BIOGEOCHEMICAL APPROACHES TO ECOSYSTEM ENDPOINTS 79 – weakness of available data due to sampling and/or measurement problems, insuf- ficient time-series of data, lack of replication; – data gaps such as no measurements on baseline environmental conditions at a study site; – toxicological data that are extrapolated from high dose experiments to relatively low exposure; – natural variations in environmental parameters due to weather, climate, stochastic events. Consequently,risk assessment processis theobligatory continuation ofthe process of quantitative calculation and mapping of critical loads of sulfur, nitrogen and acidity at various natural and agricultural ecosystems. This is connected with numerous uncertainties a priori included in the computer algorithm for CL calculations: –atthe receptor selection step the uncertainty is related to the determination of the most sensitive receptor, which protection will definitely protect other, less sensitive, ecosystems; –atthe select environmental quality criteria step the uncertainty is connected with an assessment of biogeochemical structure of ecosystems and quantitative charac- terization of biogeochemical cycles of individual elements; –atthe select computer method (model) step the uncertainty is related to the appli- cability of steady-state models to dynamic systems requiring the definite simplifi- cation of these systems; –atthe calculate critical loads step the uncertainty is usually minimum and related mainly to the possibilities of modern computer pools; –atthe compare with actual load step the uncertainty is connected with an assess- ment of modern deposition and their spatial and temporal conjugation with definite ecosystems at the selected resolution scale. 1.2. Comparative Analysis of CL and ERA Calculations of Acidification Loading at Ecosystems The existing uncertainty at all steps of an algorithm for critical load calculation and mapping influences the probabilistic character of these values and arise the necessary to combineboth approaches. Thisis illustratedin Figure 2.In the maximum degreethe given conjugation is required at the risk management step in the ERA flowchart. The probabilistic approach to the critical loads of acid forming compounds allows us to apply the set of emission reduction scenarios to minimizing the financial investments for ecosystem protection (see Figure 3). Within the defined areas, critical loads are calculated for all major combinations of tree species and soil types (receptors) in the case of terrestrial ecosystems, or water biota (including fish species) and water types in case of freshwater ecosystems. 80 CHAPTER 4 These combinations include the great variety of different ecosystems, the sensitivity of which to both acidification and eutrophication inputs by atmospheric pollutants differs greatly, determining the necessary reduction needs when CLs are exceeded by modern deposition levels. This information on ecosystem sensitivity can be compared with a pollutant de- position map, to determine, which areas currently receive deposition levels, which exceed the area’s CL. The areas of “exceedance” indicate where present levels of pollutant deposition increase the risk of damage to ecosystems. 2. BIOGEOCHEMICAL ENDPOINT IN CRITICAL LOADS CALCULATIONS FOR HEAVY METALS At present, the calculation and mapping of critical loads for heavy metals is only at the beginning and in Europe there are only a few examples of application of methods described in Section 3.2. We will refer to case studies from Germany and Russia as the most characteristic research in this direction. The typical endpoints in these calculations refer to critical concentrations of different heavy metals in the ecosystems. The determination of the given critical concentrations is still uncertain and the relevant risk assessment calculated as an exceedance of critical loads should be based on selecting values of critical concentrations (see 3.2.2). 2.1. Calculation and Mapping of Critical Loads for HM in Germany We have seen that heavy metals can cause toxic effects to living organisms when critical limits are exceeded. Present deposition rates may cause the long-term accu- mulation of heavy metals in the soil, especially in the forest humus layers and bottom sediments. Calculations based on comprehensive models show the long-range trans- port of various heavy metals in regional and continental scale. In addition to atmo- spheric deposition, in agroecosystems the input of HMs is connected with phosphorus fertilizers and application of wastewater effluents. Under increasing acidification of the forest ecosystems, many heavy metals accelerate aqueous migration in the biogeo- chemical food web. The known example is related to Cd and Cu. The accumulation of pollutants in various terrestrial and aquatic ecosystems of Germany is almost non- reversible. That is why we will consider the precautionary measures based on critical load calculations of HM. The case study in Germany may give the best example of such an approach (Schutze, 2004). Methods of Critical Load Calculations for Heavy Metals The approach, similar to that described in Section 3.2 was applied for the calculation of critical loads of HM s for German soils. Critical concentrations as ecosystems endpoints According to the heavy metals’ effects, the soil microbes, crops and ground waters as a source of drinking water, are the most important receptors. During migration in the food web, the heavy metals, especially Cd and Hg, can affect also higher organisms, BIOGEOCHEMICAL APPROACHES TO ECOSYSTEM ENDPOINTS 81 including people. After consideration of different pathways, the most sensitive links of food webs should be chosen for establishing the critical concentration in the soil’s solution (critical limit) to protect all other pathways at this concentration. The Criticalconcentrations with respect to thesoil organisms shouldbe relatedto a loweffect level on the most sensitive species. The effects on the process ofmetabolism and other processes within the organisms should be considered and also the diversity of thespecies, which is most sensitiveto theheavymetals, hasto beaccounted. Critical limits must refer to the chronic or accumulated effects. For assessment of the critical concentrations in crops and in drinking water, human-toxicological information is required. In general, for establishing critical loads we should also account the additive effects of the different metals and combination effect between the acidification and biogeochemical mobilization of the heavy metals in soils and bottom sediments. The environmental standards based on total heavy metal concentration in the soil solution seem the most important criterion for the exposition of further compart- ments of the environment. The additional effects connected with metal speciation and complexations were not considered in the study. A Monte Carlo simulation is proposed to appreciate the uncertainty in the process of establishing the critical concentrations of heavy metals in the soil solution. Models In this case study, steady-state mass balance models are applied for critical loads calculation for the heavy metals. SM V +SM Dep +SM D +SM Abf = SM E +SM Aw +SM Er +SM G +SM Vorr . where SM V is release by weathering; SM Dep is input by the atmospheric deposition; SM D is input by “usual” fertilising; SM Abf is input by the use of waste; SM E is output by harvest; SM Aw is output by leaching; SM Er is output by erosion of the soil’s parts; SM G is output by degassing; SM Vorr is changes of the heavy metal pool in soil. This mass balance presents the possible links in the biogeochemical food web for various heavy metals. Some items may be neglected, like degassing of Pb, Cd, Cu and Zn metals. However, this process is of crucial importance for mercury (see Section 3.2). The output of the heavy metals with soil erosion may also be neglected. After elimination of these processes, the simplified following equation is workable. The sum of inputs by deposition, fertilizing, and waste and rubbish as fertilizer stands as the term Critical Load’. Thus, critical loads of any heavy metals may be calculated as follows: CL SM = SM E + SM Aw − SM V + SM Vorr . The mass balance model for calculation of critical loads for heavy metals includes the weathering process, the net removal through the crop biomass harvest, leaching, and also leaf uptake and litterfall. Using the simple dynamic way, the distribution between adsorbed and dissolved phases was accounted. 82 CHAPTER 4 Figure 4. Database for initial information for calculation of HM critical loads in Germany (Bashkin and Gregor, 1999). The uncertainties in the model inputs were elaborated using the statistic distribu- tion functions for the initial parameters and also the Monte Carlo simulation. The available information for calculation of critical loads of HMs in Germany is shown in Figure 4. This figure shows also the schematic algorithms for CL calcula- tions. The application of CL model and initial information allowed the researchers to map the critical loads of various heavy metals for different ecosystems. 2.2. Calculation and Mapping of Critical Loads for Cd and Pb in the European part of Russia Biogeochemistry of heavy metals has been extensively studied in the former Soviet Union due to a widespread environmental pollution. The numerous results on ecosys- tem sustainability or sensitivity to metal inputs have been accumulated. The assessment of ecosystem sustainability to the heavy metal loading includes primarily the estimation of soil compartment (Solntseva, 1982; Elpatievsky, 1994; Glazovskaya, 1997). These researches as well as literature data from other countries showedthat theprocesses ofmetal accumulation and transformation insoil andfurther migration in biogeochemical food webs, like metal uptake by plants and metal leach- ing from the soils, are mainly dependent on geochemical properties of the soils. The following soil parameters were shown as the most important: pH, organic matter com- position (mainly thehumic andfulvicacids ratio),redox reaction, andsoil granulomet- ric composition (Davies, 1980; Sanders, 1982; Kabata-Pendias, Pendias, 1984, 1992; Adriano, 1986; Balsberg-Pahlsson, 1989; Bowen, 1989; Temminghof et al., 1997). BIOGEOCHEMICAL APPROACHES TO ECOSYSTEM ENDPOINTS 83 Glazovskaya (1997) applied an analysis of geochemical conditions in different soils anddeveloped principlesfor assessingquantitatively thesustainability ofecosys- tems under the technogenic impact of heavy metals. The soils of the main natural zones distinguished on the East European plain area were combined into 6 groups ac- cording to their ecological-geochemical sustainability under HM loading (from “very sensitive” to “insensitive”). As shown in this research, most of the soils of the East European plain area have medium or weak sustainability under metal exposure. But quantitative parameters of HM impacts on the soil (including the permissible levels of metal depositions) were not considered in this classification. The quantitative assessment of biogeochemical mass–balances of the metals in various natural, urban and agricultural ecosystems were carried out in the different regions of the Russian Federation (Bashkin et al., 1992; Elpatievsky, 1994; Kasimov et al., 1995; Priputina et al., 1999, 2002, 2004a, 2004b). Methodologically, these researches are similar to the general approach used for calculations of HM critical loads in Europe (de Vries et al., 1998a, 1998b). However, the results of these local researches could notbedirectly usedforcalculations of HMcriticalloads for thewhole area of the East European plain due to scarcity of the data needed for computation according to the steady-state mass balance equation (see Section 3.2). Nevertheless, these datawere used for estimating andmapping of HM critical loadsfor theEuropean Russia area (Priputina et al., 1999; Bashkin, 2002). Preliminary calculating and mapping of critical loads for heavy metals (Pb and Cd) in the forest ecosystems of European Russia have been accomplished using a simplified version of the steady-state mass balance model (Priputina et al., 2002). For present study effect-based critical loads evaluated in accordance to the Guid- ance (de Vries et al., 2002) have been derived using critical limits of heavy metals concentration in the soil solution (0.6–1.0 mg m −3 and 6–10 mg m −3 for Cd and Pb, respectively). Input data applied for preliminary estimations of ERA endpoints included the parameters required for computing root uptake, leaching and weathering of heavy metals in different soil types (Priputina et al., 2003). The calculations of critical loads for lead and cadmium have been accomplished for the forest ecosys- tems of several key plots located in various natural conditions of the European part of Russia; background areas from north taiga to deciduous forests zone have been taken for these evaluations (Figure 5). Calculation Methods and Critical Limits In thisstudy two different endpoints have been selected:human healthaspects (critical limits based on drinking water quality) and ecotoxicological effects on biota (critical limits based on free metal ions concentration) (Priputina et al., 2004b). Two metal fluxes (net uptake in harvestable parts of plant biomass and leaching from the considered soil layer) are included in the mass balance equation (M&M Manual, 2004): CL(M) = M u + M le(crit) . (1) 84 CHAPTER 4 Figure 5. Location of case study ecosystems used in test of critical loads calculation in the European part of Russia. In forest ecosystems these symbols stand for: CL(M) is critical load of a heavy metal (g ha −1 a −1 ); M u is metal net uptake in wood biomass under critical loads conditions (g ha −1 a −1 ); M le(crit) is critical leaching flux of metal with drainage water (from the uppermost 10 cm soil layer) (g ha −1 a −1 ). In one’s turn, metal removal from the soil because of biomass uptake is accounted in the following way: M u = f Mu ∗ Y ha ∗ [M] ha (2) where Y ha is yieldof harvestable biomass(woodbiomass netproduction) under critical load conditions (kg dw ha −1 a −1 ); [M] ha is content of the metal in the wood (g kg −1 dw); f Mu is the fraction of metal net uptake within the considered soil depth (z b or z); root uptake factor, f Mu , is assumed to be equal to 1. The critical leaching flux of HM can be calculated by the equation: M le(crit) = c le ∗ Q le ∗ M ss(crit) (3) where Q le,zb is leaching flux of water from the topsoil (the uppermost 10 cm soil BIOGEOCHEMICAL APPROACHES TO ECOSYSTEM ENDPOINTS 85 layer) (m a −1 ); [M] ss(crit) is critical limit for metal concentration in the percolating soil solution (mg m −3 ); c le is 10 g mg −1 m 2 ha −1 , factor for appropriate conversion of flux units. The annual mean water percolation Q le,zb is determined by the long-term mean annual temperature (mainly determining the potential evapotranspiration, E pot ) and precipitation (mainly influencing the actual evapotranspiration, E act ) according to Q le,zb = P m − f E,zb ·(P −2 m + (e (0.063·T m ) ·E m,pot ) −2 ) −1/2 (4) where P m is annual mean precipitation (m a −1 ); T m is annual mean air temperature ( ◦ C); E m,pot is annual mean potential evapotranspiration in humid areas at T m = 0 ◦ C; E m,pot ≈ 0.35 m a −1 in forests, possibly less in other terrestrial ecosystems; f E,zb is the fraction of total annual mean evapotranspiration above z b (·); f E,zb ≈ 0.8 for the organic top soil layer of forests. Critical concentrations of HMs in the soil solution, [M] ss(crit) , depend on the target to be protected (ERA endpoints). These values have to be derived from critical limits. Parameters of critical limits used in the calculations are presented in Table 1. Input Data The values of output metal fluxes mentioned above vary as a function of spatial distributed parameters including climate, soil and forest-type data. As a basis for computing critical loads, an overlay of three maps was made: r FAO soil map; r Runoff data map; r Forest-types map generalized from Land use map. Table 1. Overview of critical limits used for calculating critical loads in case study plots. Endpoints Indicator/critical limit Pb, mg m −3 Cd, mg m −3 Human health effects Total HM concentration in soil water below root zones (aiming at ground water protection) 10 3 Ecosystem functioning Free metal ion concentration in soil solution re-calculated to total dissolved metal concentration in soil drainage water (in view of effects on soil microorganisms, plants and invertebrates) 1.7–20.4 ∗ 1.3–3.2 ∗ ∗ Values accounted using Look-up tables (Modelling and Mapping Manual, 2004) 86 CHAPTER 4 Figure 6. Diagram of metal fluxes and input data included in calculation methods. Soil-related data (HM and BC content in soil parent materials) were included in calculations to account the values of HM weathering. Also we considered the influence of soil types on forest biomass productivity. Runoff data (at scale 0.5  × 0.5  ) were directly used to get input data on drainage water fluxes, Q le . Forest-type- related data (wood biomass growth and HM content in wood biomass) inserted into our database were subdivided depending on either coniferous, deciduous or mixed forests. Structure of thedatabaseand input dataneededfor inclusioninthe present database are shown in (Figure 6). Data on water leaching fluxes have been calculated using iteration approaches (Priputina, 2004). Water percolation parameters have beenaccounted (Manual, 2004). Annual mean air temperature and precipitation data have been obtained from IWMI World Water and Climate Atlas (2002). Two iteration versions of the map of water leaching parameters are shown in Figure 7. The data needed for estimating metal removal with harvestable part of wood biomass have been obtained from literature as well as from sampling and simulating studies. As a rule, data on yields of forests can be accounted either from forest service statistics or using special calculations when forest potential growth is estimated as a function of climate, soil and tree type characteristics (Reinds et al., 2001). However forest statistical data are not spatially distributed in most areas of Russia. Also, there are many uncertainties in estimating potential bioproductivity of the forests. That is why we used simulation procedures to account annual mean growth of wood biomass in the forests of case study plots (Table 2). Model EFIMOD (Chertov, Komarov, 1997) has been applied to compute data on annual increase of wood stock in stems and large branches of main tree types widespread in the forests of the European part of Russia. We assume that these data after some improving and completing could be applied in the national database. As a cartographic layer a generalized version of the map of forest tree dominants is used (Figure 8). BIOGEOCHEMICAL APPROACHES TO ECOSYSTEM ENDPOINTS 87 Figure 7. Spatial distribution of the data used for estimating water leaching fluxes (m a −1 ): on the left—runoff data only (iteration I), on the right—data calculated from climate parameters (iteration II). Table 2. Parameters of wood biomass growth for main tree types in the forests of European Russia; results of simulating based on EFIMOD (Chertov, Komarov, 1997). Wood biomass growth, kg m −2 (dw) Type Location Stem Branches Total Pine (Pinus silvestris) Northern regions 0.35–0.38 0.06–0.08 0.4–0.45 Central regions 0.35–0.4 0.08–0.1 0.45–0.5 Spruce (Picea abies) Northern regions 0.15–0.16 0.03–0.04 0.18–0.2 Central regions 0.5–0.52 0.08–0.1 0.6–0.62 Birch (Betula pendula) Northern regions 0.1–0.12 0.01–0.02 0.11–0.15 Central regions 0.3–0.35 0.03–0.04 0.35–0.4 Oak (Quercus robur) Central regions 0.25–0.26 0.05–0.06 0.3–0.32 Southern regions 0.25–0.3 0.05–0.8 0.3–0.4 88 CHAPTER 4 Figure 8. Forest-type-related cartographic data: on theleft—Land use map(IGBP Map ofEDC DAAC, 1997, see de Vries et al., 2002) applied in calculations’2002; on the right—fragment of the map of forest tree dominants (National Atlas, 2003, see Priputina, 2004b). Since reference national data on Pb and Cd content in the harvestable part of forest biomass (stems and branches) are very rare, sampling of stem wood has been carried out in the background forested areas of European Russia. The results are presented in Table 3. The whole set of data from this table illustrates median values of Pb content in wood biomass for all tree types, which are lower than 1.0 mg/kg (dry weight). Minimal values of average concentrations were revealed in Pine tree (Pinus silvestris). In the same regions (sampling plots) average values of Pb accumulation in spruce species (Picea abies) are higher than in pine trees. But, conjugated analysis of distribution of Pb content values in the wood of both Spruce and Pine trees did not reveal evident differences between two groups of coniferous trees (Figure 9). Median (average) values of Pb concentrations in deciduous trees are higher in comparison with coniferous ones. Maximum values of average Cd concentrations are revealed in stem wood of the Aspen tree (Populus tremula). Median values of Cd content in other types of both deciduous and coniferous trees are lower than 0.2 mg/kg (Table 3, Figure 10). [...]... +2.40 −0. 23 −0.55 Neoplasm +1. 53 +0 .37 −0. 83 Endocrine and immune −0.82 −0.05 +0.49 Thyroid gland −0. 73 −0.91 +1.09 Blood and hemophoietic organs −0.55 −0.54 +1.41 Nervous system and organs of sense −0.89 +0.28 +0.10 Cardiovascular systems −0.78 −0.20 −0.19 Peripheric nervous system −1.64 −0.49 +0. 63 Systems of the blood circulation +3. 43 −0.26 −0. 23 Chronic rheumatic heart disease −0.70 −1 .30 +0.68... Eurasia Limit concentrations, ppm Trace nutrient Number of samples Mean content in pasture crops Mn 819 73. 0 Zn 519 Cu Lower (deficit) Optimum for animal organisms Upper (excess) 30 21.0 60 937 6.4 12 Mo 537 1.25 >0.2 0.2–2.5 >2.5 Co 859 0 .32 1.0 I 39 7 0.18 1.2 According to the specific characteristics of North Eurasian biogeochemical provinces,... 186.8 117.6 137 .7 140.0 130 .2 2 13. 8 Stomach cancer — 12.4 32 .6 23. 6 19.7 12.1 19.1 32 .7 Esophagal cancer — 28.1 86.1 51.8 56 .3 52.9 49.6 69.5 Lung cancer — 4.1 9.5 8.1 9.1 7.9 13. 0 21.4 BIOGEOCHEMICAL APPROACHES 107 Table 8 Growth of cancer diseases in Kazakhstan in 1970–1990, % of 1970 rate Administrative region Lung cancer Breast cancer Skin cancer Intestinal cancer Limphomas Guriev 1 43 257 209 54... soils of North Eurasia Limit concentrations, ppm Trace nutrient Number of samples Deficit/lower Optimum Excess/upper Co 2,400 30 Cu 3, 194 60 Mn 1,629 70 4 6 30 >30 B 879 . Excess/upper Co 2,400 <7 7 30 > ;30 Cu 3, 194 <15 15–60 >60 Mn 1,629 <400 400 3, 000 > ;3, 000 Zn 1,927 < ;30 30 –70 >70 Mo 1,216 <1.5 1.5–4 >4 B 879 <6 6 30 > ;30 Sr 1,269 <600. (excess) Mn 819 73. 0 <20 20 30 > ;30 Zn 519 21.0 <20 20–60 >60 Cu 937 6.4 < ;3 3–12 >12 Mo 537 1.25 >0.2 0.2–2.5 >2.5 Co 859 0 .32 <0.25 0.25–1.0 >1.0 I 39 7 0.18 <0.07. 0.01–0.02 0.11–0.15 Central regions 0 .3 0 .35 0. 03 0.04 0 .35 –0.4 Oak (Quercus robur) Central regions 0.25–0.26 0.05–0.06 0 .3 0 .32 Southern regions 0.25–0 .3 0.05–0.8 0 .3 0.4 88 CHAPTER 4 Figure 8. Forest-type-related