Plant Species Diversity and Degree of Homogeneity after the 2010 Eruption of Mount Merapi, Indonesia

The eruption in 2010 of Merapi Mount changed the diversity of plant species. The objective of this research was to investigate the diversity of plant species on Mount Merapi National Park (MMNP). The sampling area was divided into three different levels of damage (heavy, medium, and minor). The research was conducted by using line-transect and quadrat method. The research was conducted by using line-transect and quadrat method. The research showed that Acacia decurrens has the highest Important Value Index (IVI) at the heavy damage area of Cangkringan Resort (CR) and Kemalang Resort (KR) as well as in minor damage area of Selo Resort (SR). The most important species at medium damage area of Dukun Resort (DR) were Albizia lopantha and Pinus merkusii. The pioneer species at Non-Cover Area (NCA) among others were Trema cannabina, Pityrogramma calomelanos, Imperata cylindrical, Buddleja asiatica, and Anaphalis javanica. Eupatorium riparium was the most important species of herb in all four locations based on SIMPER, but it was not dominant in NCA. The homogeneity of plant in Cangkringan Resort and Selo Resort (C < D < E) were normally distributed, whereas NCA, DR, and KR (E ≥ D) were homogeneous. Analysis of non-metric multidimensional scaling and value of ANOSIM (RANOSIM = 0.69; p = 0.0001) demonstrated that all site locations have significance differences. This research was not only focusing on survey of vegetation diversity, but also frequency class distribution. Through this study of frequency of species provides a database for conservation biodiversity and restoration program.


INTRODUCTION
The eruption of Mount Merapi in 2010 resulted in damaged areas with several categories including 1242.16 ha (20.21%) heavy damage, 1207.91 ha (19.66%) medium damage, 2543.94 ha (41.40%) minor damage, and 1151.46 ha (18.74%) lava flow area (MMNP, 2011a). Mount Merapi has dynamic ecology after the 2010 eruption. Invasive species after the 2010 eruption has been reported to be a threat in the habitat (Afrianto et al., 2017;Haryadi et al., 2019;Sutomo, 2019). On the other hand, trees distribution is impacted by average rainfall, elevation, and temperature in the springs area of Soloraya (Siswo et al., 2019). For example, there was an association between water-flow of springs and tree conditions where the spring circumstances declined after the trees degrad-ed. In a social aspect, Mount Merapi is part of belief in traditional knowledge, and it has a significant role in the surrounding community (Hendro, 2018;Wardyaningrum, 2019). In addition, the eruption also showed impacts on agriculture regarding land management (Rahayu et al., 2014;Rindrasih, 2013;Utami et al., 2018).
The succession process changes plant communities in their habitat (Raeval et al., 2012). This research aimed to identify plant species diversity after the 2010 eruption of Mount Merapi. The establishment of a plant diversity database of MMNP is a key factor in evaluating and monitoring the ecological dynamics of the area. It will also provide a systemic identification of the domination of species and the classification of native and invasive species.

Study area
This research was conducted at the area of MMNP, particularly in Yogyakarta Province (Sleman District) and Central Java Province (Magelang, Klaten, and Boyolali Districts) (Figure 1). Placement of sample units representing the types of damage that occurred after the eruption in 2010 where heavy dam-age in Cangkringan Resort (CR), Non-Cover Area (NCA), and Resort Kemalang (KR). Heavy damage area has no vegetation left because it was burnt by volcanic ash. Medium damage in Dukun Resort (DR) that it was Pinus merkusii forest where the canopy and branches of trees were burnt by volcanic ash. The minor damage in Selo Resort (SR), it was almost like an intact forest (Gunawan et al., 2013).

Sampling
Determination of the number of sampling plots was conducted using the species-area curve. Total sample plots in each region were 1.6 ha (CR and NCA), 0.8 ha (KR), 0.4 ha (DR), and 0.4 ha (SR). The determination of the sampling sites was conducted by stratified sampling with a random start method. Analysis of vegetation in all study sites was conducted by using the line transect method in sampling units measuring 20 m × 200 m with perpendicular direction to the contour line. Observations of vegetation were carried out on plots divided into plots measuring 20 m × 20 m (trees), 10 × 10 m (poles), 5 m × 5 m (saplings), and 2 m × 2 m (seedlings, shrubs, and herbs respectively). The observation was also conducted at NCA from CR using a quadrat method in sampling units measuring 100 x 100 m, each consisting of a plot sized of 20 m × 20 m. The total plot was 75 plots from 3 quadrat sampling units (Kusmana, 2017).

Data analysis
The degree of homogeneity of the plant was determined using Raunkiaer Law of Frequency. Plant species were divided into five classes (Table 1) (Khan et al., 2012;Solikin, 2015). The criteria used were: (a) The normal distribution, if A > B > C or C ≤ D < E; (b) homogeneous, if E ≥ D; (c) Heterogeneous, if B, C, D are relatively high; (d) Disturbed, if E < D (Suryawan et al., 2013). The importance value index (IVI) for herbs, shrubs, seedlings, and saplings were calculated by summing relative density and relative frequency. IVI of poles and trees were calculated by summing the relative density, relative frequency, and relative dominance (Roemantyo et al., 2012;Suryawan et al., 2015).
The model of dissimilarity in community composition was analyzed using ordination Non-Metric Multidimensional Scaling (NMDS). Non-parametric one way of analysis of similarity (ANOSIM) was used to test the significant difference. Finally, the analysis of similarity percentage (SIMPER) was calculated to evaluate the contribution of every species to dissimilarity among the group (Sutomo & Faradila, 2013;Chakira et al., 2015). All of those analyses were done using PAST 3.06.

RESULTS AND DISCUSSION
A total of 135 species belonging to 64 families was collected in this study. Among 64 families found, the dominant families were Asteraceae (10 genus, 11 species), Poaceae (9 genus, 10 species), Euphorbiaceae (5 genus, 6 species), Moraceae (2 genus, 6 species), Fabaceae (5 genus, 5 species), and Rubiaceae (5 genus, 5 species). The other plant families were considered non-dominant since the number of species was less than five (Table 2). On the other hand, the species that had not been identified until species level (taxonomically name as "sp."). This shows that future research is needed, especially regarding plant identification until species level. Previous research showed that in Kalikuning, an area affected by the eruption in 2010, the dominant plant family was Rubiaceae (Sutomo and & Fardila, 2013). Marler and del Moral (2011) also revealed that after the eruption at Mount Pinatubo, Luzon, Philippines, that area was dominated by Asteraceae, Fabaceae, and Poaceae.
The degree of homogeneity that was determined based on Raunkiaer Law of frequency ( Figure 2). The result revealed the degree of homogeneity of CR and SR (A > B > C) were normally distributed. Normal distribution means a forest community was in good condition (Suryawan et al., 2013). The frequency class distribution can be used to understand the indicator that controls the absence/presence and concentration of plant species (Danjuma et al., 2017). On the other hand, the degree of homogeneity of NCA, DR, and KR (E ≥ D) was homogeneous. The homogenous species distribution in nature means that uniform species diversity (Vaghasiya et al., 2015).
A. decurrens was one of 29 pioneer trees species after the eruption in 2010 (Gunawan et al., 2015). A. decurrens was dominant for seedling, pole, and tree in the heavy damage area (Figure 3). Even though, A. decurrens in CR was only found up to 1437 m asl after the 2010 eruption (Darmawan, 2015). In SR, which was the minor damage area, A. decurrens was also dominant in sapling, pole, and tree.   (Figure 3). It is because species of S. wallichii have the ability in adaptation from fire disturbance (Kusumastuti, 2017). S. wallichii in minor dam-age area has higher genetic diversity than in the moderate one (Baramantya et al., 2016). In addition, a number of alleles that have a lower frequency and contribution to affect the genetic variation. In nature, S. wallichii grows in altitudes between 750 and 2400 m asl (Khanduri et al., 2013). According to Khanduri et al., (2013), the reproduction of S. wallichii in altitude 750 m and 900 m asl is impacted by the variation in the floral phenology. The most important species in DR were seedling of C. calothyrsus (66.13%), the sapling of P. merkusii (112.70%), pole of A. lopantha (300.00%), and the tree of P. merkusii (300.00%). The secondary succession process in DR was showed by the blossoming of P. merkusii canopy. P. merkusii was dominant because this species was planted by Indonesian state forestry company (Perhutani) for forestry production. MMNP was trying to enrich its plant species diversity by changing a homogeneous to the heterogeneous forest.
The most contributed species to the dissimilarity in all site locations were identified using SIMPER in Table 3. SIMPER result revealed that E. riparium was found in all sampling sites with a cumulative contribution value of 17.97 %. E. riparium is reported as an invasive species (Heystek et al., 2011). This species was dominant in four site locations (CR, SR, DR, and SR). The vegetation survey showed that E. riparium was also dominant in Plawangan (Natalia & Handayani, 2013). Herbs and shrubs were dominant because the area has high light and open canopy (Sutomo & Fardila, 2013). Mount Merapi is an active volcano, and the succession will be challenging to reach a climax stage, particularly in the pyroclastic flow areas. From 2010 to 2015, it was recorded that the status of Mount Merapi became alert status because of the rise of volcanic gas in 2014. Pyroclastic flows have been predicted to occur potential high-risk areas along the Opak and Gendol River (Yulianto, 2014).
Structure of vegetation after the 2010 eruption was affected by the abiotic environment conditions (Afrianto et al., 2016). The trait-based assessment can be used to identify the early risk of invasive plant species in the forest (Junaedi & Mutaqien, 2018). The heavy damage area was dominated by A. decurrens, which was invasive species. The dominance of A. decurrens are correlated with temperature and slope conditions (Afrianto et al., 2017). The distribution of A. decurrens has been reported to be clumped (Untoro et al., 2017). The invasive species can inhibit growth and be competition to native species (Sutomo, 2019). The management of invasive species is an important part of the restoration. Thus, prevention of the widely distribution of invasive species can be done by planting native species, eradication, and biocontrol.  This research was not only focusing on the survey of vegetation diversity, but also frequency class distribution. Through the knowledge that succession is a dynamic process, the present study provides information for the conservation of native plants and restoration programs. The analysis of plant diversity can contribute to a comprehensive evaluation of its relation to the vegetation vulnerability to the volcano eruption. The other benefits from the assessment of plant species diversity are for environmental protection and sustainable management of natural resources in the area.

CONCLUSION
. The diversity of species in all sampling locations in Mount Merapi National Park was difference after the 2010 eruption. The range of homogeneity of vegetation in the Mount Merapi National Park consisted of normally distributed (CR and SR) and homogeneous (NCA, KR, and DR) vegetation.