Ostracod Fauna: Eyewitness to Fifty Years of Anthropic Impact in the Gulf of Trieste. A Potential Key to the Future Evolution of Urban Ecosystems

: For the first time the distribution and modifications of living ostracod associations present in the Gulf of Trieste (GoT) in relation to the alterations caused by human activity in the last 20 years were investigated. The results were compared with the main chemico-physical parameters (especially nitrogen and phosphorus) measured over the same period, which can lead to a general decrease in environmental quality. For a more in-depth analysis of the changes recorded by ostracods in the last 50 years, a period in which eutrophication and anoxia increased, we revisited the study carried out by Masoli in the GoT in 1967. The results obtained made it possible to verify how over the last 20 years, ostracod assemblages have suffered a decrease both qualitatively and quantitatively. Most of the species recovered show characteristics of opportunism and tolerance to environmentally stressful conditions, high organic matter concentrations and oxygen deficiency. The ostracods analyzed in 1967 showed similar results with few dominant opportunistic species. We verified how ostracods recorded in GoT, similar to Mollusks and Foraminifera, the possible environmental crisis linked to the recurrence of mucilage and hypoxic events documented for the Gulf of Trieste in the last 50 years. Finally, a comparison with the best environmental conditions found in the Marine Nature Reserve of Miramare (MPA) allowed us to emphasize the important role of protected areas to avoid the loss of biodiversity due to urbanization.


Introduction
Urbanization is one of the main causes of species extinction [1]. The expansion and growth of anthropic activities decrease biological diversity because as the same ''urban-adaptable'' species become widespread and locally abundant across several ecosystems [2,3]. Thus, a great deal of research has been focused on the human influence generating ecological degradation from coastal urbanization to pollution and eutrophication [4,5]; these fundamental changes have taken place with the evolution of freshwater systems and fluxes during the last century, with particular impact seen over the last 50 years [5,6]. More specifically, drastically increased urbanization is currently one of the main sources of anthropogenic impact [7,8], often exerting significant pressure on coastal ecological systems [9].
The need to start efficient mitigation activities to assess and manage the negative impacts of urbanization on natural habitats (i.e., to reduce the urban footprint and preserve habitats in urbanized areas) have highlighted the importance of investigating anthropic impact and the response The primary freshwater input is represented by the Isonzo/Soča River (annual flow rate of 82 m 3 s -1 ) and the contribution of several minor rivers (Timavo/Reka, Rosandra/Glinščica, Ospo/Osp, Rižana, Badaševica, Drnica and Dragonja) can be considered negligible or having only local effects. Water circulation is driven by the interplay of various forcing factors: the general circulation of the Adriatic Sea, winds (particularly the dominant Bora, N-NE direction), buoyancy fluxes together with tides.
The GoT represents a site of shelf dense water formation that contributes to the North Adriatic Deep Water.
The sediments texture varies from medium to fine sands along the coastline and the delta of the Isonzo and Tagliamento Rivers, to muds in the mid-Gulf and sandy sediments in the western open part of the GoT. Carbonate sediments dominate the sediments near the river' mouth [49,50].
The GoT is a suitable site to study anthropic impact since, in spite of its relatively small extension, it hosts two of the largest cargo shipping ports in the Adriatic Sea (Trieste and Koper). This coastal area is affected by many potential sources of organic and inorganic pollutants, discharged not only by rivers but also by sewers, industrial developments, and harbor related activities including an oil-pipeline terminal [51,52]. Moreover, the site has been recognized as an area where particular conditions related to inputs of fluvial sediments or to meteo-marine conditions led to significant algal productions and blooms, resulting in eutrophication and the subsequent hypoxic/anoxic conditions at the bottom, at least until the mid-1980s [53].

Experimental Site
The experimental site is located in the GoT (13°37' E to 13°44' E and 45°41' N to 45°44' N). In order to assess the changes which have occurred in the GoT since 1967, we applied the GIS methods to reconstruct the exact georeferenced location of the samples collected by the "Istituto di Geologia e Paleontologia" (University of Trieste) during two summer cruises conducted in 1965 and 1966, respectively (site GTCrA) and compared the qualitative analyses of ostracods reported by [46] with those ostracods recovered in 2004 and 2017 ( Figure 1). The collected sediments were wet-sieved using a 63µ m mesh, dried and weighed to determine the content of the sandy fraction. The total sandy sediments were used to collect all live benthic specimens, representative of the environmental conditions, and to avoid problems related to the potential presence of the reworked fauna originating from deposits created in connection with the Early Holocene marine transgressions and the Holocene climatic optimum as indicated by [54] in the eastern part of the northern Adriatic Sea. Ostracod analysis took into account monographs and papers from the Mediterranean literature [55,56,57]. Particular attention was paid to the papers concerning the northernmost sector of the Adriatic Sea corresponding to the GoT [46,58,59].
To characterize the biodiversity of assemblages, two faunal parameters were calculated: (1) species diversity (S), the number of species in each sample; (2) the Shannon-Weaver index (H), a measure of entropy that takes into considerations the distribution of taxa among the total individuals [60] (Table 1).
Multivariate analysis on the GTCrB and GTCrC ostracod assemblages was performed using the Xlstat software Addinsoft (2020) (XLSTAT statistical and data analysis solution. New York, USA. https://www.xlstat.com), except for the calculation of SIMPER and of diversity indices performed using PAST software (PAlaeontological STatistic -version 4.02) [61]. Cluster analysis was run for samples (Q mode). The best results were reached using Ward's method and the Bray and Curtis algorithm. Also included was the presence/absence data used in this analysis for comparative purposes and both provided the same results, thus confirming the validity of the obtained groups.

GIS analysis
Predictive distribution maps for critical species and the Shannon Index have been interpolated in GIS using the Inverse Distance weighting method (IDW). The measured relative frequency values surrounding the prediction location have been used to predict a value for any unsampled location in the study area, based on the assumption that things that are close to one another are more alike than those that are farther apart. IDW is a weighted distance average and so the predicted value is limited to the range of the values used in the interpolation.
Unlike other interpolation methods-such as Kriging,-IDW does not make explicit assumptions about the statistical properties of the input data. IDW is often used when the input data do not meet the statistical assumptions of more advanced interpolation methods.
IDW assumes that each measured point has a local influence that diminishes with distance. It gives greater weight to points closest to the prediction location, and the weight decreases as a function of distance raised to a power value (p=2 in our case). The search neighborhood can be altered by changing its size and shape and/or by changing the number of neighbors included. The maximum and minimum number of neighbor measures to include has been set and the neighborhood search has been divided into sectors to account for any directional autocorrelation or trend in the data [62]. probes, which were calibrated following the manufacturer's protocols. The data obtained were processed using Idronaut software in order to verify the quality check.

Nutrient analyses and multiprobe data acquisition
All the chemico-physical data were processed in order to determine means, median, standard deviations, and maximum and minimum values, and were graphically displayed as boxplots using the free PAST software version 2.06. Spearman correlation coefficients (r) indicate the strength and direction of a linear relationship between variables; r was considered significant when the p-value was < 0.05. The trophic state was calculated by applying the TRIX index [64]. This index combines nutrients (DIN as sum of N-NH4 + , N-NO2 − , N-NO3 − and TP expressed as μg l −1 of N and P, respectively), Chl a (μg l −1 ) and DO (absolute deviation from % saturation).

Ostracods evolution
52 species were identified in the examined area: 37 and 24 species in GTCrB and GTCrC respectively, while [46] found 25 species in the same area ( The Q-mode cluster analysis performed on the ostracod associations found in GTCrB and GTCrC reveals the presence of three groups of samples ( Figure 2).
The SIMPER analysis shows that more than 60% of the difference between clusters is defined by S. incongruens (24.4% relative contribution), A. convexa (9.9 %), C. neapolitana (9.9 %), L. ovulata neapolitana, C. whitei, L. ramosa and P. jonesii (Table 2).   The DIN: SRP molar ratio is commonly used to detect whether N and P act as factors capable of limiting primary production [67]. In this work the ratio was always higher than 16 suggesting that the system is P-limited [68].
Several criteria are commonly used to define the trophic state in aquatic systems. The TRIX index was set by [69] and according to these authors the quality varied from high, characteristic of a system with low productivity and low trophic level (TRIX: 2-4), to poor, typical of a highly productive system with high trophic levels (TRIX: [6][7][8]. In this work, TRIX ranged from 2.19 to 4.06 (average based on seasonal aggregated data), which is consistent with a low trophic level, especially during summer periods, and good water quality.
Pearson linear correlations between variables are shown in Figure

Discussion
Recent numerous works have highlighted the variations and impact on ecosystems of recent and growing urbanization at a global level. Urbanization may filter out species that are not preadapted to urban conditions, with a subsequent decrease in abundance or diversity at small (local) scale [8,70].
Alternatively, the loss of species less adapted to urban environments could be (over)compensated for by an increase in species efficient in exploiting urban resources [2,3,71]. Both phenomena may cause biotic homogenization if local communities are colonized by the same species, in turn increasing the compositional similarity of urban species assemblages and, consequently, reducing species richness of urban areas on a large scale [2,72]. The relationship between the growth of cities and the impact of human activities on adjacent marine areas is still currently difficult to analyze. [5,43] recorded how, in marine realms, the predominant cause of degradation noticed in microfossil records was nutrient enrichment and the resulting symptoms of eutrophication including hypoxia.
In particular, in the Adriatic Sea, changes were caused by eutrophication and anoxia due to human activities including agriculture, wastewater disposal, and diversion of river outflow [12,13].
In this sense, ostracods are usually intolerant to hypoxia and respond with a reduction in diversity and richness, and, in some cases, populations become monospecific [44,45].
Preprints (www.preprints.org) | NOT PEER-REVIEWED | Posted: 25 July 2020 doi:10.20944/preprints202007.0624.v1 In the northern Adriatic ecosystems and the GoT area, a review of numerous long term studies on river discharg, oceanographic features, plankton, fish and benthic compartments, collected since the 1970s revealed significant changes in mechanisms and trophic structures [53,73]. In detail, a gradual increase in eutrophication phenomena, characterized by significant hypoxic events at the bottom, was recorded during the 1970s until the mid-1980s [74,75,76], followed by a reversal of the trend, particularly marked in the 2000s [66,77]. This trend was attributed to a combination of the reduction in anthropogenic impact, mainly due to a substantial decrease of the phosphorus loads, and of climatic modifications, resulting in the decline in atmospheric precipitations and, consequently, of runoff [78,79]. However, the occurrence of significant atmospheric phenomena due to climate change are indicated by the increase in N-NO3 -and the decrease of salinity: long drought periods are followed by heavy rainfall that increase the Isonzo River discharge for short periods.   (Table S1). Among these C. subradiosa and Loxoconcha spp. are also recognized in the literature as among the most tolerant to environmentally stressful conditions [29,82]. Considering the length of time which passed between the samples being taken (1967,(2004)(2005)(2006)(2007)(2008)(2009)(2010)(2011)(2012)(2013)(2014)(2015)(2016)(2017), and in the absence of an intermediate sampling activities to indicate with certainty the precise evolution of ostracod associations over in the last 50 years, it is at any rate possible to highlight how, since the nineteen-sixties, ostracods have been exposed to a potentially compromised environment due to the possible increase in anoxia phenomena [83], with few prevailing species tolerant to environmentally

Conclusions
The Finally, it must be underlined how the best environmental conditions in recent samples have been recorded in MPA. This is not entirely surprising since meiofauna are commonly early colonists, and that the most mobile and sensitive taxon rapidly colonize sediments after favorable conditions are restored [84]. Therefore, results from our study indicate that the preservation of large and connected patches of natural habitats is the most effective measure to halt further urbanizationdriven biodiversity loss.
Future study on cores collected in selected areas of the GoT will improve understanding of the repercussions of anthropogenic activities over time on ostracod assemblages, as well as identifying additional indicator species, through seasonal sampling, that can be used to better define: a -possible causes of the recent decline in ostracod associations, b -the status and vulnerability of the ecosystem, c -to evaluate remediation activities to mitigate the negative impact of urbanization. Funding: This research did not receive external funding.