Distribution, demography and conservation perspectives of Retama raetam subsp. gussonei (Fabaceae) in Calabria (S Italy)

� Abstract: The southernmost region of peninsular Italy, Calabria, hosts two well distinct sub-populations of Retama raetam (Forssk.) Webb. subsp. gussonei (Webb) Greuter (Fabaceae), a Sicilian-Calabrian endemic shrub. The population study model applied to this work on Calabrian sub-populations of this threatened taxon, demonstrated to be a powerful tool. A preliminary survey on a sample of the studied plants enabled the fast, easy and complete collection of biometric/anagraphic data. Present number of censussed plants (1,523; 82.5% by the Ionian Sea, 17.5% by the Tyrrhenian), the complex and heterogeneous composition of the population (divided in four biometric/age classes and spread on nine different sectors), its distribution within the study area, its evolution as a result of spontaneous population dynamic or future management actions, might be hopefully long term monitored through the study model here applied.


Introduction
Calabria (Fig. 1D), located at the centre of the Mediterranean basin, is the southernmost part of Italian Peninsula.It's surrounded by about 740 km of coasts, most of which are sandy and altered by human settlement.
The Ionian sub-population, located in the northern part of the Cirò Municipality (Crotone Province), occurs close to seacoast (about 50-600 m inland from the coastline).The investigated area is encompassed SE by the final part of the Vota Ranna Stream (or Santa Venere Stream), NE by the Ionian Sea, NW by Carafuno di Cappellieri Stream and SW by the municipal road (Fig. 1A).The investigated area has altitude ranging 0-60 m a.s.l. and is crossed by the cited small streams as well as by Godano Stream, a small intermittent watercourse.Substrata are mainly fixed alluvia and dunes (Marchetti et al., 1968a-c).An inwards morphological terrace (silty clay, sands, and conglomerates) occurs, while around watercourses alluvial substrata prevail (Fig. 2A).The area is crossed by the national road S.S. 106 and by the railway following the coast line about 200-400 m inland (Fig. 1A).NE of the railway line, the landscape is dominated by exotic woody species plantations such as Eucalyptus camaldulensis Dehnh.subsp.camaldulensis, Pinus halepensis Mill., Pinus pinea L., Acacia saligna (Labill.)H.L. Wendl.The inland slope of dunes is mostly dominated by R. raetam subsp.gussonei.W of the S.S. 106 a vegetation mosaic occurs, encompassing both small Mediterranean steppe patches (especially close to S.S. 106), areas of Mediterranean maquis, and R. raetam subsp.gussonei vegetation.The inland landscape is dominated by cultivations (Gangale et al., 2008;Uzunov et al., 2009).Important threating factors on Ionian stands are weeds such as Acacia saligna and Carpobrotus edulis (L.) N.E.Br.Both of them seem able to occupy wide areas of the dune system at the expenses of native flora, Retama included.
Tyrrhenian sub-population occurs in a coastal area (around 400 m from coast line inwards) belong-ing to San Ferdinando (Reggio Calabria Province) and Nicotera municipalities (Vibo Valentia Province) (Caruso et al., 2010a).The area is crossed by watercourses, the most important of which is Mesima Stream.The dune system is altered and ecologically fragmented due to human activities (buildings, tourist facilities, roads, cultivations, exotic woody species plantation) and weeds such as Cestrum parqui L'Hér., which is probably the most hazardous.Anyway, the relict semi-natural coastal area hosts also the rare Ephedra fragilis Desf.(Caruso et al., 2010b;Caruso , 2012;Caruso & Montepaone, 2020).From the geological point of view alluvia and dunes prevail (Marchetti et al., 1968b; Fig. 2B-C), although the whole district is currently undergoing an important erosion process (Chiaravalli et al., 1990;D'Alessandro et al., 1992).
Aim of this work is to develop an easy-to-apply, low time-consuming, low-invasive scientific model to study shrub-like species population demography.Moreover, considering the present level of risk of extinction in the area of this rare endemic taxon, this work aims to provide detailed quantitative and qualitative data on Calabrian population of R. raetam subsp.gussonei, its demographic structure and geographical distribution.This may contribute to management policy needed to ensure population long term monitoring and conservation.

Materials and methods
The field work has been carried out mainly during 2007 and 2008 spring-summers, but additional information has been collected until the 2022 summer.The identification of the collected material has been done according to Flora d'Italia (Pignatti, 1982;Pignatti, 2017Pignatti, -2020) ) and Flora Europaea (Heywood, 1967).

Study area
Considering the physical (e.g.roads, railway, villages, camp areas, cultivated fields, disturbed areas, etc.) and biological barriers (e.g.exotic woody species forest) existing, inside the areas where R. raetam subsp.gussonei occurs, ecologically uniform sectors have been delimited (5 by the Ionian Sea, 4 by the Tyrrhenian) as shown in Fig. 1.

Biometric-anagraphic development model
Before starting the census operation, a preliminary study has been necessary in order to define a simple, low time-consuming, low-invasive sampling technique suitable to know, with statistically acceptable approximation, the age of each single plant belonging to the studied population.R. raetam subsp.gussonei plants have, if observed from above, a circular shape.This depends on the uniform distribution in all directions of the branches from central stem axis (Fig. 4).A simple way to estimate the age (or age range) of this shrub, is to measure (or estimate) the diameter of the projection of the plant canopy on the ground.Except the infertile juvenile plants (class 0 = 0-0.5 m), the reproductive population of the surveyed species has been therefore arbitrarily divided into 4 biometric classes, based on the diameter of the projection of the canopy on the ground (class 1 = canopy diameter 0.5-1 m; class 2 = 1-2 m; class 3 = 2-3 m; class 4 = 3-6 m) (Fig. 4; Table 1).The next step has been to verify whether these biometric classes were in fact correspondent to different age groups.For this purpose the architectural/anagraphic structure of plants has been studied in detail.A first sample, made of 30 plants belonging to all morphological classes, have central stem axis and branches sampled by a Pressler gimlet.For each sample of wood annual rings have been counted, and so has been determined the age of each  Sáenz (1996Sáenz ( -2009)).
Historia naturalis bulgarica 46 (2024) part of studied plants (Fig. 4).From these data it was possible to know, for each sampled plant, the individual age (= rings of central axis) and the age of the outer branch (= rings of branch).Consequently, has been possible to know the relationship among outer branch age, individual age and plant canopy diameter for each studied plant and all biometric classes (Fig. 4; Table 1).The biometric-anagraphic survey, as shown ahead (Table 1), demonstrated the existence of a close correlation among the average values of plant age, the outer branch age and the canopy diameter.This allowed, following the opposite path, on a second sample made of 120 plants, to know the age of a single plant just through the age of the outer branch easily sampled by Pressler gimlet.Moreover, once calculated the age range of each biometric class, it has been possible to trace the age range of each plant just measuring the canopy diameter.This demonstrates, for the surveyed taxon, the diameter of the canopy projection on the ground to be a reliable and easy-toassess parameter for an effective biometric-anagraphic classification of plants.

Population census and analysis
The reproductive Calabrian population of R. raetam subsp.gussonei, represented by the 4 biometric classes above described, has been censussed.
Seedlings and not reproductive juveniles (canopy diameter < 50 cm) have been excluded from the present analysis because difficult to count in the field and because of the already known high death rate (Peruzzi & Cesca, 2003).The occurrence of these individuals has just been recorded, but not quantified, inside all sectors.A quanti-qualitative analysis on the whole Calabrian population of the surveyed species has been carried on in the different sectors considering the biometric classes.

Biometric classes and plant development model
Biometric-anagraphic sampled measures provided data summarised in Table 1.
The study of populations of rare species and characterising of their development models are used in different study (Pogorzelec et al., 2023;Sakhraoui et al., 2024).Several plants belonging to the biometric class 0 (canopy diameter 0-0.5 m) have been found in each sector of the study area.This class, excluded by the current analysis for the reasons above explained, is composed by plants younger than 4 years (seedlings/not reproductive juveniles), usually infertile.Class 4 has not been split in different classes because very few plants with a canopy diameter wider than 4 m have been found.Besides, these wider plants grow mainly isolated inside clearings.It has been believed that uncommon size reached by these plants was mainly due to the absence of competition, and consequently was poorly significant.Biometric-anagraphic study showed that branches belonging to a single plant do not have the same age.Stem axis is the oldest part of the plant, while the outer branches are the youngest and inner branches are in between.Despite some overlapping values of plant age (e.g.classes 2, 3 and 4 in Table 1), probably due to microecological factors locally affecting morphological development of plants, Table 1 shows high values of correlation (r) among canopy projection diameter and outer branch diameter (0.999), as well as canopy projection diameter and outer branches average age (0.991) and even canopy projection diameter and plant age (0.994).These make the canopy diameter, at least for the purpose of this work, a statistically reliable measure of individual age (or age range) for this taxon in the investigated area.
Considering the distribution of plants among biometric classes it can be noticed that class 1 (498 plants; 32.7% of total), class 2 (423; 27.8%) and class 3 (533; 35%) are rather balanced (around 1/3 each), while class 4 (69; 4.5%) is the less represented.The oldest component of the population (class 4) seems to be quantitatively less important, while the youngest components are dominant.A situation like this is expected in a healthy population with a good reproductive perspective.
Analysing the population by sectors it is evident that the most plant-rich sector is I3 (457 plants; 30% of the whole population); it follows sector I2 (322; 21.1%), sector I4 (239; 15.7%), sector T3 (234; cal distribution data are available about Tyrrhenian Calabrian stands of R. raetam subsp.gussonei.Randomly interviewed local elderly report a wider distribution of R. raetam subsp.gussonei by the Tyrrhenian coast of Calabria in the past.Human activities and coastal erosion reduced this area severely fragmenting it.Some weedy species (e.g.Acacia saligna) seems now to have a stable population, at least in some tracts of the coast where frequent are ruinous attempts of Citrus sp.div.cultivation.The recent formation of small patches of R. raetam subsp.gussonei in these abandoned orchards probably could explain the frequency of young plants (class 1) and, in the long term, could lead this rare taxon to recover part of the presumed former area of occurrence.

Geomorphological role
R. raetam subsp.gussonei plays an important geomorphological role contributing to create and stabilise the dune system.Sand moved by wind is stopped by plant branches, so tending to accumulate at the plant base mixed with litter.This determines the progressive uplift of the dune top and the partial burial of plant's branches.Considering that number and age of branches increases with plant age and that in juveniles first branch is above the ground level, while in the adults is mostly covered by sand, the average dune uplift during the plant life (up to 20 years) is 50 (20-80) cm (Fig. 4).

Long term monitoring and conservation
As part of the European Natura 2000 network the Site of Communitary Importance (SCI) IT9320100 named "Dune di Marinella" has been created on the Ionian coast of Calabria (Regione Calabria, 2003) 7).This area is mainly characterised by low productive sandy soils (Fig. 2), constantly at risk of abandonment, mostly intended for herbaceous annual crops while the Cirò area is commonly renowned for high quality viticulture.On the other hand, the area could represent a future opportunity for Retama plants to colonise new habitats, therefore deserves to be included in the SCI perimeter, so providing also a more compact shape to the protected area.Second, the new official perimeter excluded some plants (Fig. 6) belonging to the I3 sector, in the NE of the Ionian distribution area.Third, in the SE corner of the new official perimeter some olive tree cultivations have been, maybe erroneously, included.Fourth, the name of this SCI can no longer be "Dune di Marinella", because "Marinella" is a locality currently far from the SCI new official perimeter.Considering these observations, is here proposed, for the SCI area, the new name "Dune di Cirò" (meaning "Dunes of Cirò" in Italian), and a new perimeter (Fig. 7), virtually resolving all the recorded issues properly.
The lack of protection measures, the altered environment, as well as the Tyrrhenian R. raetam subsp.gussonei population size and structure, expose it to a high risk of extinction.Being the area also interested by the occurrence of one of the only two known stands of Ephedra fragilis Desf. of peninsular Italy (Caruso et al., 2009;Caruso et al., 2010b;Caruso et al., 2012;Caruso & Montepaone, 2020), it should be justified the establishment of a new protected area in this territory.The proposed protected area (as a floristic protected area, regional park or, better, as a SCI), possibly named "Dune di Nicotera Marina e S. Ferdinando", would adopt measures against further anthropic pressure.
Long term conservation of this precious taxon largely depends on the environmental quality and on the possibility to maintain and even increase the present area of occurrence.The dangerous co-occurrence of invasive species such as Carpobrotus edulis (Ionian stands), Cestrum parqui (Tyrrhenian stands) or Acacia saligna (Ionian and Tyrrhenian stands) strongly affects the surveyed species actively competing for space and resources.In order to preserve the existing stands of Retama raetam subsp.gussonei the complete eradication of these alien taxa seems to be the only solution.The other woody exotic species widely planted in the surveyed area, although not invasive, occupy a large surface potentially useful to increase the present R. raetam subsp.gussonei area of occurrence.A progressive programme of eradication should be in this case a good solution.
Considering the IUCN category to which this taxon belongs and the persistence of threating factors on studied sub-populations, a long term monitoring programme must be urgently planned.Considering biology and ecology of this taxon, applying the methodology developed as part of this work, the frequency of monitoring should be 3-10 years.This would allow comparable data to be easily collected.As the eradication of weeds proceeds a 2-3-year monitoring protocol should be scheduled in order to lead dynamic management.

Conclusions
The population study model applied to this survey on Calabrian sub-populations of R. raetam subsp.gussonei demonstrated to be a very powerful tool.A preliminary survey on a sample of investigated plants made it possible a fast, easy and complete collection of biometric/anagraphic data on the whole population.
The present number of plants, the complex and heterogeneous composition of the population, its distribution within the study area, its evolution as a result of spontaneous population dynamic or future management actions, as well as the possible establishment of specific protected areas needed for the protection of this rare and endemic taxon in Calabria, might be hopefully long term monitored through the study model here applied.
and rows represent sectors (Ionian sectors: I1-I2-I3-I4-I5; Tyrrhenian sectors: T1-T2-T3-T4).Each biometric class (with the only exception of class 0) is represented by two sub-columns.The sub-column on the left side of the table contains the number of plants belonging to that class occurring in that specific sector; the sub-column on the right contains the percentage value of that class on the total amount of plants occurring in that specific sector.Similarly, each sector is represented by two rows.The upper row corresponds to the above described left column value, so contains the number of plants for that class in that sector.The lower row contains the percentage value of that sector on the total amount of plants belonging to a specific class.For example, inside the class 1, in the first column there is 6, in the second 13.6, in the lower row 1.2.This means that inside the sector I1 6 plants belonging to class 1 occur; this class represents 13.6% of total individuals occurring in the sector I1; this means also that in the sector I1 1.2% of plants belonging to the whole class 1 occur.Grey column on the right summarise data per each sector, while the grey row at bottom summarise data per each class.The two columns at the extreme right host aggregate data (and %) of Ionian and Tyrrhenian sub-populations.The overall Calabrian population of R. raetam subsp.gussonei counts 1,523 fertile plants.Ionian subpopulation is clearly prevalent with 82.5% of the

Fig. 4 .
Fig.4.Phases of morphological development in a R. raetam subsp.gussonei plant, corresponding to biometric classes adopted in this study and following the dune system evolution.

Fig. 5 .
Fig. 5. Absolute and percentage importance of biometric classes of R. raetam subsp.gussonei on the whole Calabrian population, on the Ionian sub-population and on the Tyrrhenian sub-population.

Fig. 6 .
Fig.6.Old (blue lines) and new (red lines) official perimeters of the "Dune di Marinella" SCI, the second one re-drawn in order to really include the Ionian R. raetam subsp.gussonei stands of Calabria (black).

Fig. 7 .
Fig. 7.The official current perimeter of "Dune di Marinella" SCI (red lines), overlapping the here proposed new perimeter (green lines + green dots) re-drawn according to the criticalities emerged during the present survey.

Fig. 8 .
Fig. 8. Areas by the Calabrian Tyrrhenian coast where R. raetam subsp.gussonei occurs, and proposed new protected area, divided in two sectors (the southernmost of which including also an Ephedra fragilis stand), possibly named "Dune di San Ferdinando e Nicotera Marina".

Table 1 .
Biometric classes and age (in years) range measured on Calabrian population of R. raetam subsp.gussonei and correlation coefficient (r) among some of the measured parameters.

Table 2 .
Calabrian population of R. raetam subsp.gussonei censussed within 9 sectors, 5 of which correspond to the Ionian sub-population (I), and 4 to the Tyrrhenian (T) one.