An analysis of the bryophyte flora in Sicilian archaeological areas*

Campisi, P., Dia, M. G. & Marino, M. L.: An analysis of the bryophyte flora in Sicilian archaeological areas. — Fl. Medit. 29: 143-157. 2019. — ISSN: 1120-4052 printed, 2240-4538 online. An analysis of the bryophyte diversity in the studied Sicilian archaeological areas was conducted, highlighting which species are more common and potentially harmful on the ruins. The floras are much diversified and the presence of some rare taxa highlights the role of refuge carried out by these areas, especially for the species of strongly threatened coastal habitats. Attention on the complexity of the relationships between restoration interventions on lithic structures and conservation needs of the rare and interesting taxa is point out.


Introduction
Among Italian regions, Sicily is certainly one of those that boasts a particularly rich archaeological heritage, consisting of over 50 sites distributed throughout the territory, in many cases subjected to specific protections (Grella & al. 1997).
As evidenced by several studies, the archaeological areas are of considerable interest also from the naturalistic point of view as they represent refuge areas for numerous species threatened by urban expansion and generally host a high floristic diversity (Celesti Grapow & al. 1993-1994;Lucchese & Pignatti Wikus 1995;Ceschin & al. 2006;Minissale & al. 2015).
In Italy the studies on the flora of the archaeological sites started in the Nineties and Sicily is one of the regions most concerned by this research activity (Domina 2018).
In addition to the vascular component, also the cryptogamic one was investigated in Sicily, in order to improve the knowledge of the main biodeteriogens on the remains of the ancient human-made structures (Raimondo & al. 2008).Several researches have focused in particular on bryophyte floras, starting from the Nineties, as was pointed out by Gueli & al. (2005).Some of these studies have highlighted the presence in Sicilian archaeological areas of very rare taxa and therefore of conservation interest (Dia & al. 2003;Campisi & Provenzano 2004;Dia & Campisi 2006;Campisi & al. 2008;Dia & Campisi 2009), drawing attention to the need to clarify the deteriogenic role of bryophytes to evaluate in some cases the hypothesis of their maintenance on the ruins.
Here we analyze so far studied bryophyte floras in some archaeological areas of Sicily with the aim of evaluating the bryophyte diversity of the areas and highlighting which species are more common and potentially harmful on the ruins as well as wich species are worthy of conservation.

Materials and methods
The Archaeological Parks of Solunto and of Mount Iato in Palermo province, those of Segesta and Selinunte e Cave di Cusa in Trapani province, the Neapolis Archaeological Park of Syracuse town, the Temple of Cerere in Enna province and the Greco-Roman Theatre and the Roman Amphitheatre inside Catania town have been taken into consideration in this analysis (Table 1).In view of their naturalistic interest, some of these areas, in particular Mount Iato Archaeological Park, Selinunte and Cave di Cusa Archaeological Park and Solunto Archaeological Park, are included in the list of Sites of Community Interest (ITA020027 Monte Iato, Kumeta, Maganoce e Pizzo Parrino; ITA010011 Sistema dunale Capo Granitola, Porto Palo e Foce del Belice; ITA020019 Rupi di Catalfano e Capo Zafferano).The present study was based on bibliographic and unpublished data.In particular, we have considered the papers of Privitera & al. (1996), Lo Giudice & Cristaudo (1998), Di Benedetto & Grillo (1998), Puglisi (1999), Guglielmo & al. (2003) and Aiello & al. (2003).Furthermore, the data concerning bryophyte material collected by the authors in the Sicilian archaeological areas of Selinunte and Mount Iato were also taken into consideration.
The archeological areas of Selinunte, Syracuse, Catania, Solunto and Segesta are located at lowland or hilly altitudes (10-415 m a.s.l.) on the contrary Mount Iato and Enna are located at sub-mountain altitudes (850-970 m a.s.l.).Vegetation is mostly represented by xerophilous or mesophilous grasslands and garrigues; only at Selinunte psammophilous, wetlands communities and Mediterranean maquis also occur (Table 1).
The relevant samples are kept at the Herbarium Mediterraneum Panormitanum (PAL).
The nomenclature of the species comply with Söderström & al. (2016) for liverworts and hornworts and Ros & al. (2013) for mosses, with the exception of Didymodon tophaceus complex for which Kučera & al. (2018) is followed .
Ellenberg's indicator values relating to light, temperature, moisture and substrate reaction, were taken from Düll (1991).
All available data were assembled in a Microsoft Access database, where for each taxon ecological indicator values, chorotypes, and occurrence in the Sicilian archaeological areas were recorded.
For each site, the ecological indicator mean values were calculated to draw radar diagram.In order to measure the level of similarity of the floras, a hierarchical cluster analysis using Jaccard's similarity index (Jaccard 1908), applied on a data matrix including presence-absence of moss and liverwort taxa, was performed with the Biodiversity Professional program (McAleece & al. 1997).
The bioclimatic thermotypes were derived from the thermotypes map provided by Pesaresi & al. (2014).
Moreover, with the aim of identifying the potentially most dangerous species for the integrity of the ruins, in the sites of Selinunte, Segesta and Solunto the following data were recorded by the authors in the field and analized: 1) frequency of taxa in five classes based on percentage occurrence in the total collection points of every archaeological area (I: 0-20%, II: 21-40%, III: 41-60%, IV: 61-80%, V: 81-100%); 2) percentage cover in five classes based on percentage cover of each species on the total surface colonized by the bryophytes in 30 ×30 cm areas, in every archaeological area (I: 0-20%, II: 21-40%, III: 41-60%, IV: 61-80%, V: 81-100%); 3) sporophyte or propagule production (production was considered high (+) when their occurrence was recorded in more than 30% of plants in the collected specimens, low (-) if in 30% of plants or less).
Life strategies were taken from Dierβen (2001) with same abbreviations.In all figures and tabular material studied sites were indicated with following abbreviations: Ca: Greco-Roman Theatre and the Roman Amphitheatre in Catania town; En: Temple of Ceres in Enna town; Ia: Mount Iato Archaeological Park; Sg: Segesta Archeological Park: Si: Neapolis Archaeological Park in Syracuse town; Sl: Selinunte and Cave di Cusa Archeological Park; So: Solunto Archaeological Park.

Results and Discussion
Overall, the bryophyte flora of the Sicilian archaeological areas studied so far includes 88 taxa (75 mosses and 13 liverworts).They are listed in Table 2, where the presence in each area, ecological indicators and chorotypes are reported.
From the chorological point of view, the floras are quite diversified since the Mediterranean taxa prevail in the floras of Selinunte (43.1%),Segesta (50%) and Solunto (43,8%) parks and Catania site, while the temperate taxa are the most represented in Enna site (40%) and in Mount Iato (55.6%) and Syracuse parks (36.4%) (Fig. 1).The high incidence of the latter type in the first two archaeological areas is likely to be related to their mountain altitudes, while in Syracuse park could be due to the significant presence of wet microhabitats in it.
An analysis of the average values of the Ellenberg index (Fig. 2) shows a relative uniformity of the floras with reference to the behaviour of the species with respect to the various factors, since the average values show variations of at most 1 and 1.5 units.Some significant differences may, however, be noted such as particularly high xerophily of the Solunto and Catania floras (average values of the moisture factor 3.7), the relatively low values concerning the temperature factor at the Enna site and in Mount Iato park (average value less than 6), the low value related to the reaction of the substrate in the area of Catania (6.3 average value).
A qualitative comparison of the floras, conducted through a cluster analysis, highlighted an accentuated diversification of the floras, as demonstrated by the low levels of link in the dendrogram of Fig. 3. Furthermore, it is observed that the clusters reflect the thermoclimatic conditions of the sites.The greatest similarity is found, in fact, among the floras of Bocconea 28 -2019 261 Table 2. continued.
Selinunte, Solunto and Syracuse, characterized by a lower thermomediterranean climate thermotype and to this cluster the floras of Mount Iato and Segesta (lower mesomediterranean climate thermotype), the flora of Catania (upper thermomediterranean climate thermotype) and, lastly, the one of Enna (upper mesomediterranean climate thermotype) are sequentially linked.
In addition to the high dissimilarity, the peculiarity of the floras of the Sicilian archaeological areas is due to the presence of taxa that be able to live almost exclusively in coastal areas with little disturbance.Among them four taxa, Crossidium laxefilamentosum W. Frey & Kürschner, Gigaspermum mouretii Corb., Pseudocrossidium obtusulum (Lindb.)H.A. Crum & L.E.Anderson and P. replicatum (Taylor) R. H. Zander., are of notable interest as they are very rare in Italy and candidates to inclusion in the European Red Data Book (Hodgetts 2015).
Pseudocrossidium replicatum was reported in Europe and Mediterranean basin only in Sicily (Solunto and Linosa) and Calabria (Dia & al. 2003;Privitera & Puglisi 2000;Ros & al. 2013;Hodgetts 2015).Elsewhere it is distributed in Central and South Africa, South-West Arabia and America (Zander 1981;Menzel 1986;Frey & Kürschner 1988a1988b).In the archeological park of Solunto it lives on basic soil in interstices between quartz mosaic tesserae of the houses floors.
Gigaspermum mouretii is an oceanic-Mediterranean species scattered in Syria, Israel, Cyprus, Turkey, Crete, Sicily, Morocco, Canary Islands, Spain, and Balearic Islands (Ros & al. 2013).In Italy it was reported only from two locality of Sicily (Selinunte and Capaci) (Campisi & al. 2008).At Selinunte it grows on calcarenite plinth of the temples E and F.
Crossidium laxefilamentosum was reported from Europe , Asia (Arabian peninsula) and Africa.In the Mediterranean basin it is known from Egypt, Serbia, Sicily, Tunisia and Turkey (Ros & al. 2013).In Sicily it was recorded only in two archaeological areas (Solunto and Molino a Vento near Gela) (Dia & Campisi 2009;Puglisi & al. 2013).At Solunto it lives on exposed soil among ruins of the archaeological park.
On the basis of coverage, frequency, abundance of sporophytes and propagules data as well as the life strategies of taxa living in the Segesta, Selinunte and Solunto archaeological parks, some species of mosses were identified as a potentially great threat to the state of conservation of the ruins in consideration of the direct correlation between species diffusion and coverage on the one hand and their possible biodeteriogenic action on the other.In particular, they are Tortula muralis Hedw., Scorpiurium circinatum (Bruch) M. Fleisch.& Loeske and Didymodon vinealis (Brid.)R. H. Zander, which are present in all three areas, always reaching cover class II or higher.Furthermore, Aloina ambigua (Bruch & Schimp.)Limpr., Barbula convoluta Hedw., Bryum dichotomum Hedw., Didymodon luridus Hornsch., Funariella curviseta (Schwägr.)Sérgio, Pseudocrossidium hornschuchianum (Schultz) R. H. Zander and Tortella nitida (Lindb.)Broth.occure in at least 2 areas with high frequency or coverage class (III or more) at least in one area.All these species, with the exception of the pleurocarp moss Scorpiurium circinatum, are annual or colonists, biological strategies that according to Dierßen (2001) are characterized by short-lived (<1 year-few years) with more or less high reproductive effort.They present a high production of sporophytes already in the first year of growth or after 2-3 years or form also propagules in the first years of life and, therefore, tend to expand more and more on the stone substrates, continuously creating new colonies.On the contrary, the perennial species Scorpiurium circinatum, shows reproductive effort low and begin to form sporophytes only after several years.It forms very wide moss mats that extend above all on horizontal surfaces of archaeological structures in the studied areas.
The liverworts also show a high reproductive capacity both through the sporification and with different modality of vegetative propagation (propagules production in Lunularia cruciata (L.) Lindb.and y-shaped growth form in Riccia species).Five out of seven species (Fossombronia caespitiformis De Not.ex Rabenh., Riccia glauca L., R. lamellosa Raddi, R. sorocarpa Bisch., and Sphaerocarpos michelii Bellardi) present an "annual" strategy, being characterized by a very rapid growth of the vegetative body (few months) and very high formation of spores (Dierßen 2001).However, these species spread more widely on soils and, therefore, do not reach high frequency values on the ruins, where, except for Fossombronia caespitiformis, they have a low degree of coverage due to their small size.
Regarding the periods of maximum sporification, it is observed that most bryophytes release the spores in the spring (Table 4).In some cases the maturation of the sporophytes is a little early and occurs already in later winter (February and March), in others it is pro- longed in the summer.Few species have autumnal sporification.Therefore, the analysis of these data suggests that the winter is the most suitable season to guarantee the effectiveness of the interventions on the ruins.

Conclusive considerations
Overall, this analysis confirms the naturalistic interest of archaeological areas for their significant floristic diversity and species richness, whose values are comparable to those recorded in Sicily in some natural areas (Campisi & al. 2006).The importance of these areas is also increased by the presence of some rare taxa in Europe for which specific attention would be required during the necessary, periodic, restorative interventions on remains of architectural structures.
These cleaning interventions should primarily be aimed at the removal of widespread bryophytes with numerous or extensive colonies, considering that the action of biodeterioration is certainly related to the degree of bryophyte coverage.Furthermore, the possibility of diffusion of spores and propaguls due to colony detachment operations should not be underestimated.
Nevertheless, it should be emphasized that further research is desirable to better clarify the biodeteriogenic role of different bryophyte taxa, so far ascertained only in some epilithic moss species (Hughes 1982;Altieri & Ricci 1994;Altieri & al. 1997).Many species on lithic structures, indeed, are not true saxicolous but they settle only in small accumulations of soil in grooves and fractures caused by the alteration usually due to other deteriogenic agents; hence, the possibility that they, like other terricolous bryophytes, can exercise, instead, a protective action on substrates cannot be excluded with certainty at least in some cases.The important role of moss coverings against soil erosion is, in fact, well known both in forest ecosystems and in arid habitats where they contribute to the formation of the biological soil crusts, very effective to counteract the action of atmospheric agents (Weber & al. 2016).

Table 1 .
Studied areas and relatives data of surface, altitude and vegetation types.

Table 4 .
Frequency and cover classes, sporophyte or propagule presence, time of sporophyte production and life strategies (from Dierßen 2001) of taxa growing in the sites of Segesta (Sg), Selinunte (Sl) and Solunto (So).a:annual;c: colonist; ec: ephemeral colonist; l: long-lived shuttle p: perennial; pc: competitive perennial; s: short-lived shuttle; sp: stress tolerant perennial.For abbreviations of archeological areas see Table1.