Spatial distribution and association patterns of Pinus brutia Ten. regeneration

Schlüsselbegriffe: O-Ring Statistik, Elternbäume, Naturverjüngung, Unterschicht

Download as PDF

Abstract

Pinus brutia Ten. forms a distinctive natural coniferous forest in Kurdistan region in Iraq. The aim of this study was identifying the spatial distribution and association patterns of P. brutia regeneration. On September 2016, 10 study plots with a total area of 0.05 ha of each plot were set up along a 2 km transect in the northern aspect of Zawita natural forest and the positions of P. brutia parent trees, regeneration and shrubs were recorded. We used O-ring statistics to recognize the spatial distribution and association patterns of P. brutia. Spatially, seedlings and saplings of P. brutia were distributed in an aggregation pattern in comparison to the randomized pattern of the parent trees. A pattern of repulsion was found between seedlings and saplings with parent trees. The number of seedlings decreased with the increasing stand basal area, but in a positive correlation with the presence of shrubs and with the increasing shrub crown area, particularly with Quercus shrubs. Both patterns of early regeneration of P. brutia, spatial aggregation as well as repulsion from parent trees mainly derived from the competition with parent trees. 

Zusammenfassung

Pinus brutia Ten. bildet einen ausgeprägten natürlichen Nadelwald in der Kurdistan-Region im Irak. Das Ziel dieser Studie war die Ermittlung der räumlichen Verteilung und der Assoziationsmuster von P. Brutia Verjüngung. Im September 2016 wurden 10 Probeflächen mit einer Fläche von je 0.05 ha entlang eines 2 km langen Transsekts im nord-exponierten Teil des Zawita-Waldes festgelegt und die Positionen von P. brutia-Elternbäumen, der aufkommenden Verjüngung und von Sträucher aufgezeichnet. Die räumlichen Verteilung und Assoziationsmuster von P. brutia wurden anhand von O-Ring-Statistiken untersucht. Während die Positionen der Elternbäume eher zufällig sind, zeigten die Keimlinge und Jungpflanzen ein deutliches Aggregationsmuster. Wir konnten eine Abstoßungsreaktion zwischen Keimlingen und Jungpflanzen und den Elternbäumen feststellen. Die Anzahl der Keimlinge sank zwar mit steigender Baumgrundfläche, stand aber in positiver Korrelation mit dem Vorhandensein und der Kronenfläche von Sträuchern, insbesondere mit Quercus-Sträuchern. Diese Muster der frühen Verjüngung von P. brutia – die räumliche Aggregation sowie die Abstoßung (Repulsion) von Elternbäumen – entstehen hauptsächlich durch die Konkurrenz mit Elternbäumen.

1. Introduction

One main aim of forest ecosystem studies is to understand the processes and mechanisms that drive the regeneration of tree species (Jia et al., 2016). Patterns in the distribution of regeneration are essential for establishment of regeneration. The spatial pattern of tree regeneration has been increasingly recognized (e.g., Muhamed et al., 2015; Ghalandarayeshi et al., 2017) and investigation of the effect of spatial interactions in population and community dynamics is a focus of current forestry research (McIntire & Fajardo, 2009). Advanced ecology is driven by an interest in processes that are inherently spatial, as in the case of plants. Under the assumption that processes generate patterns, intra- and inter-specific interactions between species can produce spatially recognizable patterns (Rayburn & Monaco, 2011; Kang et al., 2017). Spatial distribution patterns become apparent in the aggregation of plants. That means that plants can be aggregated (clustered) or positioned in the opposite formation, which is a systematic (regular) order. If neither of those applies, they are in random arrangement (Clark and Evans, 1954). The aggregation only relates to the spatial location, regardless of the involved species. When considering species the terms association versus segregation of plants can be used. Association means that it is more likely that the nearest neighbor of a plant is an individual of a different species than of the same species. In contrast, segregation means that different species grow in homogeneous groups (Pielou 1961).

The main aggregation patterns observed at small spatial scales have been referred to as either positive (facilitation) or negative (competition) plant-plant interactions. For example, nearly regular plant spatial patterns are usually a sign of intense competition between plants for limited resources like water and light (Stoll & Bergius, 2005; Rayburn & Monaco, 2011). While, aggregated patterns (especially interspecific) may be an indication of positive plant interactions at small scales, whereby the presence of one species alleviates a severe abiotic environment for other species (Gómez -Aparicio et al., 2005; Kefi et al., 2007). If the processes that cause segregation or aggregation are absent, or present but in balance, then random patterns are generated (Szwagrzyk & Czerwczak, 1993).The aggregated distribution of species is a common pattern, particularly for naturally regenerated seedlings in forests (Zhang et al., 2013). The location of parent trees and the modes of seed dispersal usually determine the spatial patterns of seed rain (Hubbell, 2001). Seedling establishment is then influenced by successive post-dispersal biotic and abiotic factors (Gholami et al., 2018). Niche segregation (Pielou, 1961), habitat heterogeneity (Gholami et al., 2018) and plant-plant interactions are considered to be the mechanisms that contribute to aggregation patterns of regeneration. The patterns resulting from spatial distributions and associations may vary among species, e.g., species-specific, and over a time gradient, e.g., the developmental stage, and when species grow in heterogeneous habitats (Zhang et al., 2013; Muhamed et al., 2015; Wiegand et al., 2017).

In water-limited ecosystems typical found under dry continental Mediterranean growing conditions, natural forest regeneration is severely limited at all stages, with high mortality rates among both natural and planted seedlings (Gómez -Aparicio et al., 2005; Padilla & Pugnaire, 2006). Under such environmental stress, parent trees and shrubs have been recognized as important niches for tree regeneration (Callaway, 1997; Rodríguez-García et al., 2011b). Among the facilitative effect of the presence of an overstory canopy for seedlings, the most positive effects are interspecific while the negative effects are intraspecific e.g., competition between the need for water or nutrients and light by parent trees versus offspring regeneration. Studies in Mediterranean ecosystems have investigated the influences of canopy cover on the natural regeneration mechanisms of many pine species like Pinus sylvestris L. (Barbeito et al., 2009), Pinus pinaster Ait. (Rodríguez-García et al., 2011a), Pinus halepensis Mill. (Prévosto et al., 2012), Pinus pinea L. (Adili et al., 2013), and more recently Pinus nigra (Calama et al., 2017) and Pinus brutia Ten. (Muhamed et al., 2018). The aforementioned studies indicated increasing seedling mortality in summer and in open areas and provide evidence for positive correlation with shrubs, while parent tree – regeneration association patterns varied between positive under partial canopy cover and moderate shade to clear negative association patterns. In the Kurdistan region, Zawita pine (Pinus brutia Ten.) is distributed in small enclaves and its many physiological and morphological features mean it is well-adapted to the Mediterranean-type climate (Thanos & Scordilis, 1987). In Iraq, it is one of the most widely used species for planting in dry and semi-dry lands due to its limited ecological requirements and high tolerance to drought and heat. The light requirements of its seedlings are high. According to Özdemir (1977) this species regenerates poorly under low light intensity, i.e. below 65–70 %. It occupies open disturbed sites, generally with a sparse vegetation cover, and particularly in the absence of aggressive competitors. Therefore, the heliophilous nature of the species exploits open habitats produced by fire through massive recolonization (Thanos & Doussi, 2000; Boydak, 2004; Muhamed et al., 2018). Detecting forest spatial patterns at each of the development phases may provide information on changes in plant-plant interactions from one development stage to the other, thereby uncovering the processes underlying forest future regeneration dynamics (Kazempour et al., 2018). One way to explore the spatial dynamics of forest regeneration without long-term demographic censuses is to perform a point pattern analysis based on analysing fully mapped tree positions using, for example, the centroid positions of trees or shrub stems or shrub crowns in relation to other forest understory species (Getzin & Wiegand, 2007; Muhamed et al., 2015) In forestry, the impetus for using point pattern analysis to detect tree spatial patterns is reasonable because, compared to other plant communities like diffuse growing shrubs and grass, forests comprise a wide point pattern that in many cases can be approximated by dots (Wiegand & Moloney, 2004).

To our knowledge, little is known about the regeneration dynamics of P. brutia, in Iraq, in particular on the role of parent trees and shrubs in the regeneration process. In this study, we intend to answer the following questions: What are the prevailing types of spatial distribution and association patterns in three life stages of trees in Zawita forest? Is it possible to detect mechanisms that help shape the outcome patterns of this species? The findings from this study will help understand the ecological requirements of P. brutia regeneration over life stage gradient and support the theoretical basis of P. brutia regeneration and community succession.

2. Materials and methods

2.1. Study site

The study was conducted in the northern aspect of the Zawita–Duhok province, Kurdistan region-Iraq, (N 36° 53’ 48” E 43° 08’ 48’’) at elevations ranging from 900 m up to 1600 m.a.s.l. (Figure 1). Average total rainfall is 578.5 mm and the mean annual temperature is 15 °C. Monthly average relative humidity (RH %) ranges from 18.5 % to 68% with a total average of 44.18% (Directorate of Meteorology 2014). Soils are acidic (mean pH 5.0) and poor in nutrients, especially in N, P, K+, Na+, Ca2+, Mg2+, Mn2+, Fe2+, Zn2+, and Cu2+ compared to the dominant oak forest in the area. The study site contains a ridge that is 70 % covered by Pinus brutia with some associated species, particularly sclerophyllous ones like Juniperus oxycedrus, Paliurus spina-christi, Quercus spp. and Pistacia eurycarpa, Pistacia khinjuk, Crataegus monogyna, Prunus microcarpa, Anagyris foetida, Rhus coriaria. While a comprehensive fauna survey is not conducted yet in Zawita, there is evidence for a diverse population of mammals, reptiles and birds. Zawita is famous for its pine forests commonly found in fire-related ecosystems of the eastern Mediterranean region. Zawita is considered by many authors to be the southern limit of the distribution of this species in the eastern Mediterranean (Biger & Liphschitz, 1991). Today, the pine forests of Zawita are showing signs of deterioration, as, in addition to forest fires, human recreational activities and urbanization are major threats to P. brutia forests, which could mean that natural regeneration is not sufficient to replace mature and over-mature trees.

2.2. Species 

P. brutia is restricted to a small north-eastern corner of the Mediterranean region. It grows from Greece to Iraq, but is mainly concentrated in Turkey and Cyprus. Dar Kaja Zawita or Sanoubar Zawita is local common names for P. brutia in Iraq. Is a medium size tree, reaching 20–35 m in height and with a trunk diameter of up to 1 m. The species naturally cover about 10,000 hectares distributing in Zawita and Atrush-Duhok districts (Thanos & Scordilis, 1987). P. brutia cannot resprout and regenerates from seed. It is light-demanding, grows rapidly in its early establishment phases and is a site-insensitive species that can adapt to most soil types as well as being tolerant to drought (Urgenc, 1971; Quézel, 2000; Boydak, 2004). In Iraq, P. brutia has been widely planted as a landscape tree in urban areas and for the revegetation of bare watersheds. Silvicultural experience obtained from local afforestation practices suggests that to achieve reasonable growth in arid and semi-arid zones, up to three years of irrigation are recommended after planting, particularly during summer.

2.3. Sampling and data analysis

2.3.1. Sampling methods

In the middle of September 2016, 10 rectangular plots with size 25 m by 20 m were allocated along a 2 km transect. We assumed this plot size is sufficient to ensure homogeneity in site conditions within the plot and a sufficient number of regeneration, parent trees and shrubs. For the stem census and to facilitate data collection, each plot was further divided into twenty 5 by 5m quadrates. In each of the 200 quadrate plots, all living seedlings (h ≤ 50 cm), saplings (h > 50 cm and diameter at breast height DBH < 4 cm), trees (DBH > 4 cm), and shrubs (crown diameter > 30 cm) were mapped to reference axes with origins (x, y = 0, 0) located in the lower right corner of the plot using a measuring tape (±5 cm) in a Cartesian coordinate system (Muhamed et al., 2015) (Figures 2a, 2b and Figure A2 to A8 in Appendix). The long diameter width and short diameter length were used to quantify the crown area (CA) of each pine tree and shrub within an elliptical crown shape. The stand basal area BA in m² of each plot was calculated according to Eq. 1, where d represents the tree diameter at breast height.

2.3.2. Data analysis

The univariate O-ring statistic was used to investigate the pure spatial distribution (aggregated, regular and random) regardless of the species, in two categories: parent trees among each other, regenerated trees (seedlings, saplings) among each other. The bivariate O-ring statistic was used to detect the association patterns of parent trees with seedlings and saplings. The shrub-seedlings/saplings were spatially also detected once with all shrub species combined and once with specified shrub species (Figure A1). As defined by Wiegand & Moloney (2004), the bivariate O-ring statistic O12 (r) (Eq. 2) gives the expected number of points in a pattern 1 (e.g., seedlings) at distance r from an arbitrary point of pattern 2 (e.g., trees) under the assumption of the antecedent condition null model (Wiegand & Moloney, 2004). This model randomizes the locations of the pine regenerations assuming that they could be any-

where in the plot and keeps the locations of the adult trees and shrubs fixed assuming, first, that their location did not change during pine regeneration establishment and second, that larger trees influence the distribution of the small pine regenerated individuals but not vice versa.

n1 is the number of points in pattern 1, 

is a ring with width W at distance r from an arbitrary point of pattern 1,

is the number of points in pattern 2 in

is the area of 

is the ring with radius r and W width centred at the ith point of pattern 1, n1 is the total number of points of pattern 1 in the study region 

Since we are detecting the distribution patterns of the same species cohort e.g, seedling, the univariate O-ring statistic O11 (r) is estimated by setting pattern 2 equal to pattern 1 under the assumption of complete spatial randomness (CSR) where the points are independently and randomly distributed over the entire plot (Wiegand & Moloney, 2004). In this study, we defined three fine scale spatial patterns. For the spatial distribution of parent trees and regeneration trees 10 m was selected, for the association of type shrub-regeneration 2 m was selected, for the association of type parent tree-regeneration 10 m was selected with 0.25 m intervals. All point pattern analyses were performed with the grid-based software Programita (Wiegand & Moloney, 2004, version updated Feb. 2014 from http://programita.org/., settings: ring width = 0.25, cell size = 1 m). A goodness-of-fit (GoF) test was used to avoid the risk of inflating type I errors that may occur when the value of O (r) is close to a simulation envelope (Grabarnik et al., 2011). Reducing the plant size to dot locations is one of the limitations of point pattern analysis, particularly when the plant is bigger than the spatial scales investigated. This may mask real spatial relationships at smaller scales that represent the scales that interest most ecologists who study plant-plant interaction (e.g., Wiegand et al., 2006; Muhamed et al., 2015). Therefore, to confirm our spatial patterns, a second-order polynomial models (Microsoft® Office Excel 2007) was performed to describe the effect of tree and shrub density parameters (stand basal area, the tree crown area, number of trees, shrub crown area and the number of shrubs) on total seedling and sapling density. One way ANOVA was used to detect the significance of the correlation patterns.

3. Results

3.1 Univariate spatial patterns of Pinus brutia regeneration

A total of 1024 P. brutia seedlings and 1039 saplings were recorded and mapped, with an average of 102 seedlings and 103 saplings per plot (Table A1). According to the univariate O11 (r) (Figures 3a, b, c, d and Table A2), the distribution pattern of seedlings and saplings of P. brutia in the study plots was similar. Based on the GoF tests, the regeneration displayed significant deviations from the complete spatial randomness null model. Overall, parent trees were randomly distributed up to 10 m (Table A3) while seedlings and saplings were spatially distributed in aggregation patterns within on average 2.85 m and up to m 3.8 respectively, which confirmed the visual inspection of the mapped patterns (Figures 2a, b, A2 to A8). The values of the univariate O11 (r) in this study simply mean that the observed density of P. brutia seedlings and saplings inside the rings with a radius of 0.25 m at the detected scales of distance was significantly greater than the density in the study plot.

Figure 3 a, b, c, d: Examples of univariate results of the O-ring statistic for seedlings and saplings of P. brutia in two sample plots. The solid lines represent the O-ring values according to the complete spatial randomness null model, dashed lines the 95% Monte Carlo confidence intervals of the null hypothesis of complete spatial randomness. O-ring value that fall above, below and within confidence intervals indicate aggregation, regular and random spatial patterns, respectively.

Abbildung 3 a, b, c, d: Beispiele für univariate Ergebnisse der O-Ring-Statistik für Keimlinge und Jungpflanzen der P. brutia in zwei Probeflächen. Die durchgehende Linien zeigen die O-Ring-Werte gemäß dem Nullmodell der zufälligen räumlichen Verteilung und die gestrichelte Linien das 95% Monte-Carlo-Konfidenzintervallen der Nullhypothese der zufälligen räumlichen Verteilung. O-Ring-Werte, die über, unter oder innerhalb der Konfidenzinterfalle liegen, weisen respektive auf Aggregation, reguläre und zufällige Raummuster hin.

3.2. Bivariate spatial patterns of Pinus brutia regeneration

Analysis 1: Intraspecific spatial association

From the total of 19 bivariate parent tree – regeneration (seedlings - saplings) pair tests, the GoF tests of O12 (r) function detected significant repulsion patterns in 7 parent-seedling cases out of a potential 10, and 5 arent-sapling cases out of a possible 9, as shown in Table 1 and Figures 4a, b, c, and d. Generally, the patterns of parent tree – regeneration confirmed the existence of an intraspecific segregation pattern in the 10 study plots, however, a few cases of association were found at larger scales, in plots 4, 5, 6 (Table 1).

Figure 4 a, b, c, d: Examples of bivariate results of the O-ring statistic for seedlings and saplings of P. brutia Intraspecific spatial associations. For details please see Fig. 3.

Abbildung 4 a, b, c, d: Beispiele bivariater Ergebnisse der O-Ring-Statistik zur intraspezifischen räumlichen Assoziation von P. Brutia Keimlingen und Jungpflanzen. Für Details zur Darstellung verweisen wir auf Abb. 3.

Analysis 2: Interspecific spatial associations

The overall analysis of all 468 shrubs (ignoring their species identity) with the P. brutia seedlings and saplings showed that seedlings were more spatially attracted to shrubs (Table 2). GoF tests of the bivariate interspecific shrub-seedling showed 7 cases of association, 1 of segregation and 2 independent cases out of 10 cases. The distribution of shrub- sapling spatial associations was more random (7 cases out of 9) with 1 case of association and 1 of segregation (Table 2). In assessing the role of the different shrub species in the spatial association of the natural regeneration of P. brutia, we only accounted for the results of the analysis of Quercus species and Prunus macrocarpa, since they were only the shrub species that frequently replicated in the majority of study plots. The results of the GoF tests of shrub-regeneration spatial pattern revealed significant patterns of association between Quercus species and Pinus regeneration in 7, 4 cases and 3,4 independent patterns and 0,1 segregation pattern out of 10, 9 patterns for seedlings and saplings tested respectively (Table A4). In short, the bivariate analysis under the null model of antecedent conditions (association/segregation) between parent trees on the one hand and shrubs on the other hand with seedlings and saplings of P. brutia, revealed significant negative and positive spatial associations respectively. The seedlings in the 10 plots displayed an attraction to the pooled shrubs, while for the relation between saplings and shrubs no such attraction could be detected (Figure 5).

3.3. Over- and mid-story density correlation with Pinus brutia regeneration 

Even though we mapped all shrub species, for the analysis, we only used those whose short diameter length of the crown was at least 30 cm, expecting that when crowns were smaller than that, the shade provided by the shrubs would not be sufficient to influence the regeneration process. Generally, the presence of the parent trees adversely affected P. brutia seedlings. According to the results of polynomial fitting in Figure 6 and Table 3, stand basal area and the overstory canopy area exhibit a significant negatively correlation with regeneration of P. brutia seedlings (R² = 0.65, P ≤ 0.006 and R2 = 0.75, P ≤ 0.02, respectively). In addition, a negative trend with the number of adult trees is evident in Table 3, indicating that regeneration of seedlings may decrease along a gradient of increasing tree density in our study plots. In contrast, the shrub density represented either by midstory canopy area or by the number of shrubs was significantly in positive correlation with P. brutia seedling density (Table 3). The density of Pinus seedlings increased with an increase in the number of shrubs and increase in the shrub crown area in the canopy (R² = 0.72, P ≤ 0.008 and R² = 0.55, P ≤ 0.047) respectively. In turn, no correlation was found between naturally established young saplings and the parent trees and shrub density parameters as shown respectively by the signs of the regression parameter estimates (Table 3).

4. Discussion

4.1. Univariate spatial distribution

Based on the Goodness-of-fit tests, compared to the obviously random distribution of parent trees, aggregated distribution was the dominant pattern approximately over all the study plots for both seedlings on average up to fine scale 2.85 m and saplings at larger scale 3.8 m, whose intensity faded gradually with increasing distance to up to 5 m. Aggregated distribution of species is a common pattern, particularly for naturally regenerated seedlings in forests (Zhang et al., 2013). Niche segregation (Pielou, 1961), habitat heterogeneity (Gholami et al., 2018), and dispersal mode (Hubbell, 2001) are considered to be the mechanisms that contribute to aggregation patterns of regeneration. In this study, the tendency of P. brutia to show clump distribution generally is due to the tendency of the species to regenerate in clusters in gaps avoiding the competition with parent trees for light and/or litter accumulation. However, under our study conditions, limited seed dispersal or/and biological characteristics of seeds of P. brutia might also explain a part of this pattern at small scales. For example, cones of coniferous trees are big and heavy, and the seeds are slightly heavier relative to their wing length (Richardson et al., 2000) leading to the shorter average dispersal distance, and then the seedlings that emerge from these seeds inevitably form aggregated distributions after germination (Seidler & Plotkin, 2006; Lin et al., 2011; Zhang, 2017). Previous studies suggested that intraspecific competition increases over tree life stages, causing changes in the aggregation distribution toward a predominant random pattern among saplings and to fit a regular pattern of the parent trees. This is because competition for resources over time will lead to mortality among individuals within the clumps, which will gradually alter the spatial distribution (Lan et al., 2012). This did not occur on the sites under examination of the present study, maybe because the saplings and the trees were not big enough to show strong competition among conspecific neighbors. In other words, P. brutia saplings arranged in a clumped pattern could mean that the saplings did not exhibit strong self-thinning yet, which can shift the pattern from aggregation to a random or segregation shape (Getzin et al., 2006; Yuan et al., 2018).

4.2. Bivariate spatial associations

In many cases, in severe environmental conditions, the favourable growing conditions created under the canopy of some plants appear to be the crucial factor that facilitates natural forest regeneration (Gómez-Aparicio et al., 2005; Rodríguez-García et al., 2011a; Rodríguez-García et al., 2011b; Muhamed et al., 2015). In Mediterranean ecosystems, extreme temperatures, summer drought, and irradiance are important factors affecting plant recruitment (Gómez-Aparicio et al., 2004; Puerta-Pinero et al., 2007). In this context, our expectation in this study was to find a strong spatial association pattern of Pinus regeneration with parent trees and neighboring shrubs. On the contrary, in almost every case of paired analysis, the P. brutia seedlings and saplings were found in distance from parent P. brutia trees, but in the vicinity of shrubs, particularly seedling-Quercus shrubs. These findings were confirmed by the negative correlation of both seedling and saplings with the tested density parameters of parent trees in contrast to the positive correlation with the number of shrubs and shrub crown area. This means that the microhabitat near the parent trees inhibited seedling establishment. The segregation patterns between parent trees and seedlings and between parent trees and saplings underline the significance of competition in structuring the distribution patterns of early regenerated Pinus individuals. These results are inconsistent with the heliophilous nature of P. brutia that germinates and establishes in well-illuminated places, as a pioneer and site insensitive species (Thanos & Doussi, 2000). Numerous authors have concluded that a segregation exists between the density of mature pine trees and seedlings. Palik et al. (1997) showed a negative correlation between the early seedling establishment of longleaf pine and the basal area of mature trees as a result of decreasing available light and nitrogen levels with increasing basal area. Grace and Platt (1995a, b) also detected a negative correlation between the location of seedlings with mature trees during the first year of seedling growth as a result of intraspecific competition between seedlings and the fine root biomass of mature trees. The segregation between seedlings and adult trees may also be due to increased needle litter depth around mature trees because the accumulation of litter reduces the emergence of seedlings of small-seeded tree species and prevents the radicals from reaching the soil (Grace & Platt, 1995a; Dzwonko & Gawronski, 2002; Avery, 2004). In plant-plant interactions, competition and facilitation may occur simultaneously and the net outcome of the interaction may vary both spatially and temporally (Gómez -Aparicio et al., 2005; Muhamed et al., 2015). The high number of repulsion patterns with parent trees found in this study do not exclude potential positive effects of parent trees like facilitating and sheltering the early regeneration of Pinus from direct radiation, but may mean that these positive aspects were outweighed by negative influences like light blocking, allelopathic, or thick litter layer under continuous canopy cover (Padilla & Pugnaire, 2006; Muhamed et al., 2018). Consequently, this may have promoted regeneration under the more favorable ecological conditions provided by shrubs compared to the conditions provided by parent trees. However, the effect of the depth of needle litter and light intensity was not within the scope of our study. It is likely that the main reasons behind the repulsion patterns are both the amount of litterfall that accumulates around the parent trees and in their understory (Muhamed et al., 2018) and the reduced availability of light that reaches the forest floor (Puerta-Piñero et al., 2007). This result is congruent with the Janzen-Connell theory explaining high offspring mortality rates near parent trees, which might also partly contribute in explaining the segregation patterns detected in this study (Matthesius et al., 2011). The spatial association pattern between shrubs and seedlings is comprehensible since shrubs like Quercus species with a relatively small canopy can represent a beneficial shady environment (e.g., by reducing the soil temperature and atmospheric moisture stress) and at the same time, provide sufficient light during the early establishment stages (Muhamed et al., 2018). Such results are common in Mediterranean ecosystems where pioneer shrubs can facilitate tree seedling survival and growth performance during natural establishment (Gómez-Aparicio et al., 2004). The non-significant correlation between saplings and shrubs could be due to the relatively small size of shrubs in this study compared to the well-developed root and shoot system of the saplings that help saplings continue to grow without the need for continued facilitation by shrubs. However, the shrub-seedling spatial association or the neutral shrub-sapling association may switch to segregation spatial pattern (competition) over life stage gradient, since niche requirements for the regeneration usually narrow at the seedling stage and become broader at old sapling and adult stages that may generate intensive competition with shrubs for resources over time (Quero et al., 2008). 

5. Conclusion

A pattern of aggregated spatial distribution discovered in the initial stages of P. brutia regeneration in this study is an evidence for a weak (or absent) self-thinning process in the early stages of regeneration. Zawita pine regeneration could be facilitated by shrubs while the microhabitat close to the parent trees prevented regeneration. In this context, preserving shrub cover by forest managers, particularly of Fagaceae, could enhance pine recruitment in this forest. However, further research like planting seedlings under and offside the canopy of shrubs and parent trees as well as seed dispersion study should be encouraged in order to explore the effects of shrub management on the natural regeneration of P. brutia and allow to disentangle the respective roles of competition and limited pine seed dispersal on the natural regeneration patterns of P. brutia.

Acknowledgments

This study was supported by funds from the University of Duhok- Agriculture College. We thank Duhok Directorate of Forestry and Rangeland for the permission to work and access to the study site.

References

Adili, B., El Aouni, M., Balandier, P. 2013. Unravelling the influence of light, litter and understorey vegetation on Pinus pinea natural regeneration. Forestry, 86(3): 297–304.

Avery, CR., Cohen, S., Parker, KC., Kush, JS. 2004. Spatial pattern of longleaf pine (Pinus palustris) seedling establishment on the croatan national forest. North Carolina Academy of Science, 120(4): 131–142.

Barbeito, I., Fortin, MJ., Montes, F., Canellas, I. 2009. Response of pine natural regeneration to small-scale variation on a managed Mediterranean mountain forest. Applied Vegetation Science, 12(4): 488–503.

Biger, G., Liphschitz, N. 1991. The recent distribution of Pinus brutia: a reassessment based on dendroarchaeological and dendrohistorical evidence from Israel. Holocene, 1(2): 157-161. 10.1177/095968369100100208.

Boydak, M. 2004. Silvicultural Characteristics and Natural Regeneration of Pinus brutia Ten.: Plant Ecology, 171(1/2): 153-163.

Calama, R., Manso, R., Lucas-Borja, M E et al. 2017. Natural regeneration in Iberian pines: A review of dynamic processes and proposals for management. Forestry System, 26(2): 20 pages, eR02S. https://doi.org/10.5424/fs/2017262-11255.

Callaway, RM. 1997. Positive interactions in plant communities and the individualistic continuum concept. Oecologia, 112(2): 143-149.

Clark, P.J., Evans, F.C. 1954. Distance to nearest neighbor as a measure of spatial relationships in populations, Ecology, 35, 445–453.

Dzwonko, Z., Gawronski, S. 2002. Effect of litter removal on species richness and acidification of a mixed oak-pine woodland. Biological Conservation, 106(3).:389–398.

Getzin, S., Dean, C., He, F., Trofymow, J. A., Wiegand, K., Wiegand, T. 2006. Spatial patterns and competition of tree species in a Douglas-fir chronosequence on Vancouver Island. Ecography, 29(5): 671–682. https:// doi.org/10.1111/j.2006.0906-7590.04675.x.

Getzin, S., Wiegand, K. 2007. Asymmetric tree growth at the stand level: Random crown patterns and the response to slope. Forest Ecology Management, 2429 (2/3): 165-174.

Ghalandarayeshi, S., Nord-Larsen, T., Johannsen, V. K., Larsen, J. B. 2017. Spatial patterns of tree species in Suserup Skov – a semi-natural forest in Denmark. Forest Ecology Management, 406(15): 391-401. doi: 10.1016/j.foreco.2017.10.020.

Gholami, S., Saadat, L., Sayad, E. 2018. Different microhabitats have contrasting effects on the spatial distribution of tree regeneration density and diversity. Journal of Arid Environments, 148:1-5. doi.org/10.1016/j.jaridenv.2017.10.008.

Gómez –Aparicio, L., Gómez, J.M., Zamora, R., Boettinger, J.L .2005. Canopy vs. soil effects of shrubs facilitating tree seedlings in Mediterranean montane ecosystems. Journal of Vegetation Science, 16: 191-198.

Gómez-Aparicio, L., Zamora, R., Gómez, J. M., Hódar, J.A., Castro, J., Baraza, E. 2004. Applying plant facilitation to forest restoration: A Meta‐analysis of the use of shrubs as nurse plants. Ecological applications, 14(4): 1128-1138. doi:10.1890/03-5084.

Grabarnik, P., Myllymäki, M., Stoyan, D. 2011. Correct testing of mark independence for marked point patterns. Ecological Modelling, 222(23/24): 3888–3894.

Grace, S. L., Platt, W. J. 1995a. Effects of adult tree density and fire on the demography of pregrass stage juvenile longleaf pine (Pinus palustris Mill.). Ecology, 83(1):75-86.

Grace, S. L., & Platt, W. J. 1995b .Neighborhood effects on juveniles in an old-growth stand of longleaf pine, Pinus palustris. Oikos, 72:99-105.

Hubbell, SP. 2001. The unified neutral theory of biodiversity and biogeography. Princeton Univ. Press. New Jersey – USA.

Jia, G., Yu, X., Fan, D., Jia, J. 2016. Mechanism Underlying the Spatial Pattern Formation of Dominant Tree Species in a Natural Secondary Forest. PLoS ONE, 11 (3): e0152596. https://doi.org/10.1371/journal.pone.0152596.

Kang, H., Zheng, Y., Liu, S., Chai, Z., Chang, M., Hu, Y., Li, G., Wang, D. 2017. Population structure and spatial pattern of predominant tree species in a pine–oak mosaic mixed forest in the Qinling Mountains, China. Journal of Plant Interactions, 12 (1): 78-86, DOI: 10.1080/17429145.2017.1283069.

Kazempour, L. M., Taheri, A.K., Pourbabaei, H., Pothier, D., Amanzadeh, B. 2018. Spatial patterns of trees from different development stages in mixed temperate forest in the Hyrcanian region of Iran. Journal of Forest Science, 64(6): 260-270.

Kefi, S., Rietkerk, M., Alados, C.L., Pueyo, Y., Papanastasis, V.P., El Aich, A., de Ruiter, P.C. 2007. Spatial vegetation patterns and imminent desertification in Mediterranean arid ecosystems. Nature, 449: 213-U215.

Lan, G., Zhu, H., Cao, M. et al. 2009. Spatial dispersion patterns of trees in a tropical rainforest in Xishuangbanna, southwest China. Ecological Research, 24: 1117–1124.

Lin, YC., Chang, LW., Yang, KC., Wang, HH., Sun, IF. 2011. Point patterns of tree distribution determined by habitat heterogeneity and dispersal limitation. - Oecologia 165(1): 175-184.

Matthesius, A., Chapman, H., & Kelly, D. 2011. Testing for janzen-connell effects in a west african montane forest. Biotropica 43 (1): 77-83.

McIntire, E.J.B., Fajardo, A. 2009. Beyond description: the active and effective way to infer processes from spatial patterns. Ecology, 90 (1): 46-56.

Muhamed, H., Lingua, E., Maalouf, J.P., Michalet, R. 2015. Shrub-oak seedling spatial associations change in response to the functional composition of neighbouring shrubs in coastal dune forest communities. Annals of Forest Science, SCI 72 (2): 231–241. doi.org/10. 1007/s13595-014-0419-8.

Muhamed, H., Youssef, S., Mustafa, A., Hishyer, H., Abdulqader, A., Mohammed, H., Michalet, R. 2018. Natural regeneration of Pinus brutia Ten. in a recreational public forest in Zawita- Kurdistan region, Iraq Journal of Forest Research, - Online 1–9 https://doi.org/10.1007/s11676-018-0739-x.

Özdemir, Ô. 1977. Studies on the possibilities of natural regeneration of Pinus brutia Ten.in Antalya region (in Turkish). Review of the Faculty of Forestry, University of Istanbul A/2 :( Abstract). 239-290.

Padilla, FM., Pugnaire, FI. 2006. The role of nurse plants in the restoration of degraded environments. Frontiers in Ecology and the Environment 4 (4): 196-202.

Palik, B.J. R., Mitchell, J., Houseal, G., Pederson, N. (1997). Effects of canopy structure on resource availability and seedling responses in a longleaf pine ecosystem. Forest Ecology and Management, 27 (9): 1458-1463.

Pielou, E. C. 1961. Segregation and symmetry in two-species populations as studied by nearest-neighbour relationships. Journal of Ecology, 49 (2): 255-269.

Prévosto, B., Amandier, L., Quesney, T., de Boisgelin, G., Ripert, C. 2012. Regenerating mature Aleppo pine stands in fire-free conditions: site preparation treatments matter. Forest Ecology and Management, 282:70–77.

Puerta-Piñero, C., Gómez, J.M., & Valladares, F. 2007. Irradiance and oak seedling survival and growth in a heterogeneous environment. Forest Ecology and Management, 242(2/3): 462-469. doi.org/10.1016/j.foreco.2007.01.079.

Quézel, P. 2000. Taxonomy and biogeography of Mediterranean pines (Pinus halepensis and P. brutia). In: Ne’eman, G., Trabaud, L. (eds), Ecology, Biogeography and Management of Pinus halepensis and P. brutia forest Ecosystems in the Mediterranean Basin, Backhuys Publishers, Leiden, pp.1–12.

Quero, J.L., Gómez‐ Aparicio, L., Zamora, R., Maestre, F.T. 2008. Shifts in the regeneration niche of an endangered tree (Acer opalus ssp. granatense) during ontogeny: using an ecological concept for application. Basic and Applied Ecology, 9(6):635-644.

Rayburn, A.P., Monaco, T.A. 2011. Linking plant spatial patterns and ecological processes in grazed Great Basin plant communities. Rangeland Ecology and Management, 64 (3): 276–282.

Richardson, D. 2000. Mediterranean pines as invaders in the Southern Hemisphere. In: Ne’eman G, Trabaud L. (eds). Ecology, Biogeography and Management of Pinus halepensis and P. brutia Ecosystems in the Mediterranean Basin. Backhuys Publishers, Leiden, the Netherlands. pp. 131-142.

Rodríguez-Garcia, E., Bravo, F., Spies, A.T. 2011a. Effects of overstorey canopy, plant–plant interactions and soil properties on Mediterranean maritime pine seedling dynamics. Forest Ecology and Management, 262 (2): 244-251. 10.1016/j.foreco.2011.03.029.

Rodríguez-Garcia, E., Ordóñez, C., Bravo, F .2011b. Effects of shrub and canopy cover on the relative growth rate of Pinus pinaster Ait. seedlings of different sizes. Annals of Forest Science, 68 (2): 337-346.

Seidler, T.G., Plotkin, J.B. 2006. Seed Dispersal and Spatial Pattern in Tropical Trees. PLOS Biology, 4 (11): 1-6. e344. https://doi.org/10.1371/journal.pbio.0040344.

Stoll, P., Bergius, E. 2005. Pattern and process: competition causes regular spacing of individuals within plant populations. Journal of Ecology, 93 (2): 395-403.

Szwagrzyk, J., Czerwczak, M. 1993. Spatial patterns of trees in natural forests of East Central Europe. Journal of Vegetation Science, 4 (4): 469-476.

Thanos, C.A., Skordilis, A. 1987. The effect of light, temperature and osmotic stress on the germination of Pinus halepensis and Pinus brutia seeds. Seed Science and Technology, 15: 163–174.

Thanos, C.A., Doussi, M. 2000. Post-fire regeneration of Pinus brutia forests. In: Ne'eman G, Trabaud L, editors. Ecology, Biogeography and management of Pinus halepensis and P. brutia forest ecosystems in the Meditenaneanbasin. Backhuys Publisher, Leiden, the Netherlands, pp. 291- 301.

Urgenc, S. 1971. Comparative results of long-term storage of Pinus brutia methods. Istanbul Universitei Orman Fakultesi Dergisi 21(2):82-93.

Wang, X., Wiegand, T., Anderson‐Teixeira, K.J et al. 2018. Ecological drivers of spatial community dissimilarity, species replacement and species nestedness across temperate forests. Global Ecology and Biogeography, 27 (5):581–592. https://doi.org/10.1111/geb.12719.

Wiegand, T., Kissling, W.D., Cipriotti, P.A., Aguiar, M.R. 2006. Extending point pattern analysis for objects of finite size and irregular shape. Journal of Ecology, 94(4):825-837. doi:10.1111/j.13652745.2006.01113.x.

Wiegand, T., Moloney, K.A. 2004. Rings, circles, and null-models for point pattern analysis in ecology. Oikos 104(2):209-229. doi:10.1111/j.0030-1299.2004.12497.x.

Wiegand, T., Uriarte, M., Kraft, N.J., Shen, G., Wang, X., He, F. 2017. Spatially explicit metrics of species diversity, functional diversity, and phylogenetic diversity: Insights into plant community assembly processes. Annu Rev Ecol Evol S 48: 329–351.

Yuan, Z., Wei, B., Chen, Y., Jia, H., Wei, Q., Ye, Y. 2018. How do similarities in spatial distributions and interspecific associations affect the coexistence of Quercus species in the Baotianman National Nature Reserve, Henan, China. Ecology and Evolution. 8 (5): 2580–2593. http://doi.org/10.1002/ece3.3863.

Zhang, M.T. 2017. Spatial association and optimum adjacent distribution of trees in a mixed coniferous- Broadleaf forest in northeastern China. Applied Ecology and Environmental Research, 15(3): 1551-1564. 10.15666/aeer/1503_15511564.

Zhang, Z., Hu, G., Zhu, J., Ni, J. 2013. Aggregated spatial distributions of species in a subtropical karst forest, southwestern China. Journal of Plant Ecology, 6 (2):131- 140, doi.org/10.1093/jpe/rts027.

Appendix

The following Figures A2-A8 show the spatial distribution of P. brutia regeneration and main shrub species mapped in 25 m by 20 m rectangular plots, showing the aggregation pattern of both seedling and saplings in Zawita forest. 

Die folgenden Abbildungen A2-A8 zeigen die räumliche Verteilung der P. brutia-Verjüngung und der Hauptstraucharten in 25 m * 20 m großen rechtwinkeligen Probeflächen an, die das Aggregationsmuster von Keimlingen und Jungpflanzen im Zawita-Wald aufzeigen.