Messa a punto di un metodo di valutazione dello stato dei piccoli corsi d'acqua urbani del Lazio centrale basato sui macroinvertebrati bentonici

The process of urbanization in the central portion of Latium, tied to the development of the city of Rome, is rapidly modifying the territory and its use. Subsequently natural environments and agricultural landscapes are constantly being replaced by areas characterized by great anthropization. Studies undertaken in different urban areas have demonstrated that these very changes lead to a multitude of alteration factors which, in turn, have a profound and varied impact on the ecosystems connected to small watercourses and, in particular, their biological components. The recent enacting of the European Water Framework Directive 2000/60 has established that biological communities must be considered the very fulcrum on which tutelage and management strategies of aquatic ecosystems must be centered. To this day, the communities most frequently adopted as ecological indicators for the assessment of watercourses are certainly benthic macroinvertebrates. These organisms have therefore been extensively investigated and there are studies centered on the overall community as well as on specific taxa such as Chironomidae, Ephemeroptera and Simuliidae. The present study wishes to develop a status assessment method, based on benthic macroinvertebrates, for the urban streams of central Latium; more specifically, the different tasks of study were to: characterize the abiotic conditions of the territory’s streams; quantify the macrobenthonic communities’ reactions to the changes determined by the urbanization gradient; test two different methodological approaches based on macroinvertebrates, which respectively adopt multimetric analyses and self organizing neural nets, so as to develop a methodology for the assessment of the watercourses of the area addressed; test the efficacy of the preimmaginal stages of the Simuliidae family as environmental indicators of the roman territory’s streams. First of all, 105 sites were characterized using cartographic analysis, on site observations, chemical and physico-chemical analyses of water samples, and benthic macroinvertebrate sampling. Black fly larvae and pupae were selectively collected in all 105 watercourses; in a chosen group of these, sampling was repeated every two months for an entire year. The data thus collected enabled a pre-classification of the sites, 15 of which were chosen so as to cover a wide array of conditions along the urbanization gradient present in the area considered by this study. These 15 sites were then investigated further through three seasonal sampling campaigns (conducted respectively in autumn 2007, winter 2007-2008 and spring 2008) which enabled the collection of ulterior water samples for chemical analyses and of macroinvertebrates for a quantitative evaluation of the communities (proportional multihabitat method). Chemical, physico-chemical and microbiological analyses indicate the presence of diffused watercourse pollution in areas with altitudes below 200 meters a.s.l.. It especially manifests itself with high concentrations of nutrients (nitrates: 7,71±3,59 mg/L; phosphates:1,27±1,20 mg/L in average). In accord with previous studies, the present results indicates that, in urban areas, the concentrations of phosphate ions and nitrates may be even higher than those typically found in the water sources of agricultural areas (in average respectively 8,9 mg/L and 7,5 mg/L). The extremely high concentrations may supposedly be consequence of the introduction into the water system of untreated domestic wastewaters from a multitude of sources. The amount of phosphates may, moreover, be caused by the mobilization of sediments deriving from the transformation of agricultural areas into building lands. The use of the IFF index also indicates that the functionality (self-depurant capacity) of the small watercourses of Rome’s urban area has been strongly impaired: 78% of the sites show values below the minimum threshold for mediocrity) by different factors: the removal of trees in the surrounding territory; the management of the river-side vegetation which solely wished to guarantee the hydraulic functionality of the draining system; the trivialization of the micro-habitats due to the input of fine sediments deriving from the scouring of waterproof grounds; the repercussions of different contributory factors on the biological communities. The most evident consequence, on the existing macroinvertebrate communities, of the urbanization of the territory was the loss of taxonomic variety. While, on average, 29 taxa were found in natural environments, the average was of only 4 in areas with urbanization above 40%. In such areas the communities are characterized especially by Chironomidae and Oligochaeta, the first of which, abundance wise, constitute approximately 90% of the overall communities. Plecoptera seem to be particularly sensitive and they disappear almost completely in sites with urbanization above 10% whereas the number of taxa and the relative abundances of both Ephemeroptera and Trichoptera show drastic reductions. Multivariate analysis (Principal Components) was conducted so as to identify the axes along which the community data showed greatest variation. The first axis, by definition the one capable of explaining the greatest variation, resulted highly correlated to ecological quality. An ulterior multivariate analysis (two way indicator species analysis-TWINSPAN) was conducted to group the samples on the basis of both the composition and the abundance of determined indicator taxa. These analyses enabled the identification of the descriptive metrics that could best simulate the gradient along the first axis and best discriminate among the different groups. The integration of the selected metrics enabled the formulation of a multimetric index. The data was also used to implement a self-organizing neural net (SOM), which is a classification technique, based on unsupervised learning, capable of reducing the number of dimensions of the overall data and of giving a visual representation of the results by plotting them onto a map. A U matrix was used to estimate the distances between the different neurons on the map. The inclusion of the TWINSPAN-indicator taxa in the map enabled the assessment of each group’s specific level of “environmental quality”. Ulterior data, not included in the development of the two methods, was adopted to test classification efficacy. The results indicate black flies may indeed be efficaciously adopted, in the area object of the present study, as environmental indicators to evaluate the ecological conditions of the watercourses. The comparison of species sampled and those attested in literature enabled to note the absence, in the samples collected, of certain species (Simulium reptans, S. vernum, S. liriense), perhaps due to both pollution of the watercourses and changes in use of the surrounding land. Correspondence analysis enabled the identification of altitude as the main factor influencing species’ distribution in the different sites considered. However, quality becomes the main factor when taking into account exclusively the sites in the plains: S. ornatum s.l. and the “equinum” species (S. equinum, S. pseudequinum, S. lineatum) are the ones that show the greatest tolerance to chemical water pollution. Furthermore, when urbanization alters both the chemistry of the waters and the condition of the surrounding territory, S. ornatum s.l. is the sole of these species still present in the streams; it too, however, is incapable of enduring toxic pollution and suspended solids. The mid-high course of the Aniene river and its tributaries were sampled every two months, enabling the confirmation of the spatial-temporal black fly successions observed in the central Apennine region (area belonging to the “hirtipes” group, to the “variegatum” group and to a mixed one with more species) and reported in literature. The results also indicate the larvae of S. ornatum may supplant the other species of black flies in the periods in which there is greater pressure on watercourses. This study made it possible to test three different approaches to stream assessment in the roman territory. The consistence of the two community based methodologies provides higher validity to the obtained classification. Furthermore, the black fly study results, through preliminary, open up to the use of this organisms in the development of metrics useful in the analyses of watercourses’ quality. The enhancement of study and monitoring activities, propedeutic to the beginning of biological restoration measures, can represent the basis for the tutelage and future recovery of the urban streams