspects of evolution and ecology (Peterson AT, 2006). Additional uses of species’ distribution modeling include identifying potential areas for disease outbreaks, examining niche evolution (IPCC , 2001) and informing taxonomy (Thorn J.S., Nijman V., Smith D. & Nekaris K.A.I. 2009). So, predicting wild medicinal plants distribution by combining known occurrence records with digital layers of environmental variables can bring many benefits as discussed above. However, it has much potential for application in developing conservation strategy of relevant plants in Armenia

2.7. Algorithms for Modeling Plants’ Potential Distributions

There are many methods that can be used to model the potential distribution of species. Most are data-driven methods based on a correlative approach. The correlative approach tries to build a representation of the fundamental ecological requirements of a species based on the environmental characteristics of known occurrence sites (Diaz, S., Fargione, J., Chapin III, F. S. & Tilman, D. 2006). In this case, three types of input data are required to generate a model: occurrence data, environmental data, and algorithm-specific parameters. When projected into a geographical region, the resulting map will typically show areas that are ecologically similar to those where the species is 34

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. known to occur (Fleishman, E., R. Mac Nally, and J.P. Fay 2002). Community models follow a similar approach, except that they include data from other species belonging to the same biological group as the species being modeled (Elith J, et al 2006, Ferrier S, Drielsma M, Manion G, Watson G, 2002). Although there are many algorithms that can be used to create models, unfortunately no algorithm is suitable in all situations. Algorithm suitability is determined by a variety of factors including the number of occurrence points, availability of absence data, type and number of environmental variables, and purpose of the experiment (Peterson AT, Papes M, Eaton M. 2007). There has been considerable activity in the field of species’ potential distribution modelling. New algorithms and tools are frequently created and compared with existing ones (Phillips SJ, Anderson RP, Schapire RE 2006). However, there is a practical gap between devising a new algorithm and implementing it as a usable software package.

This gap is due to the different types of expertise required for the various areas involved in developing the software. Algorithms for modelling species’ potential distribution are usually created by people with a strong background in mathematics and ecology. However, developing a usable and robust application for a new algorithm involves considerable additional effort and also requires a deep understanding of geospatial data. Ideally, such applications should be able to deal with a range of tasks such as transforming between different geospatial reference systems, handling geospatial data in different scales and extents, reading and writing geospatial data in different file formats, and facilitating data visualization. The final software package should also ideally offer pre-analysis and post-processing tools, providing support for a range of protocols and data standards for sharing and retrieving occurrence and environmental data. These tasks are common to any implementation of potential distribution modelling software, but not directly related to the algorithm itself. To date, most software development has been carried out in isolation, producing separate software packages that are targeted to a single algorithm. This is the case for DesktopGARP (Phillips SJ, Anderson RP, Schapire RE 2006). There are drawbacks in having a different software package for each algorithm. In particular, users need to learn multiple software applications to use different algorithms. Getting familiar with the parameters and methodology of an algorithm is something unavoidable for users wishing to make proper use of the algorithm itself. Since each software

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. package has different input requirements, it is usually necessary to perform specific data conversions to run the same experiment across multiple packages. This makes it difficult to compare results. Fair comparisons should ideally be performed using identical input data and an identical computational environment. Although literature mentions potential distribution modelling frameworks that can run different algorithms (Garzón MB, Blazek R, et al 2006, Thuiller W 2003) none seem to be open source and easily available. Another important factor that we must be aware of is source-sink dynamics, which may cause a species to be observed in unsuitable environments. ‘Source-sink’ refers to the situation whereby an area (the ‘sink’) may not provide the necessary environmental conditions to support a viable population, yet may be frequently visited by individuals that have dispersed from a nearby area that does support a viable population (the ‘source’). In this situation, species occurrence may be recorded in sink areas that do not represent suitable habitat, meaning that the species is present outside its fundamental niche (Pulliam, H. R. 2000). We can logically expect this situation to occur most frequently in species with high dispersal ability, such as birds. In such cases it is useful to only utilize records for modeling that are known to be from breeding distributions, rather than migrating individuals. Because correlative species distribution models utilize observed species occurrence records to identify suitable habitat, inclusion of occurrence localities from sink populations is problematic. However, it is often assumed that observations from source areas will be much more frequent than observations from sink areas, so this source of potential error is commonly overlooked. One more thing to be aware of before we move on is that some studies explicitly aim to only investigate one part of the fundamental niche, by using a limited set of predictor variables. For example, it is common when investigating the potential impacts of future climate change to focus only on how climate variables impact species’ distributions. A species’ niche defined only in terms of climate variables may be termed the climatic niche (Pearson, R.G., and T.P. Dawson), which represents the climatic conditions that are suitable for species existence. An approximation of the climatic niche may then be mapped in geographical space, giving what is commonly termed the bioclimatic envelope (Huntley, B., P.M. Berry, W. Cramer, & A.P. Mcdonald 1995). Modelling strategies for predicting the potential impacts of climate change on the natural distribution of species have often focused on the characterization of a species’ bioclimatic envelope.

2.8. Desktop GARP (Genetic Algorithm)

Among the ENM methods developed, the Genetic Algorithm for Rule-set Production (GARP) (Stockwell & Peters 1999) has demonstrated its utility for predicting species distributions (Anderson et al. 2003). GARP develops a set of if-then statements (‗rules‘) that determines whether the species is predicted present or absent according to the environmental conditions of the grid cell in question (Stockwell D.R.B. & Noble I.R. 1992). Rules can be of three types: (1) envelope rules – presence or absence is predicted if the environmental variables fall within a certain range; (2) atomic rules – presence or absence is predicted for specific values of the environmental variables; and (3) logistic rules – presence or absence is predicted using a logistic regression function of the environmental variables (Stockwell, D.R.B., and D.P. Peters. 1999). GARP initially takes a random sample, with replacement, of 1250 species presence points and 1250 grid cells without presence records. These data are divided in half for model-building and internal model validation. A random set of rules is generated, and then these are modified by mutation (changes to the values of the environmental variables in the rules) and recombination (whole portions of rules are swapped). At each step the rules are tested against the internal validation data; rules that fit the data well are more likely to be retained (Stockwell, D.R.B., and D.P. Peters. 1999). The algorithm runs until improvement in accuracy falls below a certain threshold or until a maximum number of iterations have been performed. Since the starting set of rules is generated randomly, markedly different predictions can be made using exactly the same species and environmental data. One solution to this problem has been to develop a number of replicate models for each species, and then to sum these models to generate an index of predicted environmental suitability (Anderson R.P., Gómez-Laverde M. & Peterson A.T. 2002a). However, Anderson et al. (2003) found that the accuracy of models for the same species was very variable and suggested that only the best models should be selected for the final prediction. Accurate models should predict as being present as many of the species records as possible, and should predict as being present an area that approximates the true range size of the species in question (Anderson RP, Lew D, Peterson AT 2003). Anderson et al. (2003) propose identifying the 10 most-accurate models by selecting: 1) the 20 models that have the lowest numbers of presence locations predicted absent (omission error); and then 2) the ten of these models that have a proportion of background points predicted present 37

(commission index) closest to the median value. GARP has shown mixed performance in tests of its accuracy. Many studies have found that it models species’ distributions very accurately (Randin C.F., et al 2006, Peterson A.T. & Robins C.R. 2003, Peterson A.T. & Cohoon K.P. 1999). However, in comparisons of several techniques, GARP has generally been shown to perform relatively poorly (Pearson, R.G., et al 2007, Elith J, Graham CH et al 2006) and has a tendency to over-predict distributions (Peterson A.T. & Robins C.R. 2003). On the other hand, GARP has been shown to be relatively robust to small sample sizes (Solano E. & Feria T.P. 2007) but perhaps less so than Maxent (Pearson, R.G., C.J. Raxworthy, M. Nakamura, and A.T. Peterson 2007) changes to the values of the environmental variables in the rules) and recombination (whole portions of rules are swapped). At each step the rules are tested against the internal validation data; rules that fit the data well are more likely to be retained (Stockwell, D.R.B., &D.P. Peters. 1999). The algorithm runs until improvement in accuracy falls below a certain threshold or until a maximum number of iterations have been performed. Since the starting set of rules is generated randomly, markedly different predictions can be made using exactly the same species and environmental data. One solution to this problem has been to develop a number of replicate models for each species, and then to sum these models to generate an index of predicted environmental suitability (Anderson R.P., Gómez-Laverde M. & Peterson A.T. 2002a). However, Anderson et al. (2003) found that the accuracy of models for the same species was very variable and suggested that only the best models should be selected for the final prediction. Accurate models should predict as being present as many of the species records as possible, and should predict as being present an area that approximates the true range size of the species in question (Anderson RP, Lew D, Peterson AT 2003). Anderson et al. (2003) propose identifying the 10 most-accurate models by selecting: 1) the 20 models that have the lowest numbers of presence locations predicted absent (omission error); and then 2) the ten of these models that have a proportion of background points predicted present (commission index) closest to the median value. GARP has shown mixed performance in tests of its accuracy. Many studies have found that it models species ‘distributions very accurately (Randin C.F., et al 2006, Peterson A.T. & Cohoon K.P. 1999). However, in comparisons of several techniques, GARP has generally been shown to perform relatively poorly (Pearson, R.G., 2007, Elith J, Graham CH, Anderson R et al 2006) and has a tendency to over-predict distributions (Peterson A.T. & Robins C.R. 2003). On the other hand, GARP has been shown to be relatively robust to small sample sizes (Solano E. & Feria T.P. 2007) but perhaps less so than Maxent (Pearson, R.G., et al 2007).

Chapter III: Research Methods 3.1. Introduction

The territory of the research. Researches of distribution and populations’ ecology of O. vulgare L. and M. officinalis L. have been realized in 2007-2011 in the territory of the Republic of Armenia. Species identification was verified at the Department of Plant Taxonomy and Geography Herbarium of the Institute of Botany of the National Academy of Sciences of Armenia. Voucher specimens of plants collected at these population locations are stored in the Gene Pool Laboratory under the division of Botany Department Herbarium of Armenian National Agrarian University. Soil specimen is explored at the laboratory of Environmental Conservation and Research Centre of American University of Armenia. The modeling of species distribution has been conducted in the Institute of Life Sciences and Technology at Daugavpils University during 2012-2014. The methods of the research. During the study, different research methods have been used. Conceptually, we can categories them in two groups: 1) methods applied for the field observations 2) methods used in the laboratory work. During 2007-2011, field studies were conducted to examine populations’ ecology within the Republic of Armenia following the methodology of Maxted et al. (1995). The purpose of the field trips’ observations was to study populations ecology and to re-locate wild Melissa officinalis L. and Origanum vulgare L. populations of on the basis of historical (i.e. herbarium voucher) records, and to discover new populations. Quadrat sampling plot method was identified as the key element for the assessment of populations’ ecological characteristics, such as populations’ size, density, abundance, distributional pattern etc. The main steps for the assessment is to determine quadrate size and shape, its quantity per plot and the strategy should be used for placing the quadrats. During three consecutive years (2012-2014) laboratory works, GARP (genetic algorithm) has been identified the key modelling technique for determining Origanum vulgare L. (Oregano, Lamiaceae) and Melissa officinalis L. (Lemon Balm, Lamiaceae ) potential distribution in Armenia which would serve as the basis for foretelling future climate change impact on it. In fact, the correlative approach has been the basis of this modeling, where the appropriate environmental layers have been created through ESRI ArcGIS programs. In fact, the environmental layers and the plants’ actual distributions or occurrence records are considered to be the significant input data for the genetic algorithm. It is important to mention, that combination of data gathering (prior and

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. during the study period) triangulation techniques was used. The need to achieve objectivity, reliability and validity was the rationale for using triangulation techniques (Babbie, E., Mouton, J., Voster, P & Prozesky, B. 2001). In fact, in qualitative research, data generated from a single method are often denounced as biased. Therefore, the triangulation technique used in this study provided the unique opportunity to examine the same information from many angels to improve the legitimacy of the outcomes of investigations.

3.2. Eco-geographic Survey and Field Data Collection

The eco-geographic survey is the process of gathering and synthesizing ecological, geographic and taxonomic information of this wild valuable plant in accordance to Maxted et al. (1995) methodology. Field studies were conducted in 9 regions of Armenia, focusing on the central (Ararat, Aragatsotn, Kotayk, Gegharkunik), northern (Tavush, Lori, Shirak) and southern (Vayoc Dzor, Syunik) Regions. The selection of survey territories across the country has been conducted based on historical records (Taghtajyan A. L., 1987; herbarium labels and geo-botanical, vegetation studies) as well as taking the advantage of statistical probability existence of new populations. During 2007-2011, field studies were conducted to 1) re-locate native populations of the important medicinal and culinary herb; Melissa officinalis L. and Origanum vulgare L. 2) locate new populations, 3) study population ecology and habitats in the territory of Armenia. Principles of methodology of habitat description Maxted et al. (1995) were used for obtaining standard data and elaborating habitat characteristics of species populations. In this research overall 21 habitats were surveyed and exposed the description of 5 new habitats populated by M. officinalis L. and 4 new habitats populated by O. vulgare L..

Distributional data and taxonomy of wild M. officinalis L. ; O. vulgare L. are primarily obtained from various sources e.g. Floras and monographs, geo-botanical, phytosociological and vegetation studies, herbarium labels (Department of Plant Taxonomy and Geography, Botanical Institute of the Academy of Sciences of Armenia (http://www.sci.am/), Picture 3-4) biodiversity databases (Yesson C, Brewer PW, Sutton T, Caithness N, et al. 2007.) etc.

Picture 3. Herbarium voucher of wild Origanum vulgare L.,Vayoc Dzor Region.

Picture 4. Herbarium voucher of wild Melissa officinalis L., Tavush Region.

The species geographic distribution ranges extending outside Armenia involving Mountainous Karabagh and historical territory of Armenia under Turkish government. Thus, to ensure full representation of the environmental conditions associated with the species (Pearson and Dawson, 2003; Broennimann and Guisan, 2008; Beaumont et al., 2009) and to counter any sampling bias, global scale occurrence data was obtained from the GBIF data portal (http://data.gbif.org). All occurrence data from GBIF and Herbarium voucher were compiled and

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. records with obvious errors such as zero coordinates, points outside land boundaries and textual locality references without coordinates were discarded.

3.2.1. Actual Distribution Mapping

The plant actual distribution was installed in the herbarium labels. Voucher specimens of plants collected at these population locations are stored in Gene Pool Laboratory under the division of Botany Department Herbarium of Armenian National Agrarian University (ERE). Species identification was verified at Department of Plant Taxonomy and Geography Herbarium of the Institute of Botany of the National Academy of Sciences of Armenia by the supervision of senior specialist Kamilla Tamanyan. GPS maps of present and past population distribution were created.

The data collected during field surveys included latitude, longitude and altitude, description of location including administrative unit and nearest settlement. The actual distribution of the species across the country has been mapped and current occupied niche has been determined to serve essential input for the species modeling (Elith, J. and Leathwick, J.R. (2007)). The determination of geographical coordinates of the surveyed habitats was done with Garmin 60C GPS. In fact, it has been realized four measurements at different directions for per plot and given the average mathematical value of them.

Picture 5. The plant population location in Eghegis, Vayoc Dzor Region 42

Spatial explanatory variables were measures of weather and climate, landscape and topography and geographic location of wild M. officinalis L. and O. vulgare L. habitats. Ecological data that was observed in the field trips has been processed into the generation of environmental variables to serve as essential input materials for the ecological niche modeling. The outline of frequency and the timelines of field trips over the regions is presented in the following table.

Table 1. Frequency and Timeline of Field Trips

2007 2008 2009 2010 2011

Regions

July July July July

May May June May May June May June May June

April April April April April

August August August August

June July August

Tavush ○ ● ● ● ● ○ ● ● ● ○ ○ ● ● ○ ● ○ ● ● ● ○ ○ ● ● ○ ○ Shirak ○ ● ● ○ ○ ○ ○ ● ● ● ○ ○ ● ● ● ● ● ○ ○ ● ○ ○ ● ○ ○ Lori ○ ○ ● ● ● ○ ○ ● ● ● ○ ○ ● ● ○ ○ ○ ● ● ○ ○ ○ ● ● ●

Aragatsotn ○ ○ ● ● ● ○ ○ ● ● ● ○ ● ● ● ○ ○ ○ ● ● ○ ○ ○ ● ○ ○

Kotayk ○ ● ● ● ○ ○ ● ● ● ○ ○ ○ ● ● ● ○ ○ ● ● ○ ○ ● ● ○ ○ ○ ○ ● ● ● ○ ○ ● ● ○ ○ ○ ● ● ○ ○ ○ ● ● ● ○ ● ○ ○ ○ Gegharkunik Ararat ● ● ○ ○ ○ ○ ● ● ○ ○ ● ● ○ ○ ○ ● ● ○ ○ ○ ● ● ○ ○ ○ Vayots ○ ● ● ● ○ ○ ● ● ● ● ○ ● ● ● ○ ○ ● ● ● ● ○ ● ● ● ● Dzor Dzor Syunik ● ● ● ○ ○ ● ● ○ ○ ○ ● ● ○ ○ ○ ● ● ● ○ ○ ● ● ○ ○ ○

●-indicates the presence of field trips ○-indicates the absence of field trips

3.2.2. Quadrate Sampling Plot Method

Quadrate sampling plot method was identified as the key element for the assessment of populations’ ecological characteristics, such as populations’ size, density, abundance, distributional pattern etc. The main steps for the assessment is to determine quadrate size and shape, its quantity per plot and the strategy should be used for placing the quadrant. Quadrates or transects were used to randomly sample a portion of the habitat. A quadrate size of 50 cm x 50 cm, according to the guidelines of Maxted et al. was used. Quadrates are chosen randomly by using a random number

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. generator or a random number table to select coordinates. A sample is taken randomly in order to give the sampling area an equal chance of being sampled within the study plot. The purpose for collecting the samples randomly is to avoid biasing the data during the study. Data become biased when individuals of some species are sampled more frequently, or less frequently, than can be expected at random. Such biases can cause the population size to be either over- estimated or under-estimated, and thus lead to erroneous estimates of population size.

Each time a sample is taken randomly, it allows an equal chance for the sampling area to be sampled within the study area. To achieve this, a tape measure was placed along two sides of the area being studied and random coordinates were found. Each plot length was divided into the sampling intervals. The length of one side of the quadrate formed the sampling interval.

For instance, for a 10 x 20 plot and a 50 x 50 cm quadrate, the intervals were 0, 0.5, 1, 1.5, 2…and so on up 40. To cover one m2 four 50 cm x 50 cm square quadrates were needed in order to cover 2% of the total area being sampled along with a different plot (Table 2).

Table 2. Quadrate Random Sampling Coordinates.

Populations Study area, Quadtrates/ SI*/plot Species m2 plot Regions NSa Tavush Ijevan 206 16 68 Origanum vulgare L. Lori Novoseltsovo 190 15 42 Origanum vulgare L. Syunik Chakaten 200 16 40 Origanum vulgare L. Syunik Kapa Meghri 425 34 106 Origanum vulgare L. N New Highway Syunik Artsvanik 458 36 115 Origanum vulgare L. Vayots Dzor Jermuk 423 34 71 Origanum vulgare L.

Vayots Dzor Eghegis 326 26 59 Origanum vulgare L. Aragatsotn Aparan 159 12 40 Origanum vulgare L. Gegharkunik Lichk 156 12 52 Origanum vulgare L. Tavush Getahovit 87 7 24 Melissa officinalis L. Tavush Ayrum 75 6 30 Melissa officinalis L. Aragatsotn Orgov 136 11 45 Melissa officinalis L. Kotayk Garni 165 13 66 Melissa officinalis L. Vayots Dzor Jermuk 163 13 54 Melissa officinalis L. Syunik Artsvanik 169 14 56 Melissa officinalis L. Syunik Srashen 189 15 54 Melissa officinalis L. Syunik Shikahogh 45 4 18 Melissa officinalis L. Syunik Kapan 56 5 16 Melissa officinalis L. Syunik Tsav 162 13 36 Melissa officinalis L. Syunik Karchevank 102 8 20 Melissa officinalis L. NSa - Nearest Settlement. *-Sampling Interval

Along with quadrate measurements a mathematical formula was applied to find the average number of individuals per plot: N = (A/a) * n , (1) where, N is the estimated total population size, A is the total study area, a is the area of the quadrate and n is the number of organisms per quadrate. The plant height and number of stems for plants within each population were calculated by averaging measurements taken for plants that cover 2% of the total plot area with respect to randomly selected quadrates sampled within each plot.

3.3. Environmental Niche Modelling (ENM)

Environmental niche modelling, alternatively known as species distribution modelling, (ecological) niche modelling, predictive habitat distribution modelling, and climate envelope modelling refers to the process of using computer algorithms to predict the distribution of species in geographic space on the basis of a mathematical representation of their known distribution in environmental space (= realized ecological niche).

The environment is in most cases represented by climate data (such as temperature, and precipitation), but other variables such as soil type, water depth, and land cover can also be used. These models allow for interpolating between a limited number of species occurrence and they are used in several research areas in conservation biology, ecology and evolution.

The extent to which such modelled data reflect real-world species distributions will depend on a number of factors, including the nature, complexity, and accuracy of the models used and the quality of the available environmental data layers; the availability of sufficient and reliable species distribution data as model input; and the influence of various factors such as barriers to dispersal, geological history, or biotic interactions, that increase the difference between the realized niche and the fundamental niche. Environmental niche modelling may be considered a part of the discipline of biodiversity informatics.

During 2012 - 2013 consecutive years of laboratory work, Desktop GARP genetic algorithm is applied to model potential distribution/ ecological niche of M. officinalis L. and O. vulgare L . species based on bioclimatic envelope. Among the ENM methods developed, the Genetic 45

Algorithm for Rule-set Production (GARP) (Stockwell & Peters 1999, Stockwell, Peterson et al. 2002) has demonstrated its utility for predicting species distributions (Anderson et al. 2003).

The principle of the theory Elith et al., 2006 is considered in the modeling as the evolution through natural selection, and the idea that solutions to problems evolve the same way organisms evolves, estimates the probability distribution when only presence data is available for the analysis GARP calculates habitat suitability of M. officinalis L. and O. vulgare L. species’ occurrences based on correlative approach.

The environmental conditions that are suitable for a species may be characterized using either a mechanistic or a correlative approach. Mechanistic models aim to incorporate physiologically limiting mechanisms in a species’ tolerance to environmental conditions.

The most common correlative strategy for estimating the actual or potential geographic distribution of a species is to characterize the environmental conditions that are suitable for the species, and to then identify where suitable environments are distributed in space. For instances, we know that wild Melissa officinalis L. is known to thrive in reach, slight mechanical structure and acid pH bearing soils, then simply identifying locations with this characteristics ‘soils and high precipitation can generate an estimate of the species’ distribution. Although, there are a number of reasons why the species may not actually occupy all suitable sites (e.g. geographic barriers that limit dispersal, competition from other species).

Correlative approach of Desktop GARP aims to estimate the environmental conditions that are suitable for the species by associating known’ occurrence records with suites of environmental variables that can reasonably be expected to affect the species’ physiology and probability of persistence (Pearson et al. 2007, Chase, J.M., and M.A. Leibold 2003; Payne, K., Stockwell 2002). The principal steps required to build and validate a correlative species’ distribution model are outlined in Fig. 2.

Figure 2: Flow diagram detailing the main steps required for building and validating a correlative species distribution model by GARP algorithm.

GARP inputs point locations of observations of species and environmental layers from raster grids and produces maps of possible habitats. DesktopGarp software was selected because it is user friendly and by comparison with other modeling methods, published results suggest that GARP has greater predictive capability (Soberón J, Peterson et al. 2005, Payne, K., Stockwell 2002). Two types of model input data are needed for GARP:

1) known species’ occurrence records. So, it has been created relevant environmental layers through ESRI ArcGIS programs to be used with the plant actual distribution (occurrence records) as input data of GARP based on correlative approach. In fact, the plants actual distribution has been recorded based on conducted field trips during this research and historical records (e.g. herbariums at te Department of PIant Taxonomy and Geography, Botanical Institute of the National Academy of Sciences of RA). 2) A suite of environmental variables. ‘Raw’ environmental variables, such as daily precipitation records collected from weather stations, are often processed to generate model inputs that are thought to have a direct physiological role in limiting the ability of the species to survive.

Bioclimatic Data. Raster-based bioclimatic variables, derived from the WorldClim dataset (Hijmans et al., 2005) were used to model the climate component of the realized ecological niche of species. The WorldClim dataset provides baseline climate averages of monthly temperature and precipitation data for the period from 1960 to 1999. These data have been interpolated onto GIS grids and we chose to use data with a spatial resolution of 2.5arc-min because this provided a compromise between fine spatial resolution of ecological niche models and the presumed accuracy and precision of the coordinates for the species occurrences. The WorldClim dataset also provides grids of the 19 bioclimatic variables defined by Nix (1986) and Busby (1991), which are widely used to model the ecological niches of species. However, these bioclimatic variables include many that are highly correlated. To minimize the impact of multi collinearity and over-fitting on the stability and quality of models, we selected nine minimally correlated variables (i.e. correlation coefficient < 0.75) (Table 3). We did this by calculating the correlation matrix for the 19 bioclimatic variables and selecting representative variables from highly correlated clusters. The final set of selected variables were checked for correlations low enough (less than 0.75) to avoid problem of multicollinearity or over-fitting. The selected minimally correlated environmental variables reflect ecologically important annual totals (e.g. annual precipitation), seasonality effects (e.g. precipitation seasonality) and extreme environmental factors (e.g. maximum temperature of the warmest period) (Pulliam, H.R. 2000).

Table 3. List of global bioclimatic variables from WorldClim dataset

BIO 1 Annual Mean Temperature BIO 2 Maximum temperature of warmest Month (◦ C) BIO 3 Minimum temperature of coldest Month (◦ C) BIO 4 Temperature seasonality (coefficient of variation) (◦ C) BIO 5 Temperature Annual Range BIO 6 Annual precipitation (mm) BIO 7 Precipitation of driest period (mm) BIO 8 Precipitation of wettest period (mm) BIO 9 Precipitation Seasonality (Coefficient of Variation) (mm) ALT Topography (mean elevation)

Bioclimatic variables at 2.5 arc-min resolution

3.3.1. Genetic Algorithm (GARP) For Distribution Modeling

Although, there are number of different methods that have been used created to model the potential distribution of species. However, in this study period we applied GARP based on correlative species/populations distribution modeling approach. GARP or Genetic algorithms are based on the theory of evolution through natural selection, and the idea that solutions to problems evolve the same way organisms evolve.

GARP is a software package for biodiversity and ecologic research that allows the user to predict and analyze wild species distributions. DesktopGarp is a desktop version of the GARP algorithm. The acronym stands for Genetic Algorithm for Rule-set Production. GARP searches iteratively for non-random correlations between species presence and absence and environmental parameter values using several different types of rules. Each rule type implements a different method for building species prediction models. Currently there are four types of rules implemented: atomic, logistic regression, bioclimatic envelope, and negated bioclimatic envelope rules. GARP was originally developed by David Stockwell at ERIN Unit of Environment Australia.

A set of possible solutions to a problem are formed and, through a series of iterations, the solutions are modified and tested until the best solution is found (Pearce, J., and D.B. Lindenmayer 1998). The possible solutions are different types of rule sets. GARP creates ecological niche models for species. The models describe environmental conditions under which the species should be able to maintain populations. (Stockwell, D.R.B., and D.P. Peters. 1999) The models describe environmental conditions under which the species should be able to maintain populations. GARP inputs point locations of observations of species and environmental layers from raster grids and produces maps of possible habitat. By comparison with other modeling methods, published results suggest that GARP has greater predictive capability (Lawler, J. J., D. White, R.P. Neilson, & A.R. Blaustein. 2006).

In this study the principles of GARP modeling two types of rules is applied (1) envelope rules – presence or absence is predicted if the environmental variables fall within a certain range; (2) atomic rules – presence or absence is predicted for specific values of the environmental variables.

(1) Envelope rule originates from the BIOCLIM biogeography model and is based on the idea that species have ecological tolerances beyond which they cannot survive (Knouft J.H., Losos J.B., Glor

R.E. & Kolbe J.J. 2006). A model is developed by enclosing the range of the environmental values in an envelope where species may occur. If a point is outside the range of tolerance, then the species is predicted to be absent. In the study, the potential habitats of species is predicted based on the envelop rule by applying all environmental variables together to produce the areas where the species populations may exist.

(2) Automatic rules are developed when only a single variable within the precondition of the rule is used. An example atomic rule would be: if the average snowfall is 12 inches then the species is present. This rule is used with all 9 separate variables and then by the coincidence of 10 best models were summed and ranked by the criteria of how many times each model predicted the same pixel within the exact area, classified as: High 7-9 times, Moderate habitat suitability 5-7; Low, 1-5 presence agreement.

3.3.2. Application Techniques of GARP Program

GARP Parameters. It is important to mention that GARP modeling system works through a set of eight programs: rasterize, presample, initial, explain, verify, predict, image and translate (Leathwick, J.R., Elith, J., and Hastie, T. 2006). The first two steps, rasterize and presample, prepare the input data for use in GARP. Rasterize converts species point data into contiguous raster layers. This step compresses information by clearing the data of duplicates caused by localized intensive sampling. Presample takes the newly created raster layers and creates training and testing data sets by randomly sampling the data set prepared in rasterize. The training set is necessary to construct a model while the testing set allows for the assessment of the model’s accuracy.

The DesktopGarp contains a window where we have specified all the parameters and data to be used in the experiment. Below is a sample of the interface. GARP inputs point locations of observations of species and environmental layers from raster grids and produces maps of possible habitat. By comparison with other modeling methods, published results suggest that GARP has greater predictive capability (Payne, K., Stockwell, D.R.B., 2002).The species data points panel handles the species/populations occurrence (point) data.

The Species Data Points panel handles the species/populations occurrence (point) data. A sample of this area is shown below. Sample sizes used for projecting distributional modelling ranged from 34 (Melissa officinalis L.) to 67 occurrence points or localities (Origanum vulgare L.) based on conducted field studies from 2007-2011

Figure 3. Desktop Garp Parameters

New predicted localities information can be entered by clicking the Upload Data Points button. It will open a dialog box to specify the location of the occurrence data file. Currently three formats are supported: Comma delimited, MS Excel Spreadsheets and ArcView Shapefiles. Comma delimited and Excel files should contain three columns: the first one for species/populations’ study areas’ name, the second for longitude, and the third for latitude. The list box will present all species/populations loaded and the number of data points (in parenthesis) for each one. The check box to the left of each species in the list allows the user to control which species from the list will be used in the experiment.

In this panel, the user can also specify two parameters that define how the data will be sampled and used. The first option allows the user to specify what percentage of points will be used for training, i.e., model construction. The remaining points will be used for testing. If 100% is specified for training, no significance test will be performed on the models.

The second option allows the user to specify a minimum number of points to be used for training. To enable it, check the box to the left of this option. When enabled, it will override the percentage value and use at least the specified number of points for training. This option is useful for species with few data points because it forces the program to use a minimum number of points for analysis. The algorithm typically does not perform well with fewer than 20 data points for training.

Optimization Parameters. On the Optimization Parameters panel, it is possible to specify some parameters that control the overall behavior of the genetic algorithm. A sample of this panel is shown below.

Figure 4. Optimization Parameters.

The number of runs per experiment defines how many times each distinct task will be performed within the experiment. For example, for two populations and 10 runs per experiment, 20 runs in that experiment will be executed: 10 for the first populations and 10 for the second one.

The convergence limit establishes a stop condition for iterations within the genetic algorithm. Its behavior varies depending on how difficult or easy the problem is. Usual values are between 0.01 and 0.10. If this parameter is set to 0, the algorithm will stop only when the maximum number of iterations is reached. Max iterations value establishes another stop condition for the genetic algorithm. It forces the optimization to stop at the specified iteration, even if the convergence limit 52

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. has not been reached yet. More iteration tends to yield more stable results. Usual values are between 100 and 1000. The rule type checkboxes allow the user to specify which algorithm is used to produce rules in the species model. The all combinations checkbox generates one task for each combination of the checked rules. For example, if range, logic and atomic rules are checked, DesktopGarp will create tasks where only each of those rules are used, then one for range and logic rules, one for range and atomic rules, one for logic and atomic rules, and one for all three rules combined. This is useful for analyzing the impact of each particular rule on the results. The labels below the checkbox show how many combinations will be created and also the total tasks or runs that will be executed (combinations time’s runs).

Native Range Datasets. The Native Range Datasets or Environmental Layers panel allowed us to define the environmental coverage that is used as input for the prediction. Details about the environmental database and the software parameters used in GARP are described elsewhere (Gurgel-Gonçalves & Cuba 2009). The algorithm is correlated the input data points to the values on those layers to get the final prediction. The principle capabilities of GARP’s layer application were used for the species modeling in Armenia. First, were selected all layers in the optimization in order to generate a single result. The last two alternatives using combinations of layers were applied for determining which layers are important to per species, which is essential for analyzing population’s future distributional shifts under global climate change impact.

Figure 5. Environmental Layers, according to ESRI Arc GIS program 53

Once the dataset has been chosen, DesktopGarp will automatically list all layers present on that dataset on the layers to be used list box. There, the user can control which layers will be used by clicking on the checkbox that appears to the left of each layer name. Below the layers list, there are three radio buttons that define how the selected layers will be used. The first one, all selected layers, will force DesktopGarp to use all selected layers in the optimization.

All combinations of selected layers will cause the experiment to have one task for each possible combination of the selected layers. The all combinations of selected size N radio button have similar effect, but will limit the experiment to the combinations that contains exactly N layers. The last two alternatives using combinations of layers are useful for determining which layers are important to a species. A method for analyzing that would be using linear multiple regressions to predict the error values (omission and commission), using the information on whether a particular layer was used on a task as an independent variable.

Output Parameters. The Output panel specifies the output prediction map format and the output directory for maps and other generated documents.

Figure 6. Results

The prediction maps can be generated in three formats 1) bitmaps, 2)MS Windows bitmaps, with extension ".bmp", 3)ASCII raster grids: ASCII text format, with extension ".asc", 4)ESRI Arc/Info grids: ESRI proprietary format for grid spatial data storage and management.

A separate directory is created for each grid. Another important file that is stored on the output directory is the file result.xls which stores a summary of all tasks, error messages, result parameters, statistical tests, accuracy, and more.

Projections Datasets. In the Projection Datasets panel, the user specifies which datasets will be used in the projection phase of the experiment. A sample picture of this panel is shown below.

Figure 7. Projection layers, for the predicted map

At the end of each task, DesktopGarp will project the rule set obtained during optimization onto every dataset specified on the Current Datasets list. DesktopGarp will also project the rule set onto the native range dataset, defined on the Native Range Dataset panel.

A list of available datasets is shown on the Available Datasets combo box. The Add button adds the dataset selected on the combo box. To remove a dataset from the Current Datasets list, highlight it and then click the Remove button. In Fact, using different datasets on the experiment is useful when researching invasive species, climate changes, and time analysis.

Create Custom Datasets by Dataset Manager. It is necessary to create a custom dataset that is compatible with DG format. There is a tool called Dataset Manager that is installed with DG that can be used for creating new datasets. Dataset Manager Interface with an open dataset is shown below:

Figure 8. Illustrated window for using the genetic algorithm.

The Dataset Info panel allows the user to set generic metadata about the dataset. On the bottom of this panel, there are some information on the dataset, such as its envelope and grid cell count and size.

To produce DG datasets, this tool expects the layers to be in ASCII Raster Grid format, and placed on the same directory. They must have the same geographical boundaries and the same cell size, i.e., if you stack the layers together, all cells must match perfectly on top of each other.

3.4. Methods of Statistic Analysis of Data Processing

Statisctical processing of the results was carried out in the laboratories in Daugavpils University of the Department of Computer Sciences. The Principal component analysis (PCA) statistical technique was applied to convert a set of observations of possibly correlated variables into a set of values of linearly uncorrelated variables called principal components. The number of principal components is less than or equal to the number of original variables.

This transformation is defined in such a way that the first principal component has the largest possible variance (that is, accounts for as much of the variability in the data as possible), and each succeeding component in turn has the highest variance possible under the constraint that it is orthogonal to (i.e., uncorrelated with) the preceding components. The principal components are orthogonal because they are the eigenvectors of the covariance matrix, which is symmetric. PCA is sensitive to the relative scaling of the original variables.

This technique was applied in the process of identifying the effects of environmental factors in different habitats of population’s dynamics. Also, to explicit the correlations between the impact of environmental factors and habitat factors on populations’ dynamics stochastic multiple and principal component analysis among the populations dynamics of per year were formulated.

Furthermore, we conducted a principal component analysis (PCA) with all nine environmental variables from the occurrence areas of Melissa officinalis L. and Origanum vulgare L. species. PCA reduces the dimensionality of the original set of variables with little loss of information by transforming the original variables into a new set of independent components (Robertson, M. P., N. Caithness, and M.H. Villet. 2001, Foottit & Sorensen 1992).

The components that accounted for the majority of the total variance were examined and the most highly loaded variables were analyzed. This approach was used to identify decisive environmental variables that influence most on the geographical distribution and abundance of the species

Chapter IV. RESULTS AND DISCUSSION

4.1. Changes in Distribution of wild Melissa officinalis L. and Origanum vulgare L. Populations During the Last Decade in Armenia

In respect with conducted field trips it has been identified changes of wild M. Officinalis L. and O. vulgare L. populations’ distribution and has been determined the geographical coordinates of each population irrespective of its present condition. Fig. 9 & Fig. 10 illustrate the distribution of populations of the species that were either re-located on the basis of historical records or newly discovered as a result of field surveys in areas with appropriate habitat. In some cases, historical populations could not be re-located and were presumed extinct in these locations (Table 4).

Figure 9. Changes in Distribution of Melissa Officinalis L. in Armenia for the Last Decades.

Figure 10. Changes in Distribution of Origanum vulgare L. in Armenia for the Last Decades.

During the researches and eco-geographic surveys from 2007-20011 5 new populations of M. officinalis L. and 4 new populations of O. vulgare L. were documented in Armenia. Although historical records indicated Melissa Officinalis L. (dating from 1925) and Origanum vulgare L. (dating from 1920) had been widely distributed across the country, nearly half of the populations recorded from the northern regions of M. Officinalis L. and central and northern regions of O. vulgare L. could not be re-located.

The following historically documented populations of O. vulgare L. in Urcadzor ( Ararat Region ) dating from 1964, Aghveran (Kotayk Region) dating from 1935, Narashen, Nor Bayazet (Gegharkunik Region) dating from 1923 from the central regions and Dilijan near Vartaget (Tavush Region) dating from 1961, Gamzachiman (Margahovit, Lori Region) dating from 1961 etc. from northern part could not be re-located. From the northern regions more than half of the populations of M. Officinalis L. no longer existed in cited locations. E.g. Alaverdi (documented from 1945) and Akhtala (documented from 1925) populations in Lori Region, as well as Aygedzor (documented from 1982) and Ijevan (documented from 1983) populations in Tavush Regions have not been relocated during field studies and presumed to be extinct (Table 4).

Table 4. Geographical Coordinates of Wild M. officinalis L. and O. vulgare L. Populations in Armenia

Regions NSa Altitude, m Latitude, N Longitude, E Species Aygedzor* 719 40°72'45" 45°44'05" Melissa officinalis L. Ijevan* 767 40°52′32″ 45°8′57″ Melissa officinalis L. Tavush Getahovit 905 40°90'03" 45°13'79" Melissa officinalis L. Ayrum▲ 495 41°19'02" 44°90'38" Melissa officinalis L. Ijevan 730 40°87'55" 45°14'91" Origanum vulgare L. Dilijan 1276 40°37′22″ 44°41′37″ Origanum vulgare L.

Alaverdi* 987 40° 05′ 32″ 44°29′21″ Melissa officinalis L. Akhtala* 737 41° 15′ 03″ 44°77′06″ Melissa officinalis L. Lori Novoseltsovo▲ 1490 41°05' 51" 44°27'61" Origanum vulgare L. Margahovit* 1643 39° 46′ 11″ 44° 38′ 14″ Origanum vulgare L. Gegharkunik Lichk 1940 40°16' 36" 45°22'77" Origanum vulgare L. Gavar* 1827 40° 17′ 22″ 45°19′ 33″ Origanum vulgare L. Aragatsotn Orgov 1610 40°58' 91" 44°35'72" Melissa officinalis L. Aparan 1900 40°58' 91" 44°35' 72" Origanum vulgare L. Kotayk Garni 1310 40°11' 10" 44°73'16" Melissa officinalis L. Aghveran* 1503 40° 29′ 43″ 44° 31′ 44″ Origanum vulgare L. Ararat Urcadzor* 986 39° 33′ 35″ 44° 43′23 Origanum vulgare L. Jermuk▲ 2104 39°84' 16" 45°67' 22" Melissa officinalis L. Vayots Dzor Jermuk 2098 38°55' 92" 44°44' 10" Origanum vulgare L. Eghegis▲ 1617 390 45/ 35// 450 21/ 43// Origanum vulgare L. Srashen▲ 1025 39°07'94" 46°50'11" Melissa officinalis L. Kapan 795 39°12′04″ 46°24′54″ Melissa officinalis L. Shikahogh 957 39°09'57" 46°47'42" Melissa officinalis L.

Artsvanik▲ 1530 39°26'87" 46°48' 99" Melissa officinalis L. Syunik Tsav▲ 1130 39°05 '00" 46°45 '41 " Melissa officinalis L. Karchevank 1057 38°53′47″ 46°10′44″ Melissa officinalis L. Chakaten 1050 39°18' 30" 46°43' 21" Origanum vulgare L. Kapan Meghri New Highway▲ 972 39°19' 12" 46°43' 09" Origanum vulgare L. Artsvanik▲ 1666 39°28' 29" 46°48' 48" Origanum vulgare L. ▲-indicates new population; *- indicates extinct population; a- Nearest Settlement. GPS Map 60CS device has been used to record geographical coordinates by averaging five records of per plot

It has been discovered and documented 5 new populations of M. officinalis L. and 4 new populations of O. vulgare L. species in respect with conducted field trips and researches (Table 4). 4 new populations of M. officinalis L.: Artsvanik, Srashen, Tsav (Syunik Region) and Jermuk (Vayoc Dzor Region) were located in the south and south-east regions of the country. On the other hand, a new small population was documented in Ayrum located in the northern Region of the country.3 new populations of O. vulgare L.: Eghegis (Vayoc Dzor) and Kapan Meghri New Highway, Artsvanik (Syunik Region) were located in the south and south-east regions of Armenia. Only, one new population was located in Novoseltsovo (Lori Region) from the northern part of Armenia. The collected data suggests that the abundance and distributional range of O. vulgare L. and M. officinalis L. species is expanding, particularly in the south-east regions of Armenia. The following populations of M. officinalis L.: Orgov (documented from 1991); Garni (documented from 1980) and of O. vulgare L.: Aparan; Hrazdan; Lichk were relocated during the study from the central regions of Armenia (Table 4). However remaining populations in these regions displayed trends of reduction in overall size, plant number and fragmentation during the study period. Historically documented other populations of M. officinalis L. e.g. Kapan (recored from 1959), Shikahogh (documented from 1986), Meghri near Karchevank (documented from 1958) were relocated in Syunik Region from southern part of Armenia. So, according to the historical records the species are wildly distributed across the country. However, more than half populations from the central and northern regions could not be re-located and presumed to be extinct in the result of field trips and eco-geographic surveys from 2007-2012. There are number of biotic, abiotic factors that influence on wild medicinal plants distribution and changes over the last decade in the country. The intrinsic and delicate interactions of biotic and abiotic factors and their interrelationships between wild medicinal plants’ populations in the different habitat have revealed the changes in the distributions of the wild Melissa officinalis L. and Origanum vulgare L. populations in the country throughout last decades. The abiotic factors, such as temperature, soil, light, nutrients, and water in the habitat is an example of a limiting factor of population’s dispersal ability. Population within an ecosystem will experience some kind of limiting factors to prevent continuous growth and distribution. Environmental changes (i.e drought, famine, human destruction) results in decreased rates of physiological processes, lowering the potential for survival, growth, reproduction and distribution.

Populations will undergo acclimatization to adjust to the new limiting factors through changing their behavior or physiology (Eugene P. Odum, 1971) (see in Figure 11).

Figure 11. Limiting Factors and the Plant Physiological process in Ecosystem

So, we may conclude that wild Melissa officinalis L. and Origanum vulgare L. populations located in the northern and central regions of the country have most probably undergone within limiting factors and experienced acclimatization (Hannah, L., G.F. Midgley, G. Hughes, and B. Bomhard, 2005, Davis, M.B. & Shaw, R.G., 2001). In addition, Melissa officinalis L. and Origanum vulgare L. populations of being not sufficiently capable to adjust in an environment with limiting factors ceased to exist over the last decade in the mentioned localities of the country (Harmon J.P., Moran N.A. & Ives A.R., 2009).

Other factors include geographical space, (if the plant can only survive within a given ecosystem, the size of that ecosystem will prevent further population increases), predation, climate, competition (for prey, food, mates) etc. A drastic change in the environment destabilizes or even exterminates a population. Natural calamities such as earthquake, volcanic eruptions etc. cause drastic changes in the environment leading to the destruction of the resources. The Spitak earthquake occurred in the northern region of Armenia in 1988, could be one of the reasons to cause to the extermination of wild medicinal plants populations with nearby locations.

Wild medicinal plants’ population distributions are influenced by many biotic factors too (Harper, J.L. 1977). One factor that limits the distribution of a species is its dispersal ability, i.e., how well individuals or their offspring can move from place to place. Thus the availability of suitable habitats within a potentially traversable distance affects the distribution of a species.

Another biotic factor that limits distributions is a species’ tolerance to different environmental conditions. Factors such as temperature, moisture, and light have profound effects on species’ distributions. It is notable that biotic interactions with other species (competitions, within the plant individuals and other species too) can also limit a populations’ distribution (Araújo M.B. & Luoto M. 2007). The fact that competition, predation and symbiosis with other species influence a species’ distribution was recognized a long time ago. A species invading a new environment will encounter other species with which it has never had contact. If one of these is a predator that uses unfamiliar tactics, the invading species is likely to be eaten. The immigration and emigration also can affect on the distribution and size of the plant populations (seeds, vegetative or generative organs etc.).

In general, the ability to tolerate changes in weather, biotic and abiotic factors is determined by the resilience of population (Heikinnen, R.K., M. Luoto, R. Virkkala, R.G. Pearson, and J-H. Körber, 2007, Harper, J.L., 1977). Obviously, the resilience of the plants populations from the northern and central locations is less to compare those from south locations.

Antropegenetic negative impacts are also affected highly on wild plants distribution and population growing in the country. Thus, some anthropogenic threats that were identified as part of the study included poor land management (erosion, overgrazing), increasing population pressure (impact of livestock overpopulation, improper human development), and excessive or inappropriate collection for the purposes of local sale/usage (due to lack of knowledge/training of collectors) were revealed and discussed during the study to illustrate the changes in distribution of O. vulgare L. and M. officinalis L. species across the country during the last decade in Armenia.

Due to unsustainable harvesting and destruction of natural habitats medicinal plant’ resources are continually dwindling in Armenia. Among species most at risk are plants of edible, medicinal or decorative use, and over-collection of such species has affected the semi-deserts, steppes and meadows in which they occur (National Report on the State of Plant Genetic Resources in Armenia

2008). Especially in the northern and central regions, it was observed that the wild harvesting of medicinal plant materials (e.g., O. vulgare L., Hypericum perforatum L., and Melissa officinalis L.) was conducted at inappropriate times for sale at Yerevan markets and this negatively impacted population sustainability. Records indicate that approximately 200 tons of wild edible and/or medicinal plants are sold in Yerevan markets each year (National Report on the State of Plant Genetic Resources in Armenia, 2008).Wild O. vulgare and M. officinalis are used by local inhabitants primarily in the form of an infusion (tea) as to treat disorders of the nervous and digestive systems (Gabrielian E, Zohary D, 2004). The effects of habitat loss or modification are also evident at a local scale, and a number of species/populations (including wild medicinal plants) have been affected by activities such as local deforestation, construction and road building, especially in the central and northern regions of Armenia. The study investigated a small area of high floristic diversity close to Hagartsin monastery (Dilijan, Tavush Region) has been destroyed following construction and industrial use close to the site, while around 180 species of plants and many populations of wild edible and medicinal herbs (e.g. O. vulgare L. and M. officinalis L. , Carum carvi L., Hypericum perforatum L. etc populations) that once occurred have now disappeared (Fourth National Report to the Convention on Biological Diversity, Republic of Armenia, Yerevan, 2009).

4.2. M. officinalis L. and O. vulgare L. Population’s Ecology and Dynamics

The study of population ecology under this research is covered how populations of the plants change over the time and space and interact with their environment. The significant characteristics that might reveal this specific interaction and population behavior have been posed. Such as, population size (the number of individuals in the population), structure (proportion of seedling, sampling and mature plants) density (how many individuals are in a particular area), grow pattern and dynamics etc.

Researches of population’s size, density and abundance and structure give us a comprehensive understanding of populations’ current condition and vulnerability, draw analysis of about environmental and any other constraints that can bring populations future changes and influence its evolution (Woodward, F. I., and D. J. Beerling. 1997).

Investigations at smaller spatial scales on the habitat types and ecological characteristics of wild M. officinalis L. ; O. vulgare L. (Lamiaceae) populations’ e.g. size, structure, abundance, density, grow pattern etc., especially comparisons among populations occurring inside of current occupied niche should give conclusive results on the roles of competitive interactions and evaluation of the basic concepts that are crucial for species’ distribution modeling (Guisan, A., and W. Thuiller. 2005).

The most fundamental demographic parameter of the plant population ecology is the number of individuals within population (Kunin, W.E. 1997b). Population size is defined as the number of individuals present in a subjectively designated geographic range.

Despite the simplicity in its concept, locating all individuals during a census (a full count of every individual) is nearly impossible, so ecologists usually estimate population size by counting individuals within a small sample area and extrapolating that sample to the larger population. Regardless of the challenges in measuring population size, it is an important characteristic of a population with significant implications for the dynamics of the population as a whole (Lebreton et al. 1992).

The eco-geographic data gathered from the field survey were organized into an eco- geographic conspectus, defined by Maxted et al. (1995) as a formal summary of the available geographic and ecological information of the habitats. The eco-geographic data gathered from the

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. field survey using methodology cited by Maxted et al. (3) for each of the 20 populations included in the study is summarized in Table 5.

Table 5 Habitat Characteristics of M. officinalis L. and O. vulgare L. Populations

Locati on Landscape Soil type Soil Humus Annual Species a m(abo type pH concentr precipi Regions NS ve sea ation,% tation mm level ) Getahovit 905 Forest Mountain forest 5.0-6.0 5.0-7.0 500-550 Melissa officinalis redbroun

Tavush Ayrum▲ 495 Dry Mountain forest 6.0-7.0 4.0-4.3 450 Melissa officinalis mountainous brown steppes Ijevan 730 Forest Mountain forest 5.0-6.0 9.0-11 500-800 Origanum vulgare brown

Novoseltso 1490 Forest Meadow black 5.5-6.0 7.0-9.0 650-700 Origanum vulgare Lori vo▲ earth Orgov 1610 Mountain Meadow black 4.5-5.3 3.0-6.0 470-530 Melissa officinalis Aragatsotn steep earth Aparan 1900 Mountain mountain meadow 4.0-5.5 2.5-4.0 550-600 Origanum vulgare steppe Gegharkunik Lichk 1940 Mountain Soils on the 5.3-6.5 2.5-4.0 450-500 Origanum vulgare steppe sediments of Sevan Lake Kotayk Garni 1310 Mountain Mountain black 6.9-8.1 2.0-4.0 430 Melissa officinalis steep earth Jermuk▲ 2104 Forest Forest gray soil 4.5-5.5 13-17 900-1000 Melissa officinalis Jermuk 2098 Forest black earth 7.5-8.3 13-15 900-1000 Origanum vulgare Vayots Dzor Eghegis ▲ 1617 Mountain Mountain chestnut 6.5-8.5 5.5-6.3 600-800 Origanum vulgare steppe Kapan 1486 Forest Mountain forest 4.2-5.1 5.0-7.0 600-750 Melissa officinalis brown soils Artsvanik▲ 1530 Forest Mountain forest 4.3-5.5 7.0-9.0 650-750 Melissa officinalis brown soils Shikahogh 957 Dry steppe Mountain 5.1-6.2 5.0-7.0 400-450 Melissa officinalis meadow steppe soils Srashen▲ 1025 Meadow Mountain forest 4.2-5.0 8.0-9.0 750-900 Melissa officinalis steppes brown soils Tsav▲ 1130 Meadow Mountain forest 4.5-5.5 8.0-9.0 800-950 Melissa officinalis steppes brown soils Karchevan 857 Dry steppe Semi desert 4.7-5.9 6.0-8.0 600-700 Melissa officinalis Syunik k automorph solonetzs Artsvanik▲ 1666 Forest Mountain-forest 7.5-8.0 9.0-10 600-750 Origanum vulgare brown Kapan 972 Mountain Mountain-forest 6.0-8.0 10-11 760-900 Origanum vulgare Mghri steppe brown NewHighw ay▲ Chakaten 1050 Forest Forest brown 6.4-7.5 4.0-10 560-700 Origanum vulgare a- Nearest Settlement

Populations of M. officinalis L. and O. vulgare L. were observed by growing at altitudes ranging from respectively 500 to 2100 m and 730 to 2098 m above sea level, from moist temperate forest regions (i.e., near Jermuk, with 900-1000 mm annual precipitation) until dry mountain steppe habitats (i.e., M. officinalis L.: Shikahogh (south region), Garni (central region) with only 400-450

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. mm annual precipitation; and i.e., O. vulgare L.: Lichk(central region), with only 480-500 mm annual precipitation).

Realized different multiple field observations based on quadrate sampling plot method Maxted et al 1995, over the populations have exposed the growing characteristics of the plants in different populations, as well as assessed population size, density, structure and its abundance over three vegetative period. Based on Maxted methodology three consecutive years period is sufficient for the relevant assessment and analysis of the intricate interaction between the environment, habitat factors and population grow intensity.

According to the collected data, the average plant height measured within some populations displayed a slight increase over the study period, e.g., ca. 9 cm in the Srashen and Karchevank, ca. 8 cm in the Artsvanik and Jermuk populations of M. officinalis L. located in the south regions and e.g., ca. 5 cm in the southern (Syunik, Vayots Dzor). Populations of M. officinalis L. in these habitats demonstrate a positive increase, e.g. ca 1-2 plant/m2 within the average density and stem quantities during the study (Table 6).

Table 6. M. officinalis L. Populations’ Density and Grow Intensity in Different Habitats

Population Density, plant/m2 Plant height, cm Stems quantity/plant

Region NSa 2007 2008 2009 2007 2008 2009 2007 2008 2009

Getahovit 1.30 0.95 0.89 76 70 65 1-3 1-3 1-2 Tavush Ayrum▲ 1.10 0.97 0.93 67 62 58 1-1 1-2 1-3 Aragatsotn Orgov 0.92 0.95 0.99 64 67 69 1-3 2-3 2-3 Kotayk Garni 0.87 0.83 0.79 63 56 48 1-2 1-1 1-1 Vajoc Dzor Jermuk▲ 2.77 3.05 3.57 120 124 127 3-6 3-7 4-7 Kapan 2.2 0 1.97 1.89 107 101 97 2-3 2-5 2-4 Artsvanik▲ 2.50 3.50 4.20 117 122 125 3-5 4-6 5-7

Shikahogh 1.80 2.10 1.80 85 81 77 2-3 3-4 2-3 Sjunik Srashen▲ 2.70 3.10 3.70 118 124 127 4-5 4-6 5-7 Tsav▲ 1.70 1.90 2.50 91 94 92 2-3 2-4 2- 3 Karchevank 1.5 0 2.10 2.70 108 114 117 2-3 3-4 3-5

a - Nearest Settlement;▲ –indecates new population. The plant height and stems quantity were measured at the end of stem forming -phenological phase in respect with assessment of 2% size of each population.

In particular, Srashen, Jermuk and Artsvanik populations of M. officinalis L. are exposed a high growth capacity (density, size) along with plant high growth intensity (plant height:121-125 cm, stem numbers:4-7/per plant). We hypothesize that these favorable growth characteristics may be attributed, at least in part, to the soil’s higher humus concentration and acidic environment as well

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. as relatively higher annual precipitation measured in these habitats as compared to other regions in central and northern Armenia (Table 5) (Pulliam, H. R.. 2000).

A negative trend for the average plant height was observed e.g., ca. 10 cm in the Kapan and ca. 8 cm in the Shikahogh populations of M. officinalis L. from Syunik Region.. A slight decrease ca. 0.31 plant/m2 was observed in the Kapan population’s density during the study period, where in Shikahogh population this criterion fluctuates up and down maintaining average 1.8 plant/m2 over the study period.

These populations have exposed the smallest average plants height among the other populations from south regions of Armenia (Table 6); and where the habitats are characterized with dry steppe landscape, relatively lower annual precipitation and contain poor soil more with alkaline pH environment to compare with the other populations from south regions (Table 6).

No change in average stem number and plant height was observed over the study period for plants examined in Tsav population, in Syunik region from the south part of Armenia. On the other hand, a slight increase ca. 0.8 plant/m2over the population’s density was observed during the study period.

Table 7. O. vulgare L. Populations’ Density and Growth Patterns in Different Habitat

Populations Density, plant/1m2 Plant height, cm, Stems quantity/plant

Region NSa 2007 2008 2009 2007 2008 2009 2007 2008 2009

Tavush Ijevan 0.75 1.04 0.68 43 37 45 1-3 2-4 2-4 Lori Novoseltsovo▲ 0.85 0.73 0.56 40 36 31 1-3 1-2 1-2 Chakaten 0.75 0.83 0.50 50 53 55 2-4 3-4 2-4 Kapan Meghri 0.44 0.47 0.54 54 61 65 2-3 3-5 3-6 Syunik New Highway▲ Artsvanik▲ 2.29 2.45 2.53 55 58 60 1-3 2-3 3-5 Jermuk 2.20 2.90 3.10 71 73 75 3-6 3-7 3-6 Vayots Dzor Eghegis ▲ 0.69 0.73 0.69 58 61 63 1-2 2-3 2-3 Aragatsotn Aparan 1.00 0.77 0.61 33 29 35 2-3 2-3 1-3 Gegharkunkik Lichk 1.10 0.98 0.69 27 25 30 3-4 2-3 1-3

▲-indicates new population. The plant height and stems quantity were measured at the end of stem forming -phenological phase in respect with assessment of 2% size of each population.

Populations of O. vulgare L. from the southern and south-eastern regions (Syunik and Vayots Dzor) are also demonstrated strong growth characteristics, as assessed by their overall larger population sizes, higher plant densities and average heights within the populations (Table 7).

However, a negative trend for density and the average number of stems per plant was observed in populations of O. vulgare L. located in the central regions (Aparan, Lichk). Different biotic and abiotic factors could affect the growth characteristics of the plants in different populations, but we can initially hypothesize that the poor soil quality and low annual precipitation rates measured at the Aparan and Lichk populations of O. vulgare L. contributed to the negative trend observed (Pulliam, H. R.. 2000, Vandermeer, J. H.; Goldberg, D. E. (2003).

A negative trend for the average plant height was observed e.g., ca. 15 cm in the Garni from the central region; ca. 11cm and 9 cm appropriately in the Getahovit and Ayrum populations of M. officinalis L. from the northern regions. Also, the average number of stems of per plant ca. 1-2 and the population’s density was slightly decreasing ca.0.40 plant/m2 in these populations over the study period (Table 6). Populations located near the Getahovit and Ayrum in the northern Armenia, occupied forest landscape habitats.

However unlike populations from the south regions, the low grow capacity of these populations and reduction in sizes and density could be connected with less humus concentration and alkaline pH environment in the soils (Table 5). Population located near Garni from the central region in Armenia, occupied mountain steep landscape with mountain black-earth soils, with low humus concentration, with highest alkaline soil pH among the other populations have the smallest population density, plant height and stem quantity and exhibits negative trend in populations overall size and density during the study (Table 6).

So, Through yearly measurements of population size and density, of M. officinalis L. and O. vulgare L. populations were identified that either expanded or diminished in size over the course of the five-year study (Fig. 13&14). Studying population growth might reveal the causes that bring changes in population sizes and growth rates.

The dynamics of sizes in populations is exposed mostly with exponential growth pattern. That is the continuous increase or decrease in a population in which the rate of change is proportional to the number of individuals at any given time (Vandermeer, J. H.; 2010).

Populations demonstrate increase in sizes and have mostly south location (Syunik and Vayots Dzor Regions). The highest continues increase of M. officinalis L. populations has been recorded in Artsvanik, Srashen and Jermuk habitats that gained more than 120 individuals over the study period. This number was 130 individuals in Artsvanik and 350 individuals in Jermuk habitats of O. vulgare L. populations over the study period (Fig. 13&14). According to logistic growth and rate, 69

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. we suggest that these populations could be characterized with an intrinsic rate of increase and are capable of expanding to achieve their maximum sizes (until the environmental factors constrain this type of growth Fig. 12).

Figure 12: Logistic growths explicit the exponential grow of population until it approaches its carrying capacity (K).

Factors that enhance or limit population growth can be divided into two categories based on how each factor is affected by the number of individuals occupying a given area — or the population's density (Mustrajarvi, K., Siikamaki, P., Rytkonen, S., and Lammi, A. 2001). As population size approaches the carrying capacity of the environment, the intensity of density- dependent factors increases (Fischer, M., and Matthies, D. 1998). For example, competition for resources can eventually limit population size. Other factors, seasonal weather extremes, e.g. droughts, frosts, etc affect populations irrespective of their density, and can limit population growth simply by severely reducing the number of individuals in the population (Vandermeer, J. H. 2010). In respect with logistic grow mechanism, comparatively high density (Table 6&7) of the plants populations does not afflict negatively on the exponential grow pattern, which might mean that they have not reached yet to their maximum grow capacity . So, we might contemplate that the growth would continue in the future if we exclude other factors e.g. seasonal weather extremes, natural disasters etc (Vandermeer, J. H. 2010).

Slight increase in populations sizes were recorded in other habitats located from south regions of Armenia. E.g. O. vulgare L.: in the Kapan Meghri New Highway and Chakaten populations with only ca. 60 more individuals; M. officinalis L. : in Tsav population, only with ca. 23 more individuals over the study period. In compare with these populations, relatively higher increase was observed in Karchevank population of M. officinalis L. with ca. 89 individuals. Only, in Kapan population of M. officinalis L. has exposed a slight decrease with ca. 13 individuals over the study period from south regions of Armenia. Among the other populations, the Eghegis population of O. vulgare L. with the average of 230 individuals and Shikahogh population of M. officinalis L. with ca. 90 individuals fluctuated during the study period and maintained comparatively constant sizes (Fig. 13 (C), Fig. 14(C)).

Figure 13: Populations Dynamics of Melissa officinalis L. across different habitats (A) in the North; (B) in the Central; (C) in the South Regions of Armenia, during the study period (in accordance with quadrat sampling plot method).

Figure 14. Populations Dynamics of Origanum vulgare L. across different habitats (A) in the North; (B) in the Central; (C) in the South Regions of Armenia, during the study period (in accordance with quadrat sampling plot method).

In fact, from seven populations of Melissa officinalis L. located in the south part of the country, the increase is recorded in five populations and it is quite high in three of them, one population has minor decrease and the other is constant with its size over the study years. And, from five populations of O. vulgare L. located in the south part of the country, the increase is recorded in four populations and it is quite high in two of them and the one is constant with its size over the study years. In addition, M. officinalis L. and O. vulgare L. biggest populations documented during the study are located in Syunik and Vayots Dzor Region from south part of Armenia. So, the species populations are apparently thriving in the south part of the country. We hypothesize that 72

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. these favorable growth characteristics may be attributed, at least in part, to the higher concentration of humus and annual precipitation measured in this region as compared to other regions in central Armenia (Aragatsotn, Gegharkunik) (Oostermeijer, J. G. B., Luijten, S. H., Krenova, Z. V., and Den Nijs, H.C.M. 1998 ). On the other hand, populations with displayed a reduction in their sizes over the study period were observed mostly northern and in the central regions (Fig. 13 (A,B); Fig. 14 (A,B)). The amount of decrease of O. vulgare L. included of 178 individuals in the Ijevan and, 71 in the Novoseltsovo from the northern regions, as well as 83 in the Aparan and, 77 in the Lichk populations from the central regions during the study period. And, of M. officinalis L. ca. 76 in Getahovit and ca. 67 in Ayrum populations over the study period. So, the most drastic change was observed in Getahovit population of M. officinalis L. and in Ijevan population of O. vulgare L.. In fact, trend of reduction of M. officinalis L. population was especially higher during 2010-2011 vegetative periods, when the precipitation was recorded less to compare with next vegetative years. It is almost two times higher in compare with the previous years. On the other hand, it was observed higher from 2008-2009 in Ijevan population of O. vulgare L. plant, when the temperature was recorded relatively higher. So, we may contemplate that the temperature and the annual precipitation are more crucial among the other environmental factors on O. vulgare L. and M. officinalis L. populations’ dynamics. A negative trend for the M. officinalis L. population size was observed in the Garni population from Kotayk Region located in the central Armenia, over the study period (Fig.6. B). The decrease was gradually by comprising ca. 100 individuals overall the study. In fact, Powedery mildew diseases were observed in this population (5, June 2007). The leaves observation at colba plots with room temperature during 3-4 days has been conducted to identify the disease at Gene Pool Laboratory of Plants Cultivation and Vegetable Growing Department of Armenian National Agrarian University. It is important to mention that Powedery mildew is not recorded in other populations across the country. The disease spread over 40% of the Garni population in 2009 and is remarkable, especially during the plant flowering phenological phase (June-July). In the next study years, the disease was spread by comprising 67% in 2011.

Future studies could be conducted to reveal the causes of the origin of disease in this population that might have either the population’ genetic, or other environmental reasons. Furthermore, one thing is certain that the dramatic decrease in sizes of this population obviously connected with this phenomenon as well. However, M. officinalis L.’ Orgov population (Aragatsotn Region) from the

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. central Armenia is documented as increasing gradually with ca. 20 individuals. The overall growth in the population is ca. 80 individuals over the study period.

In fact, population may display distinctive behaviors based on their size. Shrinking and small populations face a greater risk of extinction and are, thus, more susceptible to climate change impact (Whittaker, R.J., M.B. Araújo, P. Jepson et al. 2005). Additionally, individuals in small population are more susceptible to random deaths e.g. fire, floods, and disease have a greater chance of suppressing the growth of population, such as depleting grow pattern in the Garni population caused by disease.

In addition, populations with declining sizes had comparatively lower density (Table 6 & 7). In this respect, we conclude that the density does not limit population growth (Vandermeer, J. H. 2010; Vandermeer, J. H., Goldberg, D. E. 2003; Vandermeer, J. H. 1972). Habitats that are located in the central and northern regions, however, have a relatively low carrying capacity and, in other words, less favorable conditions for the plant to grow (Table 5). Thus, even a low density could be decisive when combined with unfavorable environmental and extreme weather factors. In addition, we should not exclude negative impacts from humans, which are more intense in the central and northern regions of Armenia (National Report on the State of Plant Genetic Resources in Armenia, 2011). Populations located in the central and northern regions of Armenia may be comparatively more vulnerable to the impact of global climate change (Thomas C.D., Cameron A., Green R.E., Bakkenes M., et al. 2004, Hannah, L., G. F. Midgley, T. Lovejoy, et al. 2002, Woodward, F.I. 1987). Too, it is not only the environmental or habitat and the plant biological factors that influenced on the plant population size and distributional abundance, it is also antropogenetic and other biotic factors that are different across the regions and habitats and might affect on populations growing condition.

Some anthropogenic threats that were identified as part of the study included poor land management (erosion, overgrazing), increasing population pressure (impact of livestock overpopulation, improper human development), and excessive or inappropriate collection for the purposes of local sale/usage (due to lack of knowledge/training of collectors). Especially in the northern and central regions, it was observed that the wild harvesting of medicinal plant materials (e.g., O. vulgare L., Hypericum perforatum L., and Melissa officinalis L.) at inappropriate times for

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. sale at Yerevan markets negatively impacted population sustainability. Records indicate that approximately 200 tons of wild edible and/or medicinal plants are sold in Yerevan markets each year (National Report on the State of Plant Genetic Resources in Armenia, 2008-2012).Wild M. officinalis is used by local inhabitants primarily in the form of an infusion (tea) as to treat disorders of the nervous and digestive systems (Gabrielian E, Zohary D, 2004). In addition, a significant concept is the grasp comprehension that none of this factor might separately causes the population size to alter. Only, the all components can rightly show their intricate and complex interactions among them and the correct picture of their functioning impact on the changes of the plant populations’ sizes.

4.3. Interdependent Effects of Habitat Factors and Climate on Population Dynamics of M. officinalis L. and O. vulgare L. Species in Armenia

Linking environmental variation over time and space to individual performance and population dynamics is a central objective in ecology. Spatial variation in environmental factors may occur over many different scales and has been shown to influence population dynamics in many systems (Horvitz, Tuljapurkar & Pascarella 2005, Alvarez-Buylla 1994). Temporal variation in environmental factors may be important to population dynamics both in terms of long-term trends and short-term fluctuations. Short-term fluctuations in environmental factors shown to have substantial effects on plant population dynamics include flooding ( Schleuning, Huaman & Matthies 2008, Smith, Caswell & Mettler-Cherry 2005) and other catastrophic events ( Pascarella & Horvitz 1998, Menges & Kimmich 1996, Åberg 1992). Also, temporal variation in biotic interactions has been shown to have important effects on plant population dynamics ( Kolb, Ehrlén & Eriksson 2007, Maron & Crone 2006, Horvitz, Tuljapurkar & Pascarella 2005).

Linking growth rates and population viability to key environmental factors is also important because long-term management of species will inevitably be associated with the management of environmental factors (Gotelli & Ellison 2006; Sutherland 2006, Fieberg & Ellner 2001).

In a spatial context, demographic models that include environmental factors can be used to determine how habitat quality influences population viability (Dahlgren & Ehrlén 2009). Temporal variation is often included in population models in terms of stochastic variation among years. Linking among-year variation in population performance to environmental variation is likely to 75

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. improve predictions of demographic models in many cases. For example, it enables exploration of the effects of climate change on population viability in a more direct way (Jongejans et al. 2010, Jenouvrier et al. 2009, Maschinski et al. 2006). Lastly, effects of environmental changes such as global warming are likely to depend on local environmental conditions. Examinations of how interactions between temporal variation in climatic variables and local environmental conditions affect population dynamics are therefore of major importance to enhance predictions of future local population viability and growth rates.

In this study, we investigated environmental effects on population growth rate and dynamics of M. officinalis L. and O. vulgare L. species. We linked variation in population dynamics in eleven populations of M. officinalis L. and nine populations of O. vulgare L. over five annual transitions to a set of local environmental factors as well as to variation in temperature and precipitation among years. We asked three questions: (1) Can population growth rate be linked to differences in habitat quality? (2) Can differences in local population growth rates among years be linked to climatic variation? And (3) do effects of climate depend on local habitat quality?

To investigate the effects of environmental factors in different habitats of populations, we formulated among year variations in regard with populations dynamics within environmental and habitat factors separately. In addition, to explicit the correlations between the impact of environmental factors and habitat factors on populations’ dynamics stochastic multiple and principal component analysis among the populations dynamics of per year were formulated (Tibshirani 1996).

So, we examined the effects of two habitat factors and the effects of temperature and precipitation on population dynamics. We correlated temporal variation in population dynamics to differences on among year temperature and precipitation. In addition, we included the effect on population’s dynamic of the average summer temperature and long term precipitation variations in regard with habitats. We used monthly temperature and rainfall data 1990–2011 from the meteorological stations closest to the respective populations.

It is essential to mention, that the altitudinal variation within the country results in further variation in climatic zones, in addition to existing latitudinal clines. Thus the weather, regardless the season, very often varies even in the nearby regions. Therefore, in addition to the general among year annual temperature and precipitation variations, we have discussed temperature and precipitation variations between the habitats and the effect of them on the plants population

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. dynamics. Also, the mountainous nature of Armenia results in a series of highly diverse landscapes, with variations in geological substrate, terrain, climate, soils, and water resources.

Melissa officinalis L. population dynamics, habitat features and environmental variations:

As we discussed above, the number of habitats and environmental factors could affect on the growth dynamics of M. officinalis and O. vulgare populations across the country during the study period. Among the habitat factors we examined the soil features, in particular its pH and humus concentration.

Figure 15. Melissa officinalis L. population dynamics and habitat features 1-Getahovit, 2- Ayrum, 3- Orgov, 4- Garni, 5- Kapan, 6- Artsvanik, 7- Shikahogh, 8- Srashen, 9- Tsav, 10- Karchevank, 11- Jermuk.

So, M. officinalis L. plants tolerated soils with pH ranging between 4.0 (acidic) and 8.1 (alkaline), with a preferred range between 4.0 -5.5. The species can tolerate soils with various concentrations of humus and can even grow in nutritionally poor soil, as observed for the populations located near Ayrum and Garn (Fig. 15). However, the plant populations with the growing dynamics were observed at the habitats where the soil pH is acidic e.g. Orgov from the central regions and Artsvanik, Srashen, Jermuk etc from the southern regions. In addition, in these habitats humus concentration is relatively higher, especially in the Artsvanik, Srashen, Jermuk populations within 9-13 percentages. We can conclude 77

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. that the highest increase rate and the biggest sizes of these populations can partially be connected with these favorable habitat conditions (Gabrielyan N. 2004).

Figure 16. Melissa officinalis L. population dynamics and temperature variation 1-Getahovit, 2- Ayrum, 3- Orgov, 4- Garni, 5- Kapan, 6- Artsvanik, 7- Shikahogh, 8- Srashen, 9- Tsav, 10- Karchevank, 11- Jermuk.

Figure 17. Melissa officinalis L. population dynamics and summer temperature variation 1-Getahovit, 2- Ayrum, 3- Orgov, 4- Garni, 5- Kapan, 6- Artsvanik, 7- Shikahogh, 8- Srashen, 9- Tsav, 10- Karchevank, 11- Jermuk.

In general, among year average temperature variations is crucial for the populations located in the northern regions e.g. Getahovit, Ayrum. In particular, the low annual average temperature was drastically impacted on the population’s growth dynamic located in the habitats from the northern regions across the country (Fig. 16).

During 2007-2009, when the annual temperature was recorded 12-130C the number of decreasing individuals were documented less with ca. 10 individuals in the Getahovit and 5 times less in the Ayrum populations to compare with 2009-2011 vegetative years, when the temperature was recorded 6-80C (Fig. 16). The negative impact of annual low temperature might be connected with the fact that it can contribute the early spring frosts that are common phenomenon in these regions. Researches were identified seedlings suffered much of early spring frosts and thus affect negatively on the population size (Taghtajyan A. L. 1987). This reinforces the hypothesis that environmental or weather variations impact on population dynamics are interdependent and correlated (Davison et al. 2010).

The negative impacts of among year variations in temperature and the annual low temperature on the populations growth dynamics in these habitats is connected with the number of resoans. One thing is certain; the relatively higher long term summer temperature that was observed in these habitats is decisive factors in this respect. Based on our knowledge of the species, we expected that in the summer the high temperatures and low precipitation (Fig. 19) influence negatively on population growth dynamics, as both these factors may lower soil water content (Penuelas et al. 2004).

Thus, wild M. officinalis L. populations were declining in the habitats where the average summer temperature is comparatively higher e.g. equal or above 200C degree in the Getahovit, Ayrum populations with around ca. 70 individuals from the northern regions; and in the Garni ca. 100 individuals from the central region (Fig. 17). Another factor, which could reinforce the negative impact, could be the early spring frost and abrupt weather changes which caused seedlings and samplings to suffer more which is the common phenomenon in the northern regions of Armenia (Gabrielian E 2004, Taghtajyan A. L. 1987).

On the other hand, the dynamics of populations located in the Orgov and Garni populations from the central regions of Armenia, does not affected much of the variations of yearly annual temperature over the study years (Fig. 16). For instances, irrespective of among year temperature

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. variations the increasing number of the individuals in the Orgov population were documented in average ca. 20 individuals of each study year.

Also, the decreasing numbers of individuals in the Garni population were documented ca. 24 of per study year. In addition, we cannot formulate Garni population much into the general regularities of among year temperature variations due to Powder millodw disease that was observed at the Garni population.

While observing the impacts of among year temperature variations and the annual average low temperature on populations’ dynamics located in the south regions of Armenia, we can conclude that it is not so crucial. In addition, populations’ with increasing dynamics were observed growing in the habitats, where the average summer temperature is comparatively lower. In particular, we can speculate that the most favorable summer temperature is recorded between 15- 180C at the habitats where populations observed to demonstrate the highest rate of increase with around ca. 120 individuals, during the study years, e.g. Jermuk, Artsvanik, Srashen (Fig.17).

For instances, Srashen population with relatively bigger size, during the study was documented with increasing dynamics, where irrespective of among year temperature variations the number of individuals documented similar within the study years in an average ca. 30 individuals. (Although the clear examination would reveal the fact that in 2010, when the annual average temperature was recorded 80C the number of increasing individuals were documented two times less than in 2007-2009, when the temperature was recorded higher with 4-5 degree, however in 2011, when the temperature is observed the lowest the increasing number of individuals were documented quite high e.g. ca 36 equal with the growth number that were documented in 2007- 2009. Thus we can conclude that the variations of temperature among years did not impact crucially).

In addition, we were documented in the contrary of the northern regions that the low temperature in certain habitats is influencing positively on the populations growth, e.g. Artsvanik, Jermuk, populations during 2009-2011 vegetative years when the annual temperature was recorded 6-80C, were observed to growth ca. 10 individuals more than the previous years, when the temperature was recorded 12-130C. The other habitats of populations such as Shikahogh, Karchevank, Tsav, Kapan, where the average summer temperature is recorded above 200C were demonstrated slightly increase or decrease mostly sustaining the constant size during the study years irrespective of temperature variations (Fig.17).

Figure 18. Melissa officinalis L. population dynamics and precipitation variation 1-Getahovit, 2- Ayrum, 3- Orgov, 4- Garni, 5- Kapan, 6- Artsvanik, 7- Shikahogh, 8- Srashen, 9- Tsav, 10- Karchevank, 11- Jermuk.

Figure 19. Melissa officinalis L. population dynamics and long term precipitation in Habitats 1-Getahovit, 2- Ayrum, 3- Orgov, 4- Garni, 5- Kapan, 6- Artsvanik, 7- Shikahogh, 8- Srashen, 9- Tsav, 10- Karchevank, 11- Jermuk.

Unlike the temperature variations and among habitats long term precipitation variations, the among year precipitation variations were not observed to have such a negative impacts in the habitats from the northern regions (Tavush, Lori) of Armenia. In fact, during the study wild M. officinalis L. populations were observed with declining dynamics in the habitats where long term annual precipitation was recorded relatively lower e.g. Getahovit, Ayrum from the northern regions, where the populations are declining in dynamics during the study period (Fig. 18 & 19). However, the decreased rates were documented less in 2008-2009 period e.g. ca. with 7 individuals less at Getahovit, and 5 times less at the Ayrum populations, when the annual precipitation were recorded relatively lower . This phenomenon might be connected with number of abiotic and biotic factors as well as antropogenetic impacts that is required multiple examinations in future.

Similar with the among year temperature variation the precipitation variations also were not observed to have crucial impacts on the populations dynamics located in the central (Aragatsotn, Kotayq) and southern (Vayots Dzor, Syunik) regions of Armenia (Fig. 18).

For instances, irrespective of among year temperature variations the increasing number of the individuals in the Orgov population from Aragatsotn region were documented in average ca. 20 individuals of each study year. At the same time, at the Garni population located in Kotayq region, the highest number of decreasing individuals with ca. 31 were documented in 2008 study year when the annual average precipitation was recorded the lowest within the whole period (less then 500mm).

In addition, Garni population was documented drastic decline with the highest number of ca. 100 individuals; among the other habitats it is characterized with the lowest long term precipitation 400mm (Fig. 19). Similar long term precipitation pertain is observed at the Shikahogh habitat from the south location (Syunik Region), which is the smallest one with its ca. 90 individuals and was not demonstrate the increase and sustained the constant size overall the study period (Fig. 19).

Populations with biggest sizes and growing dynamics were observed mainly at the habitats, where the long term precipitation is above 600mm e.g. Srashen, Karchevank, Artsvanik (Syunik Region), Jermuk (Vayots Dzor Region), the increase rate of these populations were documented ca. 120 individuals in the whole study period (Fig. 19). In addition, the fluctuation of annual

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. precipitation among the years of 2007-2011 does not influencing importantly on the annual growth rate on these populations located in the south regions of Armenia.

In fact, in 2008 the yearly precipitation is recorded the lowest with 490mm, however the growth rate was observed more less equal to compare with the other years of study e. g. Artsvanik, Srashen, Karchevanq. Simultaneously, the highest number of increased individuals were documented in the Artsvanik, Jermuk populations from south regions, during the years when the annual precipitation was recorded relatively higher (Fig. 18).

So, we can conclude that the long term annual precipitation has more influence in this respect and the plant populations are mostly growing well and thriving in the habitats where it is observed to be above 600mm.

Origanum vulgare L. population dynamics, habitat features and environmental variations:

Figure 20. Origanum vulgare L. population dynamics and habitat features 1-Ijevan, 2- Novoseltsovo, 3- Aparan, 4- Lichk, 5- Chakaten, 6- Kapan Meghri New Highway, 7- Artsvanik ,8- Jermuk, 9- Eghegis.

O. vulgare plants tolerated soils with a pH ranging between 6.0 (mildly acidic) and 8.5 (alkaline), with a preferred range between 7.0 -8.0. In fact, this criterion indicates that Artsvanik and Jermuk populations from the south regions of the country (Vayots Dzor, Syunik), which were

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. documented with their remarkable biggest sizes and increasing higher dynamics during the study, are the most preferable for wild O. vulgare L. plants. In addition, the humus concentrations were observed with the highest percentage among the other habitats (Fig. 20). However, the plant can tolerate soils with various concentrations of humus and can even grow in nutritionally poor soil, as observed in the Aparan, Lichk from central regions (Aragatsotn, Gegharkunik). Plants in the latter populations, however, exhibited low growth characteristics (e.g., density and plant height) as compared to those observed in other populations, and the overall size and structure of these populations decreased over the period of the study (see in details in the 4.2. section). Fig. 21 & 22 demonstrate respectively the among year annual temperature variation and the long term average summer temperature variations among the habitats as well as O. vulgare L. species’ population growth dynamics in this respect.

Figure 21. Origanum vulgare L. population dynamics and among year temperature variation 1-Ijevan, 2- Novoseltsovo, 3- Aparan, 4- Lichk, 5- Chakaten, 6- Kapan Meghri New Highway, 7- Artsvanik ,8- Jermuk, 9- Eghegis.

Figure 22. Origanum vulgare L. population dynamics and summer temperature variation 1-Ijevan, 2- Novoseltsovo, 3- Aparan, 4- Lichk, 5- Chakaten, 6- Kapan M. New Highway, 7- Artsvanik ,8- Jermuk, 9- Eghegis.

The drastic declined was documented in the Ijevan and Novoseltsovo populations respectively with ca. 180 and 100 individuals overall the study period. These habitats are characterized with less humus concentration and alkaline environment and relatively higher summer temperature (Fig. 20- 22). In particular, the highest number of decreased were documented during 2008-2009 vegetative period, when the annual temperature relatively was recorded higher. However, in 2010-2011 periods when the annual average temperature was decreasing with 3-60C (Fig. 21); the declining dynamic rates in these populations were falling 4-6 times respectively in the Ijevan and Novoseltsovo habitats.

So, we can conclude that relatively higher summer temperature at the habitats in combined with higher annual temperature in the northern regions, is affecting drastically on populations declining overall rate (Toräng, Ehrlén & Ågren 2010). Similar regularities were observed also in the central regions of Armenia. For instances, in the Aparan population the decreased rate was observed to be higher with ca. 6 individuals, and the population size is less with around ca. 10 individuals to compare with Lichk population, where the average long term summer temperature is higher with 10C to compare. As well as, in both populations the decreased individuals were observed 4-5 times

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. higher during 2007--2009 vegetative periods, rather than 2009-2011 (Fig.21), when the annual temperature is decreasing with 3-60C.

The pattern is similar with the certain habitats located in the south regions (Syunik, Vayots Dzor), which are also characterized with higher summer temperature. In fact, at the Chakaten, Kapan Meghri New Highway, Eghegis habitats, where long term summer temperature is above 200C the populations have smaller sizes from around 150 to 200 individuals and were growing slowly and overall the study period the growth rate was observed with around ca. 60 individuals at the Chakaten, Kapan Meghri New Highway,(Syunik Region) and ca. 6 individuals at the Eghegis (Vayots Dzor Region). However, relatively higher annual temperature was not observed to affect negatively the population growth rate, especially higher during the years when the annual average temperature is recorded higher with 3-60C (Fig. 21). The impact of relatively higher annual temperature was not documented to be a crucial on the population’s growth dynamics in the habitats with more favorable habitat conditions (e.g. humus concentration, pH, landscape type etc) in contradiction with the populations located from the northern and central regions of the country.

Also, the biggest populations were documented in the Artsvanik (Syunik Region) and Jermuk (Vayots Dzor Region) habitats, where the average long term summer temperature is below 200C. In fact, the growth rate is quite higher among the other populations overall the study period. And, the annual increase rate was documented quite higher when the annual average temperature is above 100C (Fig. annual tem), as it was observed also among the other populations with south locations.

In conclusion, it is significant to mention that populations growth rate diversities and regularities among the different regions of Armenia is reinforcing the hypothesis that weather, regardless the season, very often varies even in the nearby regions due to mountainous landscape diversities, where the climate is mostly continental.

The among year precipitation variations as well as the long term average precipitation appertain at the habitats were observed for identifying the influence on the populations’ dynamics of Origanum vulgare L. over the study years in the country.

Fig. 23 illustrates O. vulager L. populations are declining mostly located in the northern (Ijevan, Novoseltsovo) and central (Aparan, Lichk) regions of Armenia. The remarkable number of individual ca. around 70 and 30 were declined respectively in the Ijevan and Novoseltsovo populations during 2007-2009 growing seasons, when the annual precipitation was recorded less to compare with other years. 86

Figure 23. Origanum vulgare L. population dynamics and precipitation variation 1-Ijevan, 2- Novoseltsovo, 3- Aparan, 4- Lichk, 5- Chakaten, 6- Kapan Meghri New Highway, 7- Artsvanik ,8- Jermuk, 9- Eghegis.

Figure 24. Origanum vulgare L. population dynamics and long term precipitation variation in Habitats. 1-Ijevan, 2- Novoseltsovo, 3- Aparan, 4- Lichk, 5- Chakaten, 6- Kapan Meghri New Highway, 7- Artsvanik ,8- Jermuk, 9- Eghegis.

During 2009-2011 growing seasons the decreased number of plants for per season was documented less by comparing with previous years ca. around 20 and 10 correspondently at the Ijevan and Novoseltsovo populations, while the annual precipitations were observed to grow gradually (Fig. 23).

In the same way, the biggest decline of population dynamic were occurred in 2007-2008 and 2008-2009 growing seasons, (when the annual precipitation were documented lower (Fig. 23) by comprising respectively ca. 40 and 30 individuals in the Aparan population (Aragatsotn Region); ca. 20 and 40 individuals in the Lichk population (Gegharkunik Region). The decreased rate was documented in an average 7 individuals of both populations in per growing season of the next years. In addition, the lowest long term annual precipitation was recorded in the Lichk and Aparan (central regions); Novoseltsovo (northern region) habitats, where populations were observed to have relatively smaller sizes and were declining growth dynamics during the study period (Fig. 23). So, we may conclude that lower average annual precipitations were influenced negatively on the population growth in the northern and central regions of Armenia.

O. vulgare L. populations with growing dynamics are mostly located in the southern regions of the country (Syunik, vayots Dzor). In general, among the five populations Artsvanik and Jermuk are notable with their biggest sizes and the growing dynamics of populations during the study period. Interestingly, the highest increase in the number of individuals ca. 70 and 40 individuals at the Artsvanik ; ca. 170 and 80 individuals at the Jermuk; ca. 10 and 25 individuals in the Chakaten populations were documented respectively in 2007-2008 and 2008-2009 growing seasons, when the annual precipitation was recorded less to compare with other years (Fig. 23). During 2009-2010 and 2010-2011 growing seasons, the increasing number of individuals is less than previous years, by comprising respectively ca. 10 and 20 at the Artsvanik and ca. 13 and 7 individuals at the Jermuk populations. We can conclude that the among year precipitation variations, in particular relatively low annual precipitations were not affected negatively on the populations’ growth dynamics in these habitats due to the highest long term precipitation between e.g. 750-1000mm are characterized these habitats (Fig. 24). In addition, most of these habitats are characterized with favorable habitat features, pH environment were documented mostly alkaline and humus concentration was observed remarcabily higher too (Fig. 20).

Principal component analysis (PCA) of relationships between variation in environmental, habitat factors and stochastic population growth dynamics identified that temperature and precipitation variation influences on the populations’ growth dynamics were interdependent on the 88

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. habitat factors (e.g. soil pH, humus concentration). In addition, the negative impact of these factors did depend on the local quality, being more intense in populations with poor humus concentration and unfavorable pH environment e.g. for M. officinalis is less acidic and for O. vulagre less alkaline in the habitats mostly located in the northern and central regions of Armenia.

Several other studies with plants have shown that effects of climatic variation may depend on local environmental conditions (Davison et al. 2010, Schleuning, Huaman & Matthies 2008). Taken together, these results illustrate that interactions with the local abiotic environment should constitute an important part of predictions of population viability under climate change.

Among-year variation in population growth rate was synchronized between populations and correlated with summer temperature and precipitation. The results suggested that warm summer temperatures had negative effects on population growth rate, acting mainly through survival. In particular, the exceptionally warm summers with low precipitation resulted in very low population growth rates of M. officinalis L in the northern regions of Armenia. Several previous studies with perennial plants have documented negative effects of summer drought on population growth rates (e.g. Davison et al. 2010; Toräng, Ehrlén & Ågren 2010, Riba, Pico & Mayol 2002 ). In addition, negative plant responses to high summer temperatures may be explained by increased evapotranspiration and associated declines in soil moisture (Penuelas et al. 2004).

While observing the impacts of among year temperature variations and the low annual temperature on populations’ dynamics among the different regions of Armenia, the most crucial effects were documented in the northern and central regions. In addition, these habitats are featured with comparatively lower long term precipitation which reinforces the negative impacts of climate variations along with unfavorable habitat features, such as poor soil and faint acidic or mostly alkaline pH environment.

On the other hand, among year temperature and precipitation variations, in particular, the low annual temperature and precipitation were not observed to impact drastically in the south regions of Armenia, where the habitats are characterized with comparatively lower summer temperature and higher long term precipitation, in addition the observed habitat factors are favorable for the plant to thrive e.g. reach soils and acidic pH environment.

In general, it was observed that relatively higher summer temperature along with annual temperature variations, in particular with higher annual temperature are affecting negatively on the O. vulgare L. population growth dynamics across the different regions of Armenia. This influence 89

is particularly higher in the habitats that featured with poor soil and less alkaline pH environment. Populations were observed to be small in their sizes and vegetative growth intensity (density, stem quantity, abundance, plant height etc) and manly decreasing (from the northern and central regions) or slightly increasing/mainly maintaining constant sizes (from the south regions). In addition, the impact of relatively higher annual temperature was not observed to be a crucial on the population’s growth dynamics in the habitats with more favorable habitat conditions (e.g. humus concentration, pH, landscape type etc) in contradiction with the populations located from the northern and central regions of the country.

Our results illustrate that among year precipitation variations, especially low annual precipitation were affected negatively on the population growth dynamics from the habitats mostly located in the northern and central regions. The drastic impact of annual low precipitation is related with habitat features such as low quality of soil environment. These habitats are characterized with the lowest amount of long term precipitation.

On the other hand, populations located in the habitats with high humus concentration and alkaline pH environment were not suffered from among year precipitation variations and relatively low annual precipitations. In addition, these habitats were characterized the highest long term precipitation between e.g. 750-1000mm are characterized these habitats. So, O. vulgare L. populations on poor humus content and less alkaline pH environment suffered on more negative effects of among year precipitation variations and low annual precipitation.

Besides, temperature and precipitation are fundamental determinants of the distribution and abundance of wild M. officinalis L. and O. vulgare L. species and are an essential part of any attempt to assess climate change impact of their future distribution shifts. In addition, PCA illustrate the differences between these two variations among the species’ population growth dynamics. So, the precipitation variations were analyzed to impact more on M. officinalis L. and temperature variation on O. vulgare L. population growth dynamics.

In this study, we did not experimentally investigate the mechanism behind the relationship between population growth rate and soil pH and humus concentration. However, it is likely to be at least partly associated with differences in landscape type, soil type and water-retaining capacity. In particular, populations with growing dynamics were observed in the habitats with meadow steppes or forest landscapes mostly with forest brown soils e.g. Artsvanik, Jermuk, Srashen populations from the south regions of Armenia (see. pp 66 table 5). Future studies should therefore preferably

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. investigate causes of populations’ dynamics in effects of large-scale environmental changes such as global warming, to better understand the effects on population and community dynamics.

The fact that, under environmental and habitat conditions similar to those observed during the study, populations were observed to decrease and underscores their vulnerability towards future climate change impacts in the northern and central regions of Armenia (Bonin et al. 2007, Bensettiti et al. 2002).

In conclusion, we correlated temporal variation in population dynamics to differences in temperature and precipitation and certain habitat factors, particularly soil pH and humus concentration. Our results with the wild M. officinalis L. and O. vulgare L. species show that both local environmental factors and climatic variation among years influence population dynamics, and that the effects of climate depend on local habitat quality.

4.4. Wild Melissa officinalis L. and Origanum vulgare L. Population’s Structure

Over the last 30 years, study of population age structure has become one of the important ecological properties along with population size, density, dispersion pattern etc to understand individual’s interactions with the environment. Populations’ age structure is one of the important parameter of population demography that can impact present and future size of population. Ignoring age structure could bring the underestimation of population growth pattern and its future survival rate (Lauenroth, W.K. & Adler, P.B. , 2008).

The structure of plant populations can be evaluated in terms of age, size and form of the individuals that constitute it (Harper & White, 1974). The structure of a plant population is governed both by abiotic and biotic factors that also have a substantial bearing on the spatial pattern, age grouping and genetic structure of plant populations (Hutchings, 1997). Additionally, these groups of factors also regulate the spatial and temporal changes in the number of individuals in the populations (Watkinson, 1997; Silvertown & Charlesworth, 2001).

Therefore, the plant population’s age structure is simultaneously the outcome of past demographics events and an indicator of its demographics future. Demography is the study of the age structure and growth of populations, especially as it relates to the births and deaths of individuals. Present age structure reflects previous temporal variation in recruitment and morality and, because vital rates in many populations are age dependent, future age structure can be projected from present structure (Bacchetta, G. Fenu, R. Gentili, Mattana, E. & Sgorbati S., 2013). Odum (1971) states that the ratio of various age groups within the population determines the current reproductive status of a species population, in addition to its future role in the ecosystem, e. g. whether or not the species would be maintained or phased out.

During the study, we have assessed and determined the age structure and identified changes over three consecutive years, in 11 populations of M. officinalis L. and in 9 population of O. vulgare L.. In accordance with quadrat sampling plot method, at each site, all individual plants within each 1-m2 quadrat were identified and mapped to cover 2% of per plot within each year using a pantograph (Hill 1920).

The basal cover of grasses was mapped as polygons with indeterminate shape. When possible, seedlings, samplings and matures were explicitly labeled during mapping (Lauenroth & Adler 2008; Chu et al. 2013). It has been determined the age ratio in each population and the average mathematical structure of each population based on relevant mathematical formulas. So, the results of this estimation are illustrated in the following figures.

Melissa officinalis L. Populations Age Structure

Origanum vulgare L. Populations Age Structure

Figure 25. M. officinalis and O. vulgare L. population’s age structure in respect with quadtrat sampling plot method. (the av. data of 2007, 2008 and 2009 study years).

Dynamics in wild M. officinalis L. Population Structure (according to 2007, 2008 and 2009 study results)

Figure 26. Populations’ age structural distribution and its modification in the (A) Northern; (B) Central; and (C) Southern Regions of Armenia.

Dynamics in wild O. vulgare L. Population Structure (according to 2007, 2008 and 2009 study results)

Figure 27. Population’ age structural distribution and its modification in the (A) Northern; (B) Central; and (C) Southern Regions of Armenia.

The age distribution of population is the proportion of individuals at different ages. In respect with the average proportion of seedling, saplings, mature plants it has been determined 95

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. demographically rapidly growing, stable or depleting populations of M. officinalis and O. vulgare species across the country (Lauenroth, W.K. & Adler, P.B. 2008). Fig. 26 exposed that rapidly growing populations are mostly located in the south regions of the country (Syunik and Vayots Dzor Regions). For instances, Artsvanik, Srashen, Jermuk populations of M. officinalis L., with about 40% of average youngster plants are considered as growing populations. In particular, the increase is documented 5-8% (seedlings-sampling) youngster plants in the Artsvanik, 4- 4% in the Srashen, 5-6% in the Jermuk populations in the overall study period (Fig.26 (C)). Jermuk and Artsvanik populations of O. vulgare L. species with around 45% of youngster plants are also rapidly growing populations located in the south regions of Armenia (Fig.27 (C)). The increase rate is documented 4-7% (seedlings-sampling) youngster plants in the Jermuk and 5- 6% in the Artsvanik populations in the overall study period. In fact, the average growth rate of youngster plants is comprised above 10% in three years study course, which is sufficient to classify them as demographically rapidly growing population in the future, if we exclude external environmental or biological factors (Lauenroth, W.K. & Adler, P.B. 2008, Odum E.F. 1971). The number of annual increase of seedlings and samplings is different in the populations of O. vulgare L. species. For example, seedlings are growing gradually with 2% of annual increase in the Jermuk and Artsvanik populations. However, samplings are growing rapidly with almost 5% in 2007-2008 vegetative years in the Jermuk; and with 4% in the Artsvanik populations. This type of grow might be connected with number of environmental or biological factors. Researchers recommend future continues monitoring to identify and monitor number of specific factors that influence on this intricate pattern of structural dynamics and identify future possible shift. On the other hand, the annual increase is documented almost equal e.g. 3-4% (seedlings- sampling) youngster plants in the Artsvanik , 2-3 % in the Srashen and in the Jermuk populations of Melissa officinalis L. species.

Other populations of Melissa officinalis L. species, located in the south (Syunik Region) of Armenia are also considered to be growing populations in terms of age proportion and structure, e.g. Karchevank, Shikahogh, Kapan and Tsav populations is exposed with about 30% of average youngster plants (seedling and sampling) in the overall study period. The increase is documented 4- 4% (seedlings-sampling) youngster plants in the Tsav, 3- 3% in the Kapan, 3-5% in the Shikaogh and 5-3% in the Karchevank populations in the overall study period.

Fig. 25 & Fig. 27 display O. vulgare L. populations with south locations e.g. Eghegis, Kapan Meghri New Highway, Chakaten with the average 30% of youngster plants could be considered stable populations in terms of age structure. As, the clear examination will reveal the fact that the annual increase of youngster plants (seedling-sampling) is relatively low by comprising 1-1% in the Chakaten; 2-2% in the Kapan Meghri New Highway, where is in the Egeghis population the youngster plants proportion is constant during 2007-2008 vegetative years, then slightly grow in 2009 with 2% of each. Thus these populations would be considered demographically stable to compare with Artsvanik and Jermuk population of O. vulgare L..

Demographically declining populations of O. vulgare L. species is considered Ijevan, Novoseltsovo from the northern and Aparan, Lichk from the central regions of Armenia. In fact, the average youngster plants proportion is documented between 15-20%. In addition, field observations revealed, youngster plants proportion’s declining tendency during the study period. In fact, the decrease of seedlings and samplings comprise around 2% at the Ijevan, Novoseltsovo (northern regions), and almost 3% at the Aparan, Lichk (central regions) populations. It is significant to mention that the proportion of youngster plants is less e.g. with 5% in the Ijevan population to compare with the others. Thus the situation in the Ijevan population is more drastic in terms of demographic decrease. Anyway, mentioned four populations are at future higher risk towards weather, climatic change, as well as habitat and other environmental factors (e.g. light, water, nutrients and other environmental limiting factors etc.). Simultaneously, field studies were revealed that demographically declining populations of M. officinalis L. species are located mostly in the northern regions of the country. With the only 17% of average youngster plants and 2-1% (seedling-sampling) of gradually recorded decrease in the study period, Getahovit and Ayrum populations are in declining condition and vulnerable towards future environmental deterioration, or other risks. In fact, the most drastic decrease of youngster plants is documented in the Ayrum population with almost 10% from 2007 to 2009. This number is two times less in the Getahovit population. Orgov population of M. officinalis L. species from the central regions of Armenia could be considered as gradually growing population with around 20% of youngster plants in 2007 vegetative season, which rise up almost twice overall the study period by growing gradually with 2% of per year.

On the other, Garni population with the average 25% of youngster plants and gradually depleting size of youngster plants e.g. with 1-2% of per year, is considered as declining population in age structure. Although, in 2007 the proportion of youngster plants number of 28% is documented higher to compare with Orgov population. Interestingly, the decrease is documented almost the same proportion in the Getahovit and Garni population, however in the Garni population the proportion of youngsters with 10% higher. Thus, Getahovit is considered to be at relatively higher risk.

There are numbers of significant biological as well as habitat factors that could affect on wild Melissa officinalis L. and Origanum vulgare L. population’s dynamics in structure. The observed dynamics in age structure during the study is characterized with vegetative sprouting in the plant populations, which includes number of genetically distinct individuals, of vegetative reproduced individuals, of leaves, branches, tillers, stems, flowers, etc., also movement of seeds into and out of the population and storage in the soil. It is observed and revealed age heterogeneity psychological behavior of the plants population, which is, in particular new youngster plants (seedlings/samplings) generation, connected with number of biological factors and combined with relevant environmental/weather conditions, resources availability, competition, disturbance, availability of propagates (biogeography) etc. In population ecology, the structure–abundance (S–A) relationship is the relationship between the abundance of populations and the structure of population of their heterogeneity of ages within a region.. This relationship is perhaps one of the well-documented relationships in population demography and applies interpopulations’ demographic growth (among the different habitats). In population ecology this relationship is described with Demographic Index that displays population is either demographically growing or declining. The analysis of results is revealed a significant correlation between population abundance and its structural dynamics in our studies (Stover, J.P., Kendall, B.E. & Fox, G.A., 2012, Kunin, W.E. 1997b). Fig. 28 is exposed structure dynamics either increasing or decreasing proportions of youngsters in population in respect with Odum E.P. (1971) demographic index during the three years of study course with various average abundance criteria. Population with increasing demographic index DI≥10+ e.g. Artsvanik, Jermuk Srashen of M. officinalis L. and Artsvanik, Jermuk of O. vulgare L. species is documented with the highest average abundance number 3 plant/m2 during the three study years period.

We can conclude that the equilibrium of index structure of these populations in combined with favorable environmental conditions and resources availability of the habitat factors such as in the south regions and plant vegetative growth intensity, (heights, stems) could reinforce the hypothesis that the abundance of populations may enforce population’s demographic growth in future in these locations (Stover, J.P., Kendall, B.E. & Fox, G.A., 2012). Demographic Index that explicit youngsters dynamics in proportion of population that covered minimum of three year study course in respect with Odum E.F. (1971) in each of the eleven population of M. officinalis L. and nine populations of O. vulgare L. were observed during the study period.

Figure 28. M. officinalis and O. vulgare Population Abundance and Structure Dynamics in respect with Demographic Index. Population is demographically considered as DI ≥ 8-10+ rapidly growing, DI ≤6-8+ gradually growing, DI ≤4-6+ stable , DI ≤4-6- gradually declining and DI ≥8-10- rapidly declining.

Kapan, Tsav, Shikahogh, and Karchevank populations of M. officinalis L. from south location and Orgov from central location with demographic Index ≤6-8+ are considered as demographically gradually growing and also possess relatively higher abundance 2 plant/m2 . However, populations with the lowest abundance number ≤1plant/m2 are documented as demographically rapidly DI≥8-10- and gradually DI ≤4-6- declining e.g. Ijevan, Novoseltsovo 99

(from northern region), Aparan, Lichk (from central regions) populations of O. vulgare L., and Ayrum, Getahovit (from northern regions) and Garni (from central region) populations of M. officinalis L. are declining (Fig. 28). Respectively, Chakaten, Kapan Mghri New Highway, Eghegis populations from south locations of O. vulgare L. species are documented as stable populations in respect with demographic index of and abundance is higher in comparison with demographically declining populations (Fig. 28). The analysis of structural dynamics of populations during the study can give us a comprehensive approach to the various causes and impacts on it. Habitat factors could be one of these various causes that might impact on population demographic growth type and its dynamic during the study period (Lauenroth, W.K. & Adler, P.B. 2008). In this respect, we can conclude that habitats mostly in the south locations are more favorable and possess better available resources for the plant species demographic growth rather than habitats with northern and central location in the country. However, we should not exclude also that weather conditions e.g. abrupt weather changes, early spring frost that are common phenomenon in the northern and central regions could affect negatively on youngster plants, in particular on seedlings propagation and grow. Among the other various biotic, abiotic, environmental and other factors one of the reasons in decline of the age structure is Powedery mildew diseases that is documented and observed during the study period in Orgov population of M. officinalis L. species (refer 4.2 section pp 76). In particular, seedlings and saplings are susceptible towards the Powedery mildew diseases. Antropogenetic impacts, especially overgrazing, could affect negatively on the youngster plants proportion within a population. Therefore, demographically growing populations are located mostly in the south regions (Syunik, Vayots Dzor) of Armenia. For instances, Artsvanik, Srashen, Jermuk populations of M. officinalis L., with about 40% of average youngster plants and Jermuk and Artsvanik populations of O. vulgare L. species with around 45% of youngster plants. Simultaneously, field studies were disclosed declining populations are mostly located in the northern (Tavush, Lori) and central regions (Kotayq, Aragatsotn, Gegharkunik) of Armenia. In fact, Ijevan, Novoseltsovo (Tavush, Lori) and Aparan, Lichk (Kotayq, Aragatsotn, Gegharkunik) populations of O. vulgare L.; and Getahovit, Ayrum, (Tavush, Lori) populations of M. officinalis L. species are documented with the average youngster plants (seedlings, saplings) between 15-20% of proportion.

100

4.5. Distribution Modeling of Melissa officinalis L. and Origanum vulgare L. in The Republic of Armenia

In 1957, Hutchinson defined the fundamental ecological niche as a multi-dimensional range of environmental conditions within which a species can survive and grow (Pearson and Dawson, 2003). However, observed occurrences of species can only provide an approximation of the realized niche of a species (i.e., the range of environmental conditions). Several methods have been used to approximate the realized niche as a statistical model between occurrence and observed environmental variables at occurrence locations (Guisan and Zimmermann, 2000; Austin, 2002; Elith and Leathwick, 2009). If the environmental conditions encapsulated within the fundamental niche are plotted in geographical space then we have the potential distribution or niche. While suitable environmental conditions determine a species’ fundamental niche, biological factors such as competition tend to reduce the fundamental niche into the realized niche (Hampe, A. 2004, Guisan, A. & Zimmermann, N.E. 2000).The potential distribution of a species can be seen as the geographical expression of its realized niche at a particular time i.e., where there is a fulfillment of both abiotic and biotic requirements (Huston, M.A., 2002, Huntley, B., P.M. Berry, W. Cramer, & A.P. Mcdonald, 1995).

Distribution modelling or environmental niche modelling (ENM) used to predict environmental suitability as function of given environmental variables as well as to produce hypotheses. ENM was used to generate a series of models of present-day annual mean temperature and precipitation patterns in Armenia, and to predict suitable habitats for M. officinalis L. and O. vulgare L. in Armenia given these present-day climatic conditions (Thuiller W. 2003). In addition, mean temperature and precipitation variables were manipulated in line with future climate change predictions to model the effects of changing these variables on habitat suitability (Thorn J.S., Nijman V., Smith D. & Nekaris K.A.I. 2009). So, nine environmental variables and 34 occurrence points of M. officinalis L and 67 occurrence point’s O. vulgare L. species were used to generate models using the GARP algorithm based on correlative approach.

The GARP approach models the ecologic niche of species based on relating point- occurrence data to the electronic maps of relevant ecological dimensions, producing a heterogeneous set of rules that describe the potential distribution of species (Peterson et al., 2002b).

101

It has been determined the predicted ecological niche/ potential niche of each species as well as habitat suitability of them in Armenia during the study (Fig. 29 & 30).

Figure 29. The portion of predicted potential niche of M. officinalis L. across the different regions of Armenia.

Figure 30. The portion of predicted potential niche of O. vulgare L. across the different regions of Armenia.

Fig. 29 &30 were illustrated the predicted potential niche of wild medicinal plant species n different regions of Armenia. In general, the territory of potential niche of O. vulgare and M. officinalis species is predicted less in the northern e.g. Tavush, Lori and central e.g. Aragatsotn, Kotayk, Gegharkunik, Ararat regions to compare with south regions e.g. Vayots Dzor, Syunik.

Within GARP, input data are further divided randomly and evenly into training and intrinsic testing 102

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. data sets. GARP works in an iterative process of rule selection, evaluation, testing, and incorporation or rejection. A method is chosen from a set of possibilities (e.g. logistic regression, bioclimatic rules), applied to the training data and a rule is developed or evolved (Stockwell, 1999). GARP relates ecologic characteristics of occurrence points to those of ecologic characteristics sampled randomly from the rest of the study region, developing a series of decision rules that best summarize factors associated with presence (Peterson et al., 2000). The method (Anderson et al. 2003), for analyzing that would be using linear multiple regressions to predict the error values (omission and commission), using the information on whether a particular layer was used on a task as an independent variable. The χ2 statistic was used to calculate the probability of a random prediction being similar as the one generated by GARP. For each species we produced 10 models. The best subset procedure (Anderson et al. 2003) was used to select 1 of 10 models with the highest predictive values. The 10 best GARP models resulting from ecological niche analysis for these species presented statistically significant χ2 values (p < 0.01), indicating that the models are quite predictive and that they summarize the ecological requirements of each species. The highest levels of commission and omission errors were observed for species (Table 8).

Table 8. Statistical parameters for the 10 models

Species χ2 p Commission a Omissionb M. officinalis 67.7-87.2 < 0.01 19.4-27.9 0-7.2 O. vulgare 133.7-299.0 < 0.01 11.0-21.3 0-3.5

a: percentage of the predicted area that exceeds the recorded occurrence; b: percentage of test points that were predicted absent, but are presence records of the species. Values present the range of the 10 models in the best subset selection; p: probability that a random prediction has the same number of correct predicted points as the one generated by DesktopGarpapplication (version 1.1.6); χ2: chi-square statistics.

These models were imported into ARC VIEW (version 3.3) and summed using the Map Calculator function in order to generate a single cumulative predictive map ranging from zero (predicted absence) to 10 (the coincidence of the 10 best models, the highest predictive agreement of presence). We considered the coincidence of the 7-9 best models (moderate to high predictive agreement of presence) to be appropriate for describing the species' potential distribution. This

103

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. procedure added a component of conservatism to the predictions of GARP, which otherwise could over-extrapolate the potential areas. Fig. 31 & 32 were illustrated habitat suitability levels of each species determined by the number of coincidences of ten models in the same pixel within the exact area.

Predicted Geographic Distributions of Wild Melissa officinalis L. in Armenia

Figure 31. The 10 best annual models were summed and ranked by the criteria of how many times each model predicted the same pixel within the exact area, classified as: High, dark green plots (7-9 times); Moderate habitat suitability, green plots ( 5-7); Low, light green plots (1-5 presence agreement). The blank areas represent the absence predicted by the models.

104

Predicted Geographic Distributions of Wild Origanum vulgare L. in Armenia

Figure 32. The 10 best annual models were summed and ranked by the criteria of how many times each model predicted the same pixel within the exact area, classified as: High, dark green plots (7-9 times); Moderate habitat suitability, green plots ( 5-7); Low, light green plots (1-5 presence agreement). The blank areas represent the absence predicted by the models.

Within GARP, input data are further divided randomly and evenly into training and intrinsic testing data sets. GARP works in an iterative process of rule selection, evaluation, testing, and

105

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. incorporation or rejection. A method is chosen from a set of possibilities (e.g. logistic regression, bioclimatic rules), applied to the training data and a rule is developed or evolved (Stockwell, 1999). GARP relates ecologic characteristics of occurrence points to those of ecologic characteristics sampled randomly from the rest of the study region, developing a series of decision rules that best summarize factors associated with presence (Peterson et al., 2000b).

Importantly, most of the high suitable points fell into the south territory of Armenia. South regions of Armenia, in particular Syunik region are characterized as having the highest number of predicted potential niche of the species. Predicted potential niche occupied 63% (M. officinalis) and 60% (O. vulgare) of the territory from south regions. Potential niche of M. officinalis L. from central regions with ca. 2% and northern regions with ca. 1% less than is recorded of O. vulgare L.

In general, the number of potential generations decreased at sites of greater proximity to the central and northern regions of Armenia. The absence of O. vulgare, M. officinalis in two central (Armavir, Ararat) and Shirak regions located in the north western of the country may have various causes, including historical restrictions (e.g., geographic barriers and/or lack of sufficient dispersal opportunities) and biotic interactions (such as competition with related species).

Fig. 33 & 34 were explicated the amount of territory of high, moderate and low habitats in different regions of Armenia.

106

Figure 34. Habitat Suitability Pattern of Figure 33. Habitat Suitability Pattern of Melissa Origanum vulgare L. in in the (A) northern; (B) officinalis L. in the (A) northern; (B) central and central and (C) southern Regions of Armenia. (C) southern Regions of Armenia.

The smallest area of high habitat suitability of M. officinalis L. and O. vulgare L. species distribution is documented in the Northern Region of Armenia: with 20km2 predicted area of M.

107

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. officinalis L. and 40km2 of O. vulgare L. is indeed low with its surveyed proportion into the potential niche. Unlike, northern regions high habitat suitable area of O. vulgare L. is predicted 4 and of M. officinalis L. 6 times higher in the central regions (Kotayk, Aragatsotn).

However, the general pattern indicates that high habitat suitable areas have not identified by modeling in certain regions located in the central and northern territory of Armenia (Gegharkunik, Ararat) (Fig. 34). Predicted high habitat territories with its highest range from 1200km2 for M. officinalis L. to 2000km2 for O. vulgare L. species are modeled in the south regions of Armenia.

Moderate habitat suitability with its largest territories of 2250 km2 for M. officinalis L., and 1600 km2 for O. vulgare L. is also predicted in the south region of Armenia. It is about 2/4 times (for O. vulgare L.) and 7/4 times (for M. officinalis L.) higher respectively from the central/ northern regions of Armenia. On the other hand, low habitat suitable areas are predicted around 150 km2 for O. vulgare L. and 860 km2 for M. officinalis L. species in the south regions are documented as the smallest areas across the country. This number is modeled 6/5 times (for O. vulgare L.) and 1.4/1.2 times (for M. officinalis L.) higher from respectively central/northern regions of the country.

So, the northern regions where high habitat suitability is missing and the plant potential niche is mostly encompasses low and moderate suitable habitats are characterized mainly as dry mountainous steppes landscapes with warm, dry summers and mild winters, suggesting that it would be comparatively less suitable for M. officinalis L. to thrive.

On the other hand, the plant growing high suitable areas are mostly spread on mountain/ meadow steppes as well as forest landscapes, ridge tops among steppes mostly in south part of the country as well as with less territory in central and in the north-east parts.

Therefore, here the climate change impact might not be so crucial. In respect with this, north part of the country would be more stressful towards weather and climate change. Central regions could be less vulnerable in this point, with exception of Gegharkunik region.

Distribution modelling of closely related species/populations occurring in adjoining or slightly overlapping areas is useful for exploring the factors affecting the species’ geographical distributions and for providing directional hypotheses that can be tested in future studies. By examining the congruence or discordance between predicted and actual distributions, it is possible 108

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. to evaluate the potential role of ecological and historical factors in determining a species’ geographical distribution (Anderson et al. 2002). Furthermore, we conducted a principal component analysis (PCA) with all nine environmental variables from the occurrence areas of Melissa officinalis L. and Origanum vulgare L. species. PCA reduces the dimensionality of the original set of variables with little loss of information by transforming the original variables into a new set of independent components (Robertson, M. P., N. Caithness, and M.H. Villet. 2001, Foottit & Sorensen 1992). The components that accounted for the majority of the total variance were examined and the most highly loaded variables were analyzed. This approach was used to identify decisive environmental variables that influence most on the geographical distribution and abundance of the species (Table 9).

Table 9. Variables correlation with the first two principal components (PC1 and PC2) which reflects the importance of the environmental variables used do produce the ecological niche models of Origanum vulgare L. and Melissa officinalis L.

Variables PC1 PC2 Normalized difference vegetation index -0.16 -0.75a Altitude -0.87a 0.19

Annual mean temperature 0.88a -0.18

Mean diurnal range -0.56 -0.36 Temperature seasonality -0.73a 0.12 Max. temperature of warmest month 0.70a -0.37 a Min. temperature of coldest month 0.96 0.00 a Mean temperature of wettest quarter 0.73 -0.23 Mean temperature of driest quarter 0.95a -0.15 Temperature annual range -0.71a -0.31 a Mean temperature of warmest quarter 0.85 -0.25 a Mean temperature of coldest quarter 0.96 -0.26 Annual precipitation -0.32 -0.84a Precipitation of driest month -0.02 -0.77a

Precipitation of wettest month -0.53 -0.31 Precipitation seasonality 0.67 0.40 Precipitation of wettest quarter -0.07 -0.83a Precipitation of driest quarter -0.49 -0.31 Precipitation of warmest quarter -0.59 -0.15

Precipitation of coldest quarter 0.43 -0.41 a: correlation > 0.7.

The temperature, altitude, NDVI and precipitation values of O. vulgare and M. officinalis occurrence areas were compared and our results showed clear ecological niche differences between these species. In the multivariate analysis, including all environmental variables from the 109

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Doctoral Dissertation in Biology.-Daugavpils University:1-128. occurrence areas for O. vulgare and M. officinalis the first two principal components (PC) summarized 66% of the environmental variance and also showed clear niche differences between these species (Fig. 33 & 34). PC1 explained most of the total variation and was correlated with altitude and temperature variables, mainly temperature in the coldest and driest months. PC2 explained 19% of the environmental variance and was correlated with NDVI and precipitation variables (Table 9). Low values of precipitation/NDVI combined with high values of temperature in the coldest and driest months distinguish the ecological niche of O. vulgare from that of M. officinalis. PCA was also useful for measuring niche amplitude for each species, as previously reported by Ibarra-Cerdeña et al. 2009. The maximum variation in niche amplitude occurred along the first axis (PC1), mainly associated with a temperature gradient. Moreover, a precipitation-humidity gradient was observed along the second axis (PC2). Our results support the hypothesis that the warmest and driest areas of the country produce ideal conditions for O. vulgare L. occurrence. In contrast, the colder areas with the highest altitude, precipitation and NDVI values seem to be favorable for M. officinalis L.. Further comparative phylogeographical studies may also evaluate the role of demography and climatic events in shaping the diversity and distribution of M. officinalis and O. vulgare populations in Armenia. Further studies were shown that the climate and geography of northern and central regions of Armenia make them highly sensitive to the climate change impact (Houghton J.T., Y. Ding, D.J. Griggs et al. 2001). It is predicted that, as a result of global warming, the average temperature in Armenia will rise by 2-3ºC, and rainfall will decrease by 10-15%, within the next 50-100 years (1th National Report, March 1999; “Adaptation to Climate Change Impacts in Mountain Forest Ecosystems of Armenia”UNDP/GEF/00051202). Distribution modeling identified mostly northern and central regions of Armenia less suitable for these species distribution and abundance thus this support the hypothesis that existing populations might face with the highest risk of extinction under global climate change impacts. Comparison of the GARP modeled the prediction of the distribution of M. officinalis and O. vulgare species across the country and defined ecological risk (Global Climate Change scenario) suggests that the central and especially northern are characterized by a hot and semi-arid environment is the highest risk areas.

110

Chapter V. CONCLUSIONS

1. As a result of carried out research the hypothesis about the fact that, that O. vulgare L. and M. officinalis L. (Lamiaceae) distribution is decreased from the northern and central regions of Armenia where they were historically recorded was confirmed. During the field trips in 2007- 2011 five population of O. vulgare L. :Urcadzor, Aghveran, Narashen, Nor Bayazet from the central and Dilijan (Vartdaget), Gamzachiman from northern regions could not be re-located. And, four population of M. Officinalis L. : Alaverdi, Akhtala, Aygedzor, Ijevan from northern regions could not be re-located and presumed to be extinct. Remaining population would face a greater risk of extinction and are more susceptible towards climate change impact was confirmed by GARP modeling. 2. As a result of carried out research the hypothesis about the fact that, there are more than historically known populations of O. vulgare L. and M. officinalis L. located mostly in the south regions of Armenia was confirmed. Field studies were revealed and found five new populations of M. officinalis L.: Artsvanik, Srashen, Tsav, Jermuk located in the south and Ayrum located in the northern regions of Armenia. At the same time, four new populations of O. vulgare L. species: Eghegis, Kapan Meghri New Highway, Artsvanik were discovered in the south regions and Novoseltsovo in the north region of Armenia. The collected data suggests that the abundance and distributional range of O. vulgare L. and M. officinalis L. species is expanding, particularly in the south-east regions of Armenia. 3. The third hypothesis of the research is the presumption that populations from south regions have higher carrying capacity determined by populations’ growth and habitat peculiarities were confirmed. Artsvanik, Srashen, Jermuk populations of M. officinalis L. and Artsvanik, Jermuk populations of O. vulgare L. are characterized with an intrinsic rate of increase and are capable of expanding in future is foreseen. The expanding density and abundance (plant height, stems number) of these populations has less influence on the exponential growth rate rather than habitat characteristics were confirmed. For instances, the average density in the collected data was determined the same e.g. 2.5plant/m2 (and the stem numbers in average 3-6 /plant) in the Artsvanik and Jermuk populations of O. vulgare L., however during the course of the study the rate of exponential growth was documented in different sizes: with the number of 130 individuals in Artsvanik and 350 individuals in Jermuk populations. This data suggests the habitat characteristics and weather conditions on populations’ growth rate have more influence

111

except the density dependant factor, and populations have higher carrying capacity due to habitat peculiarities therefore theoretically these populations expansion located in the south regions is predicted in future. 4. The identification of particular factors of habitats that could be assessed quantitatively (e.g., annual precipitation, soil pH, humus concentration) or qualitatively (e.g., interaction of the plant grow pattern and the environment, anthropogenic threats) that had either positive or negative effects on the survival and fitness of the located populations were carried out. Created essential data can provide necessary database for the relevant approaches of these valuable wild medicinal plants future domestication in Armenia. This would foster the improvement of their sustainable use and traditional medicine in the country. 5. The fourth hypothesis of the research is the pressumtion that population dynamics is linked to differences in habitat quality and in among years climatic variation. And, that the effects of climate depend on local habitat quality was confirmed. Researches and Principal component analysis (PCA) of relationships between variation in environmental, habitat factors and stochastic population growth dynamics identified that temperature and precipitation variation influences on the populations’ growth dynamics were interdependent on the habitat factors (e.g. soil pH, humus concentration). 6. The results suggested that warm summer temperatures had negative effects on the species population growth rate, acting mainly through survival. In particular, the exceptionally warm summers with low precipitation resulted in very low population growth rates in the northern and central regions of Armenia. This influence is particularly higher in the habitats that featured with poor soil and not relevant pH environment. 7. While observing the impacts of among year temperature variations and the low annual temperature on populations’ dynamics among the different regions of Armenia, the most crucial effects were documented in the northern and central regions. In addition, these habitats are featured with comparatively lower long term precipitations which reinforce the negative impacts of climate variations along with unfavorable habitat features, such as poor soil and not relevant pH environment. 8. So, we correlated temporal variation in population dynamics to differences in temperature and precipitation and certain habitat factors, particularly soil pH and humus concentration. Our results with the wild M. officinalis L. and O. vulgare L. species show that both local environmental factors and climatic variation among years influence population dynamics, and that the effects of climate depend on local habitat quality.

112

9. The fact that, under environmental and habitat conditions similar to those observed during the study, populations were observed to decrease and underscores their vulnerability towards future climate change impacts in the northern and central regions of Armenia. 10. It is observed and revealed age heterogeneity psychological behavior of the plants population, and documented demographically growing, stable and declining population, which is, in particular youngster plants (seedlings/samplings) generation rate or proportion dynamics in age structure and connected with the number of biological factors in combined with relevant environmental/weather conditions, resources availability, competition, disturbance, availability of propagates (biogeography) etc. So, demographically growing populations are located mostly in the south regions (Syunik, Vayots Dzor) of Armenia. For instances, Artsvanik, Srashen, Jermuk populations of M. officinalis L., with about 40% and Jermuk and Artsvanik populations of O. vulgare L. species with around 45% of youngster plants. Simultaneously, field studies were disclosed declining populations are mostly located in the northern (Tavush, Lori) and central regions (Kotayq, Aragatsotn, Gegharkunik) of Armenia. In fact, Ijevan, Novoseltsovo (Tavush, Lori) and Aparan, Lichk (Kotayq, Aragatsotn, Gegharkunik) populations of O. vulgare L.; and Getahovit, Ayrum, (Tavush, Lori) populations of M. officinalis L. species are documented with the average youngster plants between 15-20% of proportion. 11. The analysis of results is revealed a significant correlation between population abundance and its structural dynamics. Population with increasing demographic index DI≥8-10+ e.g. Artsvanik, Jermuk Srashen of M. officinalis L. and Artsvanik, Jermuk of O. vulgare L. species is documented with the highest average abundance number 3plant/m2 during the three study years. Populations with the lowest abundance number ≤1plant/m2 are documented as demographically rapidly DI≥8-10- and gradually DI ≤4-6- declining e.g. Ijevan, Novoseltsovo (from northern region), Aparan, Lichk (from central regions) populations of O. vulgare L., and Ayrum, Getahovit (from northern regions) and Garni (from central region) populations of M. officinalis L. species. 12. Among the various causes the impact of habitat factors, environmental, weather conditions and the plants growing biological characteristics have been discussed to reveal the specific interaction on the species population future demographic growth and dynamics. Researches shows that habitats mostly in the south locations are more favorable and possess better available resources for the species population demographic growth rather than habitats with northern and central location in Armenia.

113

13. The extensive phenotypic variability and adaptive ability displayed by individual plants within these populations suggests that M. officinalis and O. vulgare in Armenia represents a rich genetic resource for the species as a whole. 14. The identification of population ecological properties and growth pattern reveal the interaction with its environment, in particular necessary for assessing population resilience and vulnerability in different habitats under global climate change impacts. In this respect, populations located in the northern and central regions of Armenia might face with higher risk of extinction. 15. Occurrence point, 67 of O. vulgare and 34 of M. officinalis, along with environmental variables are imported into the genetic algorithm and modeled the spatial distribution of environments that are suitable for the species. The highest habitat suitability of the species within the predicted environment (potential niche) is modeled in the south regions of Armenia by GARP defined on correlative approaches in 2012-2013 study years. In addition, mean temperature and precipitation variables were manipulated in line with future climate change. In fact, northern and central regions are predicted less suitable for species distributional range and identified most stressful habitats under the global climate change impacts and anthropogenic threats. 16. Principal component analyses (PCA) indicate 1) the level of humidity; and 2) the temperature the most decisive environmental variables that influence on the predicted geographical distribution of the species. In particular, the level of humidity influenced more on M. officinalis L.; and the temperature on O. vulgare L. predicted distribution in Armenia. 17. The researchers recommend continuous monitoring of specific sites and the effects of the independent variables that influence populations’ dynamics of wild Melissa officinalis L.; Origanum vulgare L. and identifying threats to conservation. 18. This research has provided a baseline dataset that can be used for the development of further ex situ and in vitro strategies to conserve unique genotypes, as well as assess the sustainability of wild populations with regard to the IUCN Red Book Criteria, of this important medicinal and culinary species in Armenia.

114

ACKNOWLEDGEMENTS

This dissertation is dedicated to the 100 years of Memory of the Armenian Genocide. The memorial date is on 24 April 1915. I would like to express the deepest appreciation to my academic supervisors, Dr. biol. prof. Arvids Barsevskis and Prof., Dr. sc. agr. Andreas Melikyan, they have been tremendous mentors for me. I would like to thank you for encouraging my research and for allowing me to grow as a research scientist. Your advices on both researches as well as on my career have been priceless. Words cannot express how grateful I am to my very special and cherished mentor Dr. agr.sc, prof. from Texas A&M University, who has the attitude and the substance of a genius he continually and convincingly conveyed a high spirit of adventure in regard to researches and constructive thinking. His encouragements and contributions were tremendous for the development of my academic English Language and my studies in abroad. Without his guidance and persistent help this dissertation would not have been possible. I thank my colleagues from Armenian National Agrarian University (Yerevan, Armenia) and Daugavpils University for huge support and collaboration. I am thankful to the “Environmental Conservation and Research Center” of American University of Armenia for organizing field expeditions and Dr. biol. Nina Stepanyan-Gandilyan, senior specialist Kamilla Tamanyan from Department of Systematic and Geography of Higher Plants Institute of Botany, National Academy of Sciences of Republic of Armenia for their kindly support of participating in the species identification and verification works. I am grateful to the director of doctoral study programme “Biology” Dr. biol. prof. Arvids Barsevskis, Head of the Department Systematic Biology Dr. biol. Uldis Valainis, European Social Fund within the project Nr. 2012/0004/1DP/1.1.2.1.2/11/IPIA/VIAA/011, “Support of the implementation of doctoral studies at Daugavpils University” and its leader Dr. paed. Eridiana Olehnovica for supports, responsiveness and understanding. Significant advices and instructions in the formation of my publications have given by reviewers and co-authors, therefore I would like to thank them for critical comments, constructive collaboration and the given opportunity to improve my manuscripts. In particular, I am very grateful to Dr. pharm Sara Crockett from Institute of Forensic Medicine, Medical University Graz, (Graz, Austria) and Dr. hort. sc. Tomas Ayala Silva from the Subtropical Horticulture Research Institutute of USAD for their huge support. Special thanks to my beloved family, parents Anahit Hayrapetyan and Vazgen Abrahamyan, sisters and brothers who supported my studies and incented me to strive towards my goals. I would like to thank to all of my friends for rising up my spirit and enthusiasm to achieve the desired progress in my research.

115

REFERENCES

1. Åberg, P. (1992) Size-based demography of the seaweed Ascophyllum nodosum in stochastic environments. Ecology, 73, 1488–1501. 2. Alicia, V., Daniel G., María G., Johan, E., (2014). Contrasting effects of different landscape characteristics on population growth of a perennial forest herb, Ecography, Wiley Online Library, 37, 3, 230. 3. Anderson RP, Lew D, Peterson AT, Evaluating predictive models of species’ distributions: criteria for selecting optimal models. Ecol Model, 2003, 162:211–232 4. Anderson R.P., Gómez-Laverde M. & Peterson A.T. (2002a). Geographical distributions of spiny pocket mice in South America: insights from predictive models. Global Ecology and Biogeography, 11, 131-141. 5. Austin, M. P. et al. (1990). Measurement of the realized qualitative niche: environmental niches of five Eucalyptus species. Ecol. Monogr. 60: 161_177. 6. Austin, M. P. 1987. Models for the analysis of species’ response to environmental gradients. Vegetatio 69: 35_45. 7. Anderson RP, Laverde M, Peterson AT (2002) Using niche-based GIS modeling to test geographic predictions of competitive exclusion and competitive release in South American pocketmice. Oikos 93:3–16. 8. Araújo M.B. & Luoto M. (2007). The importance of biotic interactions for modelling species distributions under climate change. Global Ecology and Biogeography, 16, 743-753. 9. Araújo, M.B., and R.G. Pearson. Equilibrium of species' distributions with climate. Ecography 28, 693-695, 2005. 10. Araújo, M.B., R.G. Pearson, W. Thuiller, and M. Erhard. (2005a). Validation of species climate envelope models under climate change. Global Change Biology 11, 1504-1513. 11. Araújo, M.B., W. Thuiller, P.H. Williams, and I. Reginster. 2005b. Downscaling European species atlas distributions to a finer resolution: Implications for conservation planning. Global Ecology and Biogeography 14, 17-30. 12. Araújo, M. B., and P.H. Williams. 2000. Selecting areas for species persistence using occurrence data. Biological Conservation 96, 331-345. 13. Babbie, E., Mouton, J., Voster, P & Prozesky, B., (2001).The practice of social research. Oxford University press, Cape Town.

116

14. Bacchetta, G. Fenu, R. Gentili, E. Mattana, S. Sgorbati, Preliminary assessment of the genetic diversity inLamyropsis microcephala(Asteraceae), Plant Biosystems - An International Journal Dealing with all Aspects of Plant Biology, 2013, 147, 2, 500CrossRef 15. Bakkenes M., Alkemade J.R.M., Ihle F., Leemans R. & Latour J.B. (2002). Assessing effects of forecasted climate change on the diversity and distribution of European higher plants for 2050. Global Change Biology, 8, 390-407. 16. Beaumont L.J., Hughes L. & Pitman A.J. (2008). Why is the choice of future climate scenarios for species distribution modelling important? Ecology Letters, 11, 1135-1146. 17. Berry, P.M., T.P. Dawson, P.A. Harrison, and R.G. Pearson. 2002. Modelling potential impacts of climate change on the bioclimatic envelope of species in Britain and Ireland. Global Ecology and Biogeography 11, 453-462. 18. Best A.S., Johst K., Münkemüller T. & Travis J.M.J. (2007). Which species will successfully track climate change? The influence of intraspecific competition and density dependent dispersal on range shifting dynamics. Oikos, 116, 1531-1539. 19. Biodiversity: Climate change and the ecologist. 2007 Nature, 448, 550-552. 20. Bishop JG, Schemske DW (1998). Variation in flowering phenology and its consequence, Ecol., 79:534-546. 21. Bourg, N.A., W.J. McShea, and D.E. Gill. 2005. Putting a cart before the search: Successful habitat prediction for a rare forest herb. Ecology 86, 2793-2804. 22. Broennimann O., Treier U.A., Müller-Schärer H., Thuiller W., Peterson A.T. & Guisan A. (2007). Evidence of climatic niche shift during biological invasion. Ecology Letters, 10, 701- 709. 23. Boyce, M.S., P.R. Vernier, S.E. Nielsen, and F.K.A. Schmiegelow. 2002. Evaluating resource selection functions. Ecological Modelling 157, 281-300. 24. Campbell, N. E., & Reece, J. B. (2002). Biology (6th ed.). San Francisco: Benjamin Cummings. 25. Canhos VP, Souza S, De Giovanni R, Canhos DAL, Global Biodiversity Informatics: setting the scene for a “new world” of ecological forecasting. 2004; Biodiversity Informatics 1:1. 26. Caswell, H. , Matrix Population Models. Sinauer, Massachusetts, 1989. 27. Chase, J.M., and M.A. Leibold (2003). Ecological niches: Linking classical and contemporary approaches University of Chicago Press, Chicago. 28. Climate Change Problems, collected articles, II issue, GEF, within the framework of the UNDP/GEF/ARM/95/G31A/1G/99, Yerevan, “Lusabac”, pages 353.

117

29. Costa GC et al., (2010). Sampling bias and the use of ecological niche modelingin conservation planning: a field evaluationin a biodiversity hotspot. Biodiversity and Conservation, 19(3):883-899. http://dx.doi.org/10.1007/s10531-009-9746-8 30. Cotton, C. M.; Ethnobotany: principles and applicayions. Wiley, West Sussex, 1996. 31. Cunningham, A. B., Applied ethnobotany:People, Wild Plant Use and Conservation. Earthscan Publications Ltd, London and Sterling, VA, 2001. 32. Dahlgren, J.P. (2010) Alternative regression methods are not considered in Murtaugh (2009) or by ecologists in general.Ecology Letters, 13, E7–E9. 33. Dahlgren, J.P. & Ehrlén, J. (2009) Linking environmental variation to population dynamics of a forest herb. Journal of Ecology, 97, 666–674. 34. Darwin, C., On the Origin of Species. 1st ed. 1964 facsimile edition, Harvard University Press, Cambridge Mass, 1858. 35. Davison, R., Jacquemyn, H., Adriaens, D., Honnay, O., de Kroon, H. & Tuljapurkar, S. (2010) Demographic effects of extreme weather events on a short-lived calcareous grassland species: stochastic life table response experiments.Journal of Ecology, 98, 255–267. 36. Davis, M.B. & Shaw, R.G., (2001) Range shifts and adaptive responses to Quaternary climate change. Science, 292, 673–679. 37. Desktop GARP program parametrs http://www.nhm.ku.edu/desktopgarp 38. Diaz, S., Fargione, J., Chapin III, F. S. & Tilman, D. (2006).Biodiversity loss threatens human well being, PloS Biology 4,1300-1305. 39. Doak, D.F. & Morris, W.F. (2010) Demographic compensation and tipping points in climate induced range shifts. Nature,7318, 959–962. 40. Dullinger, S., Dirnböck, T. & Grabherr, G. (2004) Modelling climate change-driven treeline shifts: relative effects of temperature increase, dispersal and invasibility. Ecology, 92, 241–252. 41. Easterling, M.R., Ellner, S.P. & Dixon, P.M. (2000) Size-specific sensitivity: applying a new structured population model.Ecology, 81, 694–708. 42. Eaton M.D., Soberón J. & Peterson A.T. (2008). Phylogenetic perspective on ecological niche evolution in American blackbirds (Family Icteridae). Biological Journal of the Linnean Society, 94, 869-878. 43. Ellner, S.P. & Rees, M. (2006) Integral projection models for species with complex demography. The American Naturalist,167, 410–428. 44. Elith J, Graham CH, Anderson RP, Dudķk M, Ferrier S, Guisan A, Hijmans RJ, Huettmann F, Leathwick JR, Lehmann AL, Li J, Lohman LG, Loiselle BA, Manion G, Moritz C, Nakamura 118

M, Nakazawa Y, Overton JMcC, Peterson AT, Phillips SJ, Richardson K, Scachetti-Pereira R, Schapire RE, Soberón J, Williams S, Wisz MS, Zimmermann NE, (2006). Novel methods improve prediction of species’ distributions from occurrence data. Ecography 29:129–151. 45. Elith, J. and Leathwick, J.R. (2007) Predicting species' distributions from museum and herbarium records using multiresponse models fitted with multivariate adaptive regression splines. Diversity and Distributions 13, 165-175. 46. ESRI ProgramTools http://www.esri.com/(3.10.2009) 47. ESRI Arc GIS Tools http://www.esri.com/software/arcgis/index.html(17.03.2011) 48. Eugene P. Odum, FUNDAMENTALS OF ECOLOGY, W. B. Saunders, Comp. Philadelphia, London, 1971 ( Third Edition ) pages, 574 49. Fayvush, G., Danielyan T., Nalbandyan A. (2004) Armenia as a producer of medicinal plants: possibilities and perspectives. Available online (accessed 12 April 2010):http://www.natureic.am/NCSA/Publication/Medical_Plants_eng.pdf 50. Ferrier S, Drielsma M, Manion G, Watson G, Extended statistical approaches to modelling spatial pattern in biodiversity in northeast New South Wales. II. Biodivers, 2002, Conserv 11(12):2309–2338. 51. Fielding, A.H. and J.F. Bell (1997). A review of methods for the assessment of prediction errors in conservation presence/absence models. Environmental Conservation 24, 38-49. 52. Ficetola G.F. & Padoa-Schioppa E. (2009). Human activities alter biogeographical patterns of reptiles on Mediterranean islands. Global Ecology and Biogeography, 18, 214-222. 53. Ficetola G.F., Thuiller W. & Miaud C. (2007). Prediction and validation of the potential global distribution of a problematic alien invasive species — the American bullfrog. Diversity and Distributions, 13, 476-485. 54. Fleishman, E., R. Mac Nally, and J.P. Fay. Validation tests of predictive models of butterfly occurrence based on environmental variables. Conservation Biology, 17, 2002, 806-817. 55. Freville, H., Colas, B., Riba, M., Caswell, H., Mignot, A., Imbert, E. & Olivieri, I. (2004) Spatial and temporal demographic variability in the endemic plant species Centaurea corymbosa (Asteraceae). Ecology, 85, 694–703. 56. Gabrielian E, Zohary D, Wild relatives of food crops native to Armenia and Nakhichevan. Flora. Mediter 14:5–80, Palermo 2004, p. 80. 57. Garzón MB, Blazek R, Neteler M, de Dios RS, Ollero HS, (2006) Furlanello C Predicting habitat suitability with machine learning models: the potential area of Pinus sylvestris L. in the Iberian Peninsula. Ecol Model 97:383–393.

119

58. Gotelli, N.J. & Ellison, A.M. (2006) Forecasting extinction risk with nonstationary matrix models. Ecological Applications, 16, 51–61. 59. Guisan, A., O. Broennimann, R. Engler, M. Vust, N.G. Yoccoz, A. Lehman, and N.E. Zimmermann. 2006. Using niche-based models to improve the sampling of rare species. Conservation Biology 20, 501-511. 60. Guisan, A., and W. Thuiller (2005). Predicting species distribution: Offering more than simple habitat models. Ecology Letters 8, 993-1009. 61. Guisan, A., T.C. Edwards Jr., and T. Hastie (2002). Generalized linear and generalized additive models in studies of species distributions: Setting the scene. Ecological Modelling 157, 89-100. 62. Guisan, A. & Zimmermann, N.E. (2000). Predictive habitat distribution models in ecology. Ecol. Model., 135, 147–186. 63. Hampe, A. 2004. Bioclimatic models: what they detect and what they hide. Global Ecology and Biogeography 11, 469-471. 64. Hannah, L., G.F. Midgley, G. Hughes, and B. Bomhard. 2005. The View from the Cape: Extinction Risk, Protected Areas, and Climate Change. BioScience 55, 231-242. 65. Hannah, L., G. F. Midgley, T. Lovejoy, W. J. Bonds, M. Bush, J. C. Lovett, D. Scott, and F. I. Woodward. 2002. Conservation of biodiversity in a changing climate. Conservation Biology 16:264–268. 66. Harmon J.P., Moran N.A. & Ives A.R. (2009). Species response to environmental change: impacts of food web interactions and evolution. Science, 323, 1347-1350. 67. Heikinnen, R.K., M. Luoto, R. Virkkala, R.G. Pearson, and J-H. Körber (2009). Biotic interactions improve prediction of boreal bird distributions at macro-scales. Global Ecology and Biogeography, 16, 754-763. 68. Herborg L.M., Rudnick D.A., Siliang Y., Lodge D.M. & MacIsaac H.J. (2007). Predicting the range of Chinese mitten crabs in Europe. Conservation Biology, 21, 1316-1323. 69. Hijmans, R. J., S.E. Cameron, J.L. Parra, P.G. Jones, and A. Jarvis (2005). Very high resolution interpolated climate surfaces for global land areas. International Journal of Climatology 25, 1965-1978. 70. Hirzel, A.H., J. Hausser, D. Chessel, and N. Perrin. 2002. Ecological-niche factor analysis: How to compute habitat-suitability map without absence data. Ecology 83, 2027-2036.

120

71. Hoegh-Guldberg, O., Hughes, L., McIntyre, S., Lindenmayer, D. B., Parmesan, C., Possingham, H. P., Thomas, C. D., (2008) Assisted colonization and rapid climate change. Science 321:345–346. 72. Houghton, J.T., Ding, Y., Griggs, D.J. , Noguer, M., Linden, P.J. , Dai, X., Maskell, K., & Johnson, C.A., (2001) Climate change 2001: the scientific basis. Contribution of Working Group I to the Third Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Cambridge University Press, Cambridge, UK. 73. Hughes, L; Biological consequences of global warming: is the signal already apparent? Trends in Ecology and Evolution, 15, (2000), 56–61. 74. Hutchinson GE, Concluding remarks. Cold Spring Harb Symp Quant Biol 22:415–442, 1957. 75. Huntley, B., P.M. Berry, W. Cramer, and A.P. Mcdonald. Modelling present and potential future ranges of some European higher plants using climate response surfaces. Journal of Biogeography 22, 967-1001, 1995. 76. Huston, M.A., Introductory essay: critical issues for improving predictions. In: Predicting Species Occurrences: Issues of Accuracy and Scale (eds Scott, J.M., Heglund, P.J., Morrison, M.L.,Haufler, J.B., Raphael, M.G., Wall, W.A. & Samson, F.B.). Island Press, Covelo, CA, 2002, pp. 7–21. 77. Hutchinson GE, Concluding remarks. Cold Spring Harb Symp Quant Biol 22:415–442, 1957. 78. IUCN, WHO, WWF (1993). Guidelines on the Conservation of Medicinal Plants, IUCN, Gland, Switzerland, 50 p. 79. Jongejans, E., de Kroon, H., Tuljapurkar, S. & Shea, K. (2010) Plant populations track rather than buffer climate fluctuations. Ecology Letters, 13, 736–743. 80. Jonzen, N., Pople, T., Knape, J. & Sköld, M. (2010) Stochastic demography and population dynamics in the red kangarooMacropus rufus. Journal of Animal Ecology, 79, 109–116. 81. Kolb, A., Ehrlén, J. & Eriksson, O. (2007) Ecological and evolutionary consequences of spatial and temporal variation in pre-dispersal seed predation. Perspectives in Plant Ecology, Evolution and Systematics, 9, 79–100. 82. Kozak, J.H. and J.J. Wiens. 2006. Does niche conservatism promote speciation? A case study in North American salamanders. Evolution 60, 2604-2621. 83. Klar N., Fernández N., Kramer-Schadt S., Herrmann M., Trinzen M., Büttner I. & Niemitz C. (2008). Habitat selection models for European wildcat conservation. Biological Conservation, 141, 308-319.

121

84. Knouft J.H., Losos J.B., Glor R.E. & Kolbe J.J. (2006). Phylogenetic analysis of the evolution of the niche in lizards of the Anolis sagrei group. Ecology, 87, S29-S38. 85. Lauber, K. & Wagner, G. (1998) Flora Helvetica. 2nd edn. Paul Haupt, Bern, Switzerland. Maron, J.L. & Crone, E. (2006) Herbivory: effects on plant abundance, distribution and population growth. Proceedings of the Royal Society B: Biological Sciences, 273, 2575–2584. 86. Lawler, J. J., D. White, R.P. Neilson, and A.R. Blaustein (2006). Predicting climate-induced range shifts: model differences and model reliability. Global Change Biology 12, 1568-1584. 87. Lawton, J. L. 2000. Concluding remarks: a review of some open questions. Pages 401-424 in M. J. Hutchings, E. John, and A. J. A. Stewart (editors). Ecological Consequences of Heterogeneity. Cambridge University Press. 88. Leathwick, J.R., D. Rowe, J. Richardson, J. Elith and T. Hastie (2005) Using multivariate adaptive regression splines to predict the distribution of New Zealand's freshwater diadromous fish. Freshwater Biology 50, 2034-2052. 89. Leathwick, J.R., Elith, J., and Hastie, T. (2006) Comparative performance of generalized additive models and multivariate adaptive regression splines for statistical modelling of species distributions. Ecological Modelling 199, 188-196. 90. Lehman, A., J.M. Overton, and J.R. Leathwick (2002). GRASP: generalized regression analysis and spatial prediction. Ecological Modelling 157, 189-207. 91. Loiselle, B.A., C.A. Howell, C.H. Graham, J.M. Goerck, T. Brooks, K.G. Smith, and P.H. Williams (2003). Avoiding pitfalls of using species distribution models in conservation planning. Conservation Biology 17, 1591-1600. 92. Longley PA, Goodchild MF, Maguire DJ, Rhind DW, (2005). Geographic information systems and science, 2nd edn. John Wiley & Sons, Chichester, 517 p. 93. MacArthur, R. H. Geographical Ecology: Patterns in the distribution of species. New York: Harper & Row, 1972. 269p. 94. Mander, M., Marketing of Indigenous Medicinal Plants. FAO, Rome, (1998). 95. Maschinski, J., Baggs, J.E., Quintana-Ascencio, P.F. & Menges, E.S. (2006) Using population viability analysis to predict the effects of climate change on the extinction risk of an endangered limestone endemic shrub, Arizona cliffrose.Conservation Biology, 20, 218–228. 96. Massot, M., J. Clobert, and R. Ferrière. (2008). Climate warming, dispersal inhibition and extinction risk. Global Change Biology 14:461–469. 97. Maxted N, van Slageren MW, Rihan JR, Ecogeographic surveys. In: Guarino L, Ramanatha Rao V, Reid R (eds.) Collecting Plant Genetic Diversity Technical guidelines. CAB

122

International, Wallingford International Plant Genetic Resources Institute, Rome, Italy, (1995), Chapter 14, pp 255-287. 98. Millennium Ecosystem Assessment (2005).Ecosystems and Human Well-being: Syntheisis, Island Press, Washington, Dc. 99. Miles L., Grainger A. & Phillips O. (2004). The impact of global climate change on tropical forest biodiversity in Amazonia. Global Ecology and Biogeography, 13, 553-565. 100. Mustin K., Benton T.G., Dytham C. & Travis J.M.J. (2009). The dynamics of climate- induced range shifting; perspectives from simulation modelling. Oikos, 118, 131-137. 101. Myneni, R.B., Keeling, C.D., Tucker, C.J., Asrar, G. & Nemani, R.R. (1997) Increased plant growth in the northern high latitudes from 1981 to 1991. Nature, 386, 698–702. 102. National Report on the State of Plant Genetic Resources in Armenia, Ministry of Agriculture of The Republic of Armenia, Yerevan, (2008), p. 49. 103. National Statistical Institution of Armenia, Yerevan, (2008). 104. Nix, H.A., A biogeographic analysis of Australian elapid snakes. In: Longmore, R., (Ed.). Atlas of Elapid Snakes of Australia. Australian Government Publishing Service, Canberra, 1986. 105. Pachauri, R.K. & Reisinger, A. (2007) Climate Change 2007: Synthesis Report. Contribution of Working Groups I, II and III to the Fourth Assessment Report of the Intergovernmental Panel on Climate Change. IPCC, Geneva, Switzerland. 106. Parmesan, C. (2006) Ecological and evolutionary responses to recent climate change. Annual Review of Ecology, Evolution, and Systematics, 37, 637–669. 107. Payne, K., Stockwell, D.R.B., (1996). GARP modeling system user guide and technical reference. URL: http://biodi.sdsc.edu/Doc/GARP/Manual/manual.html (Last accessed Dec. 14, 2002. 108. Pearce, J., and D.B. Lindenmayer (1998). Bioclimatic analysis to enhance reintroduction biology of the endangered helmeted honeyeater (lichenostomus melanops cassidix) in southeastern Australia. Restoration Ecology 6, 238-243. 109. Pearson, R.G., C.J. Raxworthy, M. Nakamura, and A.T. Peterson. Predicting species' distributions from small numbers of occurrence records: A test case using cryptic geckos in Madagascar. Journal of Biogeography 34, 102-117. 2007. 110. Pearson, R.G., and T.P. Dawson (2003). Predicting the impacts of climate change on the distribution of species: Are bioclimate envelope models useful? Global Ecology and Biogeography 12, 361-371.

123

111. Penso, G. 1980. WHO Inventory of Medicinal Plants Used In Different Countries. Geneva. Switzerland, WHO. 112. Penuelas, J., Gordon, C., Llorens, L., Nielsen, T., Tietema, A., Beier, C., Bruna, P., Emmett, B., Estiarte, M. & Gorissen, A. (2004) Nonintrusive field experiments show different plant responses to warming and drought among sites, seasons, and species in a north–south European gradient. Ecosystems, 7, 598–612. 113. Peterson AT, Papes M, Eaton M Transferability and model evaluation in ecological niche modeling: a comparison of GARP and Maxent. 2007, Ecography 30(4):550–560. 114. Peterson AT, Uses and requirements of ecological niche models and related distributional models.Biodiversity Informatics, 2006, 3:59–72. 115. Peterson A.T. & Kluza D.A. (2003). New distributional modelling approaches for gap analysis. Animal Conservation, 6, 47-54. 116. Peterson A.T. & Robins C.R. (2003). Using ecological-niche modeling to predict barred owl invasions with implications for spotted owl conservation. Conservation Biology, 17, 1161-1165. 117. Peterson A.T. & Cohoon K.P. (1999). Sensitivity of distributional prediction algorithms to geographic data completeness. Ecological Modelling, 117, 159-164. 118. Phillips, S. J., P. Williams, G. Midgley, and A. Archer. Optimizing dispersal corridors for the cape Proteaceae using network flow. 2008, Ecological Applications 185:1200–1211. 119. Phillips SJ, Anderson RP, Schapire RE Maximum entropy modeling of species geographic distributions. Ecol Model 190:231–259, 2006. 120. Pistrick, K. 1987: Untersuchungen zur Systematik der Gattung Raphanus L. — Kulturpflanze 35 225-321.— 2001: Pastinaca (Pp. 1322-1323), Origanum (Pp. 1983-1989), Satureja (Pp. 1997-1999), Melissa (Pp. 1995-1997). — In: Hanelt, P. (ed.), Mansfeld’s encyclopedia of agricultural and horticultural crops, 1st English edition. — Berlin. 121. Pulliam, H. R.. On the relationship between niche and distribution. Ecology Letters, 3:349- 361, 2000. 122. Randin C.F., Dirnböck T., Dullinger S., Zimmermann N.E., Zappa M. & Guisan A. (2006). Are niche-based species distribution models transferable in space? Journal of Biogeography, 33, 1689-1703. 123. Raven, P. H., & Johnson, G. B. (2002). Biology (6th ed.). 124. McGraw-Hill, Raxworthy, C.J., C. Ingram, N. Rabibosa, and R.G. Pearson, (2007). Species delimitation applications for ecological niche modeling: a review and empirical evaluation using Phelsuma day gecko groups from Madagascar. Systematic Biology in press.

124

125. Raxworthy, C.J., E. Martinez-Meyer, N. Horning, R.A. Nussbaum, G.E. Schneider, M.A. Ortega-Huerta, and A.T. Peterson. 2003. Predicting distributions of known and unknown reptile species in Madagascar. Nature 426, 837-841. 126. Rees, M. & Ellner, S.P. (2009) Integral projection models for populations in temporally varying environments. Ecological Monographs, 79, 575–594. 127. Reilly J., Stone P.H., Forest C.E., Webster M.D., Jacoby H.D. & Prinn R.G. (2001). Uncertainty and climate change assessments. Science, 293, 430-433. 128. Robertson, M. P., N. Caithness, and M.H. Villet. (2001). A PCA-based modeling technique for predicting environmental suitability for organisms from presence records. Diversity and Distributions 7, 15-27. 129. Rodríguez J.P., Brotons L., Bustamante J. & Seoane J. (2007). The application of predictive modelling of species distribution to biodiversity conservation. Diversity and Distributions, 13, 243-251. 130. Seneviratne, S.I., Lüthi, D., Litschi, M. & Schär, C. (2006) Land-atmosphere coupling and climate change in Europe.Nature, 443, 205–209. 131. Schleuning, M., Huaman, V. & Matthies, D. (2008) Flooding and canopy dynamics shape the demography of a clonal Amazon understorey herb. Journal of Ecology, 96, 1045–1055. 132. Schweiger O., Settele J., Kudrna O., Klotz S. & Kühn I. (2008). Climate change can cause spatial mismatch of trophically interacting species. Ecology, 89, 3472-3479. 133. Segurado, P. and M.B. Araújo. (2004). An evaluation of methods for modelling species distributions. Journal of Biogeography in press. 134. Skelly D.K., Joseph L.N., Possingham H.P., Freidenburg L.K., Farrugia T.J., Kinnison M.T. & Hendry A.P. (2007). Evolutionary responses to climate change. Conservation Biology, 21, 1353-1355. 135. Small, E., Smartt, J. & Simmonds, N. W. (ed.), (1995). Evolution of crop plants, 2nd edn. Longman, UK. Hemp. Pp. 28-32. 136. Culinary herbs. Ottawa, Ontario & Cronquist, A. (1976). A practical and natural taxonomy for Cannabis. — Taxon 25: 405-435. 137. Soberón J, Peterson AT , Interpretation of models of fundamental ecological niches and species’ distributional areas. Biodiversity Informatics, 2005 https://journals.ku.edu/index.php/jbi/article/view/4

125

138. Solano E. & Feria T.P. (2007). Ecological niche modeling and geographic distribution of the genus Polianthes L. (Agavaceae) in Mexico: using niche modeling to improve assessments of risk status. Biodiversity and Conservation, 16, 1885-1900. 139. Smith, M., Caswell, H. & Mettler-Cherry, P. (2005) Stochastic flood and precipitation regimes and the population dynamics of a threatened floodplain plant. Ecological Applications, 15, 1036–1052. 140. Steiner F.M., Schlick-Steiner B.C., VanDerWal J., Reuther K.D., Christian E., Stauffer C., Suarez A.V., Williams S.E. & Crozier R.H. (2008). Combined modelling of distribution and niche in invasion biology: a case study of two invasive Tetramorium ant species. Diversity and Distributions, 14, 538-545. 141. Stockwell, D.R.B., Peterson, A.T. (2002) Effects of sample size on accuracy of species distribution models. Ecological Modelling 148, 1-13, 2002. 142. Stockwell, D.R.B., and D.P. Peters. 1999. The GARP modelling system: Problems and solutions to automated spatial prediction. International Journal of Geographical Information Systems 13, 143-158. 143. Stockwell D.R.B. & Noble I.R., (1992). Induction of sets of rules from animal distribution data: a robust and informative method of data analysis. Mathematics and Computers in Simulation, 33, 385-390. 144. Swets, J.A. (1988). Measuring the accuracy of diagnostic systems. Science 240, 1285-1293. 145. Stockwell DRB, Noble IR , 1992, Induction of sets of rules from animal distribution data: a robust and informative method of analysis. Math Comput Simul 33:385–390. 146. Sutherland, W.J. , 2006. Predicting the ecological consequences of environmental change: a review of the methods.Ecology, 43, 599–616. 147. Taghtajyan A. L., Flora of Armenia, VIII sector, National Academy of Sciences of Armenia, Botanical Institute, Yerevan, 1987, Pages 418 148. The Primary Directions of Strategy For Reducing Vulnerability of Armenian Agriculture under Global Climate Change, Yerevan, (2008), pages 31. 149. Thomas C.D., Cameron A., Green R.E., Bakkenes M., Beaumont L.J., Collingham Y.C., Erasmus B.F.N., de Siqueira M.F., Grainger A., Hannah L., Hughes L., Huntley B., van Jaarsveld A.S., Midgley G.F., Miles L., Ortega-Huerta M.A., Peterson A.T., Phillips O.L. & Williams S.E. (2004). Extinction risk from climate change. Nature, 427, 145-148.

126

150. Thorn J.S., Nijman V., Smith D. & Nekaris K.A.I., 2009. Ecological niche modelling as a technique for assessing threats and setting conservation priorities for Asian slow lorises (Primates: Nycticebus). Diversity and Distributions, 15, 289-298. 151. Tool Database of Desktop Garp http://ebmtoolsdatabase.org/tool/desktop garp(15.03.2010.) 152. Thuiller, W., D.M. Richardson, P. Pysek, G.F. Midgley, G.O. Hughes, and M. Rouget. 2005. Niche-based modeling as a tool for predicting the global risk of alien plant invasions. Global Change Biology 11, 2234-2250. 153. Thuiller, W., M.B. Araújo, R.G. Pearson, R.J. Whittaker, L. Brotons, and S. Lavorel. 2004. Uncertainty in predictions of extinction risk. Nature 430, 33 (doi: 10.1038/Nature02716). 154. Thuiller W, (2003) BIOMOD—optimizing prediction of species distributions and projecting potential future shifts under global change. Glob Chang Biol 9:1353–1362. 155. Vandermeer, J. H. Niche theory. Annual Review of Ecology and Systematics, (1972) 3:107- 132. 156. Verbyla, D.L., and J.A. Litaitis. (1989). Resampling methods for evaluating classification accuracy of wildlife habitat models. Environmental Management 13, 783-787. 157. Vos, C., P. Berry, P. Opdam, H. Baveco, B. Nijhof, J. O’Hanley, C. Bell, and H. Kuipers, (2008). Adapting landscapes to climate change: examples of climate-proof ecosystem networks and priority adaptation zones. Journal of Applied Ecology 45:1722–1731. 158. Viard-Cretat, F., Gallet, C., Lefebvre, M. & Lavorel, S., (2009). A leachate a day keeps the seedlings away: mowing and the inhibitory effects of Festuca paniculata in subalpine grasslands. Annals of Botany, 103, 1271–1278. 159. Walther, G.R., Post, E., Convey, P., Menze, 1, A., Parmesan, C., Beebee, T.J.C., Fromentin, J.M., Hoegh-Guldberg, O. & Bairlein, 2002 F. Ecological responses to recent climate change. Nature, 416, 389–395. 160. Whittaker, R.J., M.B. Araújo, P. Jepson, R.J. Ladle, J.E.M. Watson, and K.J. Willis. (2005). Conservation biogeography: Assessment and prospect. Diversity and Distributions 11, 3-23. 161. Witten, D.M. & Tibshirani, R., (2009). Covariance-regularized regression and classification for high dimensional problems. Journal of the Royal Statistical Society. Series B, 71, 615–636. 162. Woodward, F. I., and D. J. Beerling. 1997. The dynamics of vegetation change: health warnings for equilibrium 'dodo' models. Global Ecology and Biogeography 6, 413-418. 163. Woodward, F.I. 1987. Climate and plant distribution. Cambridge University Press, Cambridge, 188pages.

127

164. Yesson C, Brewer PW, Sutton T, Caithness N, Pahwa JS, BurgessM, GrayWA,White RJ, Jones AC, Bisby FA, Culham A, (2007). How global is the Global Biodiversity Information Facility. PLoS ONE 2(11):e1124. 165. http://www.biogeomancer.org/understanding.html

128

Daugavpils University Institute of Systematic Biology

ARMINE ABRAHAMYAN

DISTRIBUTION MODELING AND POPULATION ECOLOGY OF WILD MELISSA OFFICINALIS L. (LAMIACEAE) AND ORIGANUM VULGARE L. (LAMIACEAE) IN ARMENIA

DOCTORAL DISSERTATION IN BIOLOGY FOR A SCIENTIFIC DEGREE (BRANCH: ECOLOGY)

APPENDIX

Daugavpils 2015 Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Appendix of Doctoral Dissertation-Daugavpils University:1-22.

1. Theoretical and Experimental Framework of Environmental Variables of Species Distribution Modeling in Armenia

1.1. Physical Geography of The Republic of Armenia

Since early geological history the land surface of Armenia, and the surrounding Armenian plateau, has been mountainous, with further mountain building occurring during the Cenozoic era (particularly after the Miocene). These complex tectonic shifts have resulted in a country dominated by a by a series of mountain massifs and valleys. The tectonic movements which created the series of folded ridges which dominate the country, also resulted in extensive volcanic activity. The climatic changes over the last million years have also left their mark on the country, with evidence of two glacial periods (Riss and Wurm) preserved on almost all mountains over 3000m.

Four main geographic/geological regions can be recognized within Armenia including: 1) Mountainous ridges and valleys in the north-east of the country (highest altitude 3101m), which occur mainly in the basin of the River Kur (including the ranges of Virahajots, Bazumi, Pambak, Gougarats, Aregouni, and Sevan) and which are subject to extensive erosion. 2) Regions of volcanic origin within Asia Minor, including the mountain ranges of Ashotsk, Aragats, Geghama, Vardenis, Sunik and Mount Aragats (4095m). These areas are covered by lava of relatively recent origin (upper Pliocene). Such regions are characterized by gentle slopes, and little evidence of erosion, although larger rivers have carved out deep gorges and canyons. 3) A series of ridged mountains adjacent to the River Arax (ridges on the left bank along with the Urts-Eranossian, Teksar, Vaik, and Zangezour mountain ranges, including the peak of Kapoutdjugh at 3094m) constitute the Minor Caucasian system. This area is prone to intense erosion. 4) The Ararat Valley represents the lowest part of the Ararat depression (which is still undergoing tectonic movement). This area is covered with alluvial and prolluvial sediments.

Being a mountainous country (with altitudes ranging from 375 to 4090 meters above mean sea level) with a characteristic relief straddling two different physical-geographic regions, Armenia has a wide variety of climatic conditions and soils, ultimately manifested in its rich flora and fauna as

Abrahamyan A. (2015): Distribution Modeling and Population Ecology of Wild Melissa officinalis L. (Lamiaceae) and Origanum vulgare L. (Lamiaceae) in Armenia. Appendix of Doctoral Dissertation-Daugavpils University:1-22.

well as a variety of landscapes and types of vegetation. There are six climatic zones from dry subtropical to high mountainous snowcaps and from warm humid to subtropical forests, as well as humid semi-desert steppes. The lowest point being 375m above sea level (near the Debed River in the north of the country) and the highest recorded point being at 4095m (northern peak of Mount Aragats. Overall, the average altitude across the country is 1850m, but the variations in altitude (up to 3700m, but more generally 1500-2000m; have important effects on the climatic and landscape zones within the country (Figure 1, Table 1).

Figure 1. Physical Map of Armenia

Furthermore, the position and gradient of slopes have important implications for 3

the distribution of biodiversity in the country. The steepest slopes found are within mountain folds, but in contrast, over 74% of the land (21, 000 km2) consists of slopes of up to 120, which are generally under cultivation. Among the ridged mountains and valleys of the Minor Caucasus, most forests occur on north-facing slopes.

Table 1. Absolute altitude range of territory of Republic of Armenia

Altitude above sea Area (km2) % level (m)

up to 500 20 0.1

500-1000 2900 9.8

1000-1500 5430 18.3

1500-2000 9300 31.3

2000-2500 7290 24.5

2500-3000 3800 12.6

3000-3500 970 3.3

above 3500 30 0.1

TOTAL 29740 100

1.2. Climate Zones

A great range of climatic zones have been recorded within Armenia. The country is located centrally in the sub-tropical zone, and thus is prone to arid (desert and semi-desert) conditions. However, the altitudinal variation within the country results in further variation in climatic zones, in addition to existing latitudinal clines. The climate in Armenia is mountainous, continental. Since the country is located high above sea level, weather, regardless the season, very often varies even in the nearby regions.

On the Ararat plain and in the basin of Arpa river summers are often hot. Average July temperature is 26 °С, maximum 42 °С. Winters are cold: average January temperature is -4 °С with low

precipitation (350 mm per year). In the low hills summers are warm (average July temperature is 20 °С) and winters are cold (average January temperature is -7 °С) with heavy precipitation. In the middle hills of the central part of the country (altitude of 1500-1800m) summers are also warm. Average July temperature is 18-20 °С, but winters are cold: average January temperature is -10 °С with heavy snowfall.

In the northern and southeastern middle hills the weather is mild and humid. Average January temperature is -4-0 °С, in the mid-summer +18-19 °С. On the southeast and northeast of the country summer is long and very hot. Average summer temperature is about 24 °С, and winters are mild and without snow (average winter temperature is 0 °С).

In the highlands the climate is cold and humid. Average January temperature is -14 °С, average July temperature is not higher than +10 °С. In the winter Armenia receives a lot of snow, which reaches 30-100 sm in the highlands and lies until the mid-spring. Generally, rainfall in Armenia is infrequent. The capital city receives 33 cm of rain annually, though more rainfall occurs in the mountains. The maximum rainy season in Armenia occurs from spring to early summer.

In general, the country receives a high amount of sunshine; ranging from 2600 hours per year (in Yerevan) to 2800 hours per year (shore of Lake Sevan). The average temperature throughout the year varies geographically from 2.70C (Mount Aragats) to 140C (at Meghri). July and August are usually the warmest months (Table 2), while average minimum temperatures recorded in winter vary geographically (from -3.10C at Meghri in the north-east, to -18.90C at Berdashen).

Table 2. Maximum average monthly temperatures in summer and annual rainfall in different altitude zones

Altitude zone Average monthly temperature in summer Annual rainfall (mm)

Low-level 240-260C 250-300

Mid-level 150-200C 400-600

High-level 0 0 700-1000 10 -15 C

Average annual precipitation is around 600 mm, but varies in different altitudinal zones (Table 2). Most precipitation occurs in the spring, while the second half of the summer is dry. Relative humidity averages 60% (ranging from 44% in summer to 80% in winter). Long-lasting snows exist on mountains over 1300m. In these areas snowfall may reach 2m, whilst snowfall reaches 0.5m on the lower steppes.

Figure 2. Monthly average min and max temperature

Comparison of long-term average amount of precipitations and temperature (1961-1990) with average annual precipitation and temperature in 1990-2011:

The increase of annual average temperatures against long-term annual averages has been observed in Armenia (Fig. 3). It was observed that the annual temperatures were lower than the long term temperature in one or two particular years within the whole period.

Figure 3. Increase of annual average temperatures against long-term annual averages 6

Figure 4. Variation of annual precipitation against long-term annual precipitation

It is important to evaluate how climate has varied and changed in the past. The monthly mean historical rainfall and temperature data can be mapped to show the baseline climate and seasonality by month, for specific years, and for rainfall and temperature. The chart above shows mean historical monthly temperature and rainfall for Armenia during the time period 1960-1990. The dataset was produced by the Climatic Research Unit (CRU) of University of East Anglia (UEA, http://sdwebx.worldbank.org/climateportal/index.cfm?page=country_historical_climate&ThisRegio n=Asia&ThisCCode=ARM). Also, the monthly weather datasets has been provided by the Weather Stations of Armenia.

Main landscape zones:

The mountainous nature of Armenia results in a series of highly diverse landscapes, with variations in geological substrate, terrain, climate, soils, and water resources. These landscapes support a great variety of habitats, which support distinctive flora and fauna, and different human use. Seven distinct landscape zones are described in Armenia: deserts, semi-deserts, dry steppes, steppes, woodlands, sub-alpine and alpine lands (Table 3).

Table 3. Areas covered by different landscape zones

Landscape belts Altitude range % of national territory (m above sea level)

Deserts and semi-deserts 700-1300 10

Mountain steppes 1300 37

(wet grassland) (375-700) -

(dry grassland) (1300-1600) -

Forests, trees and scrubland 600-2500 20

Alpine and sub-alpine meadows 2100 28

Deserts and Semi-deserts occur in the Ararat Valley and adjacent mountain slopes at altitudes of 1200-1300m, in the Vaik lowlands, and the Meghri gorge. Sand accumulations in the Arax area result in a desert landscape, which are also found in saline lowlands. In these landscapes climate is dry and continental, with hot summers and moderately cold winters. The soils are generally of the semi-desert grey type, and have been managed for cultivation over the last millennia.

Dry mountainous steppes are found at higher altitudes than semi-deserts (above 1500m) in the Ararat Valley, and some other areas, but are also found at lower altitudes (above 800m) in the lowlands to the north-east of the country, which were originally forested. The climate in the dry steppes is characterized by warm, dry summers and mild winters. A range of soils are found, but in the Ararat Valley these lands are typically stony. Irrigation of dry steppes allows cultivation of crops, vegetables and fruit, and these landscapes have also suffered severe human impact.

Mountain steppes are the dominant landscape for most of the country, particularly at altitudes above 1500m (and at altitudes up to 2000m in the north, 2400-2500m in the south). Meadow steppes occur in the highlands, while patches of forest also occur on ridge tops among steppes in the north-east and Sjunik regions. Climate is generally moderate, with warm, cool summers, and moderate or cold winters. Soils generally have a humus content of between 6-7%. Steppes are used for agriculture (including cultivation of crops, vegetables, frost-tolerant fruit trees (in lower altitudes) and fodder

plants (in highland areas).

Forests generally cover the mid-zone of mountains, occurring at altitudes between 500m and 2100m in the north (up to 2500m in the south). In central Armenia forests occur in small areas rather than as a continuous zone, and forests can be found on steep slopes, and other areas with limited human access. Soil types include red soil in the lowlands and forest gray soils in the highlands.

Sub-alpine meadows occur at higher altitudes than steppes and forests, including highland mountain ranges. Climate is moderate with short, cool summers and long, cold winters. Much of the land here is meadow, with soils of high humus content.

Alpine meadows occupy the highest altitudes above sub-alpine meadows (up to 3000m in the north, 3800m in the south). These meadows represent the principal pasturelands for the country, with meadow and alpine vegetation. Climatic conditions are severe, with long, cold winters, and annual temperatures average less than -40. Snow cover lasts up to 9 months, and permanent snows may occur in some areas.

Azonal landscapes cover over 10% of the territory of the country, and occur independently of altitude (unlike the previously described landscapes). These include wetlands, as well as saline and alkaline lands, which cover about 25,000 ha, including areas in the Ararat Valley where the underground waters are close to the earth surface, resulting in water vaporization and salt precipitation. Upland wetlands are dominated by fresh (non-brackish) water, while lowland wetlands (particularly those around the River Arax) are usually drained in summer, resulting in high salinity.

Distribution of Soil Types in The Republic of Armenia

Figure 5. Soil Types in Armenia

Soil and geology: The soil types and geology of the population were identified using the geographical information system data provided by Environmental Conservation and Research Center of American University of Armenia. The soil forms were confirmed on the ground by digging soil profiles.

2. Historical Distribution of Wild Melissa officinalis L. and Origanum vulgare L. Species in The Republic of Armenia

Due to unsustainable harvesting and destruction of natural habitats medicinal plant’ resources are continually dwindling in Armenia. For sketching the historical distribution of the wild medicinal plants in Armenia we adopted A. L. Takhtajan (1954-2001) of Flora of Armenia in volume 8 (Fig. 5 & 6).

Figure 6. M. officinalis L. Historical Distribution in Armenia.

Figure 7. O. vulgare L. Historical Distribution in Armenia.

Distributional data and taxonomy of wild M. officinalis L. ; O. vulgare L. are primarily obtained from various sources e.g. Floras and monographs, geo-botanical, phytosociological and vegetation studies, herbarium labels of the Department of Plant Taxonomy and Geography, Botanical Institute of the Academy of Sciences of Armenia (http://www.sci.am/).

3. M. officinalis L. and O. vulgare L. Populations’ Size, Density and Grow Intensity in the Different Habitats

Tbale 4. O. vulgare L. Populations’ Sizes, Density and Grow Intensity in Different Habitats

Population Study area, m2 Size, plant Density, Plant height Stems plant/ 1m2 cm quantity/plant Region NSa Tavush Ijevan 206 214 1.0 45 2-4

Lori Novoseltsvo▲ 326 234 0.72 50 3-5

Syunik Chakaten 200 169 0.85 55 2-4

Kapan Meghri 425 206 0.50 65 3-6 New Highway▲

Artsvanik▲ 458 1115 3.0 70 3-5

Vayots Dzor Jermuk 423 1197 3.0 75 3-6

Eghegis ▲ 326 233 0.71 63 3-5

Aragatsotn Aparan 159 128 0.80 35 1-3

Gegharkunik Lichk 156 145 0.92 30 1-3

a- Nearest Settlement ▲ _indecates new population. The plant height and stems quantity were measured at the end of stem forming -phenological phase in respect with assessment of 2% size of each population.

Table 5. M. officinalis L. Populations’ Sizes, Density and Grow Intensity in Different Habitats

Population Study area, m2 Size, Density, Plant height, cm Stems 2 a Plant plant/1m quantity/ Region NS plant

Tavush Getahovit 87 92 1.0 70 1-3 Ayrum▲ 75 89 1.2 62 1-3 Aragatsotn Orgov 136 268 0.95 67 2-3

Kotayk Garni 165 226 0.83 57 1-2

Vajoc Dzor Jermuk▲ 163 491 3.13 124 4-7

Kapan 56 94 2.0 102 2-5 Sjunik Artsvanik▲ 169 528 3.4 121 4-7 Shikahogh 45 89 2.0 81 3-5 Srashen▲ 189 569 3.0 123 4-7 Tsav▲ 162 323 2.0 92 2-4 Karchevank 102 203 2.0 113 3-5

a- Nearest Settlement, ▲ –indecates new population. The plant height and stems quantity were measured at the end of stem forming -phenological phase in respect with assessment of 2% size of each population.

Figure 8. Populations Dynamics of M. officinalis L. in Armenia over the study years

Changes in Origanum vulgare L. Populations’ Sizes

Figure 9. Populations Dynamics of O. vulgare L. in Armenia over the study years

4. Field Trips Framework (2007-2011) in The Republic of Armenia

Figure 10. Field trips in respect with different regions of Armeni.

4.1. Melissa officinalis L. Populations in Different Region of Armenia

Picture 1. Srashen population in Syunik Region, 2007, photo by A. Abrahamyan

Picture 2. Artsvanik population in Syunik Region, 2008, photo by A. Hayrapetyan

Picture 3. Ayrum population in Tavush Region, 2010, photo by A. Abrahamyan

Picture 4. Jermuk population in Vayots Dzor Region, 2008, photo by A. Abrahamyan

Picture 5. Orgov population in Aragatsotn Region, 2009, photo by A. Abrahamyan

Picture 6. Getahovit population in Lori Region, 2010, photo by A. Hayrapetyan

Picture 7. Garni population in Kotayq Region, 2009, photo by H. Abrahamyan

Picture 8. Garni population in Kotayq Region, 2009, photo by A. Abrahamyan

Picture 9. Shikahogh population in Sunik Region, 2011, photo by H. Abrahamyan

Picture 10. Karchevank population in Sunik Region, 2007, photo by A. Abrahamyan

4.2. Origanum vulgare L. Populations in Different Region of Armenia

Picture 11. Jermuk population in Vayots Dzor Region, 2010, photo by A. Hayrapetyan

Picture 12. Artsvanik population in Syunik Region, 2010, photo by A. Hayrapetyan

Picture 13. Aparan population in Aragatsotn Region, 2009, photo by A. Abrahamyan

Picture 14. Origanum vulgare L., Ijevan, Tavush norten Region, 2007, photo by A. Abrahamyan.