1. GLI IMPATTI DELL’INQUINAMENTO ATMOSFERICO SUGLI ECOSISTEMI NATURALI E ANTROPICI 23 - 24 Gennaio 2014 ENEA, Via Giulio Romano, 41 – Roma Marcello Vitale, Fabio Attorre, Michele De Sanctis Dipartimento di Biologia Ambientale Sapienza Università di Roma Con i contributi di: Maria Francesca Fornasier - ISPRA, Roma Alessandra De Marco - ENEA, CR Casaccia, Santa Maria di Galeria - Roma) Topic: I carichi critici per la valutazione degli impatti sulla Biodiversità Carichi critici, eccedenze e biodiversità: sviluppo di nuovi approcci e metodologie

2. Nitrogen emissions and deposition of nitrogen compounds have decreased since 1990 but relatively little compared to sulphur emissions. Agriculture and transport are the main sources of nitrogen pollution (EEA, 2007). In addition, nitrogen components can lead to eutrophication of ecosystems. When this pollution exceeds certain levels ('critical load'), it is damaging to biodiversity. Critical load exceedance is still significant (1). (1) The critical load of nutrient nitrogen is defined as 'the highest deposition of nitrogen as NOX and/or NHY below which harmful effects in ecosystem structure and function do not occur according to present knowledge' (ICP, M&M, 2004). Rationale

3. Available at: www.eea.europa.eu/publications/eea_report_2007_2.www.eea.europa.eu/publications/eea_report_2007_2 A critical load is defined as “a quantitative estimate of an exposure to one or more pollutants below which significant harmful effects on specified sensitive elements of the environment do not occur according to present knowledge” (Nilsson and Grennfelt, 1988). Exceedances of critical loads by current or future nitrogen loads indicate risks for adverse effects on biodiversity. Rationale

4. Excess nitrogen is one of the major threats to biodiversity. Excessive levels of reactive forms of nitrogen in the biosphere and atmosphere constitute a major threat to biodiversity in terrestrial, aquatic and coastal ecosystems. On land it causes loss of sensitive species and hence biodiversity by favouring a few nitrogen tolerant species over less tolerant ones. In coastal waters it leads to algal blooms and deoxygenated dead zones in which only a few bacteria may survive. Questions Is it possible to evaluate a plant population changes induced by climate and air pollutants into the plant community? How are actual biodiversity indices useful for this assessment, taking into account limited database either in time or in space? Rationale

5. Study areas

6. Kg/ha/y Test sites locationNitrogen depositions in Italy

7. Significant differences among N exceedances (above, BOF N and under BSC N canopy) of the monitored sites have been found by ANOVA test * * * * * * Study areas

8. Temporal trends of exceedances above (BOF N) and under canopy (BSC N) calculated for the three test sites Critical Loads for the three test sites Study areas

9. LOM1 Study areas

10. The values ​​ of the diversity indices are not always comparable between them and depend on the extent to which they can actually vary. The evenness is a measure of diversity normalized on a fixed scale (e.g. from 0 to 1) and allows to carry out these comparisons. From the Shannon-Wiener index, the Pielou’s evenness can be defined as:

12. Ellenberg, in a series of publications (Ellenberg 1979, 1988; Ellenberg et al. 1991), defined a set of indicator values for the vascular plants of central Europe. These have been widely used, both in central Europe and in adjacent parts of western Europe. The basis of indicator values is the realised ecological niche. Plants have a certain range of tolerance of temperature, light, soil pH, and so on. If we wish to make inferences about the ecological conditions pertaining at a site, much useful information can be obtained from the flora. These values are not i.e. mean pH values, but are on an arbitrary scale reflecting soil pH though not directly based on measurements. However, an advantage of indicator values is that they may be more sensitive to the requirements of plants than is a selected physical variable.

13. * * * * * * * * * * * *

14. * * * * * *

16. Site=LOM1 Correlations Marked correlations are significant at p <.05000 N=9 H'JH'dune H'dune std Exceed BSC N 0.270.26 0.27 Exceed BOF N 0.330.490.59 0.70 Ellen. Light 0.100.06-0.39-0.41 Ellen. Temp -0.03-0.150.09-0.02 Ellen. Cont 0.170.29-0.27-0.15 Ellen. Soil Moist 0.070.100.070.12 Ellen. Soil pH -0.16-0.320.07-0.07 Ellen. Soil N 0.01-0.060.260.20 Site=ABR1 Marked correlations are significant at p <.05000 N=10 H'JH'dune H'dune std Exceed BSC N -0.050.300.400.36 Exceed BOF N -0.39-0.400.010.06 Ellen. Light 0.120.410.250.21 Ellen. Temp -0.000.16-0.15-0.14 Ellen. Cont -0.22-0.05 -0.04 Ellen. Soil Moist -0.130.190.01-0.02 Ellen. Soil pH 0.000.12-0.13-0.16 Ellen. Soil N 0.600.570.250.17 Site=LAZ1 Marked correlations are significant at p <.05000 N=10 H'JH'dune H'dune std Exceed BSC N 0.510.49-0.08-0.16 Exceed BOF N 0.510.440.310.21 Ellen. Light -0.56-0.47-0.24-0.13 Ellen. Temp -0.67-0.72 0.290.39 Ellen. Cont -0.90 - 0.88 0.010.16 Ellen. Soil Moist 0.780.75 0.08-0.05 Ellen. Soil pH 0.260.29-0.34-0.37 Ellen. Soil N 0.210.16-0.27-0.30 Correlation analysis showed that generally diversity indices did not correlate with the Ellenberg’s indicators except H’ in the LAZ1 site, but H’ dunestd correlated with N exceedances above canopy in LOM1.

17. The application of more advanced statistical analyses (such as the Random Forest in the Data Mining statistical tools) did not result into a clear causal pattern between biodiversity indices and environmental and pollution factors. Likely this was due to a very poor dataset. NoxNred % of the total It should be taken into consideration that N depositions may have different distributions along the peninsula regard their oxidative state and, as a consequence, different effects on vegetation (De Marco et al., 2013) De Marco, Screpanti, Attorre, Proietti, Vitale (2013) Assessing ozone and nitrogen impact on net primary productivity with a Generalised non-Linear Model. Environmental Pollution, 172: 250-263.

19. REMARKS These simple analyses highlighted that different diversity indices (one based on the species proportion and the other based on proportional coverage area) could suggest different trends, which were not in correlation with climatic data and N depositions for three herbaceous communities growing under different ecological niches; It seems that one decade of data it is not sufficient for assessing a change in herbaceous community’s composition; Inferences made on the air pollutant-induced effects in affecting plant community’s composition should be carried out with extreme caution;

20. Wide database with plant community information, air pollutant and climate should be used for non linear statistical analyses, in order to provide inferring statistical models; The spatial pattern of plant distribution should be taken in consideration when plant diversity indices are applied for different monitoring plots. Actually, this structural parameter is not considered; The application of different biodiversity indices and ecological indicators should be applied in extensive way to other European herbaceous plant communities, in order to assess if air pollutants, and/or climate, and or anthropogenic activities are causal effects for a changing plant communities. REMARKS

21. Grazie per l’attenzione !