Flows of Chemical Substances in Latvian Pine Forests Ecosystems FACULTY OF GEOGRAPHY AND EARTH SCIENCES DEPARTMENT OF ENVIRONMENTAL SCIENCE
UNIVERSITY OF LATVIA FACULTY OF GEOGRAPHY AND EARTH SCIENCES DEPARTMENT OF ENVIRONMENTAL SCIENCE
Flows of Chemical Substances in Latvian Pine Forests Ecosystems
29 Introduction Actuality of the work Ecosystems of forests are under continuously impact from substantial natural and anthropogenic influence. One of major factors affecting the function of ecosystems is environmental pollution and flows of nutrients from the atmosphere. Studies into the effects of atmospheric deposition on ecosystems started in the 1960's. The prevailing assumption at that time was that atmospheric deposition was an additional source of nutrients to ecosystems and therefore beneficial (Johnson and Lindberg, 1992). Later studies indicated that acid deposition actually caused a net loss of nutrients from forests ecosystems and caused soil acidification (Cowling, 1982; Cowling and Nilsson, 1995). In areas where the supply of nutrients from the soil by weathering is low, precipitation is the main source of nutrients to the ecosystem (Parker, 1983). The greatest number of studies about cycling of substances, including pollutants in forest ecosystems have concerned boreal and nemoral zones. Still a relatively small number of studies have been conducted in the boreo-nemoral zone that unlike the above-mentioned zones, has a different climate and length of vegetation season. The territory of Latvia is interesting as an area of study due to a rapid decrease of atmospheric pollution after the restoration of independence, and thus a decrease of substance flows in forest ecosystems from the atmosphere. The results of substance flows in ecosystems of pine forests, reviewed in the doctoral thesis, provide new well-grounded scientific information about cycling of Chemical elements and substances in pine forests ecosystems of the boreo-nemoral zone during the period when atmospheric pollution decreased. The results of the study provide a possibility to improve methods used in forest monitoring and integrated monitoring. Basic information sources As a basis of the study, data from integrated monitoring stations of Latvia have been used, at the same time performing more detailed research of separate nature components. An integrated monitoring programme in Europe was developed in 1992 involving 22 countries. Latvia has participated since 1994. The network of integrated monitoring consists of confluence basins of small water courses to provide a possibility of modeling water amount and its Chemical composition changes in forest ecosystems as a result of atmospheric pollution. Two integrated monitoring stations, in Rucava (Liepājas region) and Taurene (Cēsu region), have been established in Latvia. The Rucava catchment receives higher loads of long distance transboundary air pollution arriving from Central and Western Europe.
Integrated monitoring in Latvia includes 25 sub-programmes (http//:www. meteo.lv). The data of seven sub-programmes were used in the doctoral thesis:
The objective of the study is to study turnover of substances in ecosystems of pine forests in the boreo-nemoral zone when atmospheric pollution has decreased. Main tasks of the work:
Novelty of work
Approbation of doctoral work results: The results of the doctoral work are summarized in 3 scientific publications and 7 conference theses. Reports on study results were presented at 5 international conferences and 3 Latvian conferences. The results have been published in:
Conference abstracts
32
Reported conferences:
33 • University of Latvia 62 scientific conference. Latvia, Riga February, 2004. Oral presentation „Ķīmisko vielu izsēšanās ar nobirām priežu mežu ekosistēmās Latvijā".
.54 Literature Review Small particles and gases of both biogenic and anthropogenic origin are emitted to the atmosphere and deposited to ecosystems as wet and dry deposition (Bredemeier, 1988). Traditionally studies on precipitation Chemistry have focused on S and N compounds, which have an acidifying effect, and basic cations (Ca2+, Mg2+, K+), which have neutralizing effect on the environment (Ulrich, 1983; Ukonmaanaho et al., 1998). Nutrients are returned to the soil for recycling via two major pathways: litterfall and throughfall. In forests ecosystems rain water passing through a forest canopy changes substantially not only its quantity, but also its quality (Lövblad et al, 1994; Ukomnaanaho et al, 1998). Throughfall usually consists of 90-60% of the precipitation depending of the characteristics of forest stand (species, canopy etc.). The growth and productivity of forests ecosystems mainly depend on the amount, the nature and the rate of decomposition of forest litterfall (Kavvadias et al., 2001). As in forest ecosystems pollutants and nutrients from precipitation and dry deposition are firstly collected with forest canopy, the litterfall falling on the soil each month contains Chemical substances accumulated in needles and leaves of trees. Litterfall supplies more matter to the forest floor than throughfall but the nutrients in litter are released through decomposition and mineralization while throughfall nutrients are nearly all dissolved and readily available for uptake (Ukonmaanaho, 2001; Pajuste, 2004). Atmospheric pollutants have also an important role in forming the Chemical composition of litterfall, determining the content of heavy metals and other Chemical substances in needles and leaves. Small amounts of heavy metals are always present in the environment, but their increased concentrations are known to cause disturbances in all living organisms, and can result in decreased litter decomposition and subsequently slower nutrient cycling in the whole ecosystem Pennanen, 2001). The Chemistry of the soil water reflects atmospheric inputs and a number of processes in the soil, including weathering, ion exchange, mineralization and immobilization (Ulrich, 1983). The amount of throughfall and its quality affect the properties of the top soil and availability of nutrients in the soil (Bird and Bresolin, 1998). Changes of soil acidity (H+) and deposition of sulphur, nitrogen and basic cations may substantially impact the cycling of nutrients. Considering that in most forest ecosystems deposition of N still is noticeably high, we expect that in the future the most part of forests ecosystems shall reach
35 Saturation with nitrogen and N leaching with all its harmful results (Eichhorn et ah, 2001), for instance, soil acidification, leading to leaching of base cations and aluminium from the soil (Stoddard, 1994) and increased movement of nitrogen compounds and base cations into surface waters (Beier et ah, 2001). Deposition of basic cations, in its turn, decreases the acidifying effect of S and N in forest soils (Nissinen and Ilvesniemi, 1990; de Vries et ah, 1995). Soils have a noticeable ability to accumulate heavy metals (Merrington and Alloway, 1997). Heavy metal participation in biological cycling of substances is affected with the ability of the plant-root system to uptake them from the soil solution and with the degree of how hard these substances are captivated by the soil absorption complex (Smith, 1995). Metals in forest soils demonstrate two distinct patterns of release. In the first pattern a crucial role is played by soluble organic acids. They release in great quantities when the organic matter in the top soil is mineralized. The second pattern is associated with soil acidity. The more acidic the soil is, the greater the amount of metals is released (Bergkvist et al., 1989). The relative sizes of the throughfall, stemflow, litterfall and soil water fluxes that define biogeochemical cycles depend on several interacting factors, including levels of deposition, stand composition, structure and age, and soil type, and differ for each nutrient in the distribution and amount of rainfall (Parker, 1983).
Material and Methods The study used data from integrated monitoring (IM) performed in Latvia by the Latvian Environment, Geology, and Meteorology Agency in cooperation with the Faculty of Geography and Earth Sciences of the University of Latvia. The location of IM catchments, the sampling procedures and analysis methods for bulk precipitation, throughfall, stemflow, forest litterfall, needles, soil and soil water followed the standard methods of the Integrated Monitoring Manual (Manual for Integrated Monitoring, 1998). In addition to the IM sub-programme of litterfall chemistry, monthly samples of litterfall were analyzed. Biogenic elements (Ca, Mg, К, Ntot) and concentrations of Pb and Zn in 1M BaCl2 and HN03 extract were determined by atomic absorption spectrometry for mean samples of each month. Data handling and statistical analysis Annual precipitation and soil water mean concentrations were calculated for each site. Annual deposition values were calculated by multiplying the concentrations from collected samples by the corresponding amount of precipitation (in mm). The output fluxes of chemicals from the soil were calculated from soil water concentrations and water flow values (l/s/km2). The pH values were converted to H+ before calculation of the mean pH value. In the study mean annual and monthly concentrations of Chemical elements in litterfall and mean annual concentrations in needles and soil were used. Litterfall fluxes were calculated by multiplying elemental concentrations by litterfall mass fluxes (kg). The nonparametric Mann-Kendall test was used for detecting trends in the time series. Paired-Samples T-test was used for testing differences between mean annual concentrations in bulk precipitation, throughfall and stemflow, and for testing differences between mean concentrations in first (C), second (C+l) and third (C+2) year needles samples and for testing differences between mean concentrations in 0, E and В horizon of soil. The Mann-Whitney U-test was used for testing differences between IM plots and seasonal differences of elements concentration in litterfall. Spearman's rank correlation was used for testing differences between throughfall and soil water. A significance level of/КО. 05 has been chosen to represent a statistically significant difference between data sets. Programs MS" Excel and SPSS 15.0 for Windows were used in data processing.
37 Results and Discussion Input of Chemical substances and elements in forest ecosystem with precipitation Sulphate and nitrate in the atmosphere can have significant impact on the pH of precipitation. Atmospheric deposition of S and N compounds in pine forests has decreased in Latvia. During the same time, a substantial decrease in precipitation acidity was observed. The mean annual pH of bulk precipitation (BP) during the period of 1994-2004 varied from 4.5 (± 0.9) in 1996 to 5.9 (±0.8) in 2004 at the Rucava IM site and from 4.8 (± 1.0) in 1997 to 5.8 (± 0.7) in 2002 at Taurene IM site. Similarly, pH significantly increased also in throughfall (TF). The mean annual pH of TF at Rucava increased from 4.4 (± 0.4) to 5.4 (± 0.8), but at Taurene from 4.5 (± 0.5) to 5.8 (± 0.7). The largest changes were observed in the western part of Latvia, reflecting a decrease of long-range transboundary air pollution impact from Western Europe. Changes in precipitation acidity are not characteristic for stemflow (SF) waters. SF average pH parameters in Rucava per years varied from 3.9 (± 0.2) to 4.2 (± 0.3), and in Taurene from 3.8 (± 0.1) to 4.4 (± 0.7). This indicates an important role of dry deposition of pollutants and biological processes in forming Chemical composition of stemflow water. SF waters at both monitoring stations are more acid than BP and TF waters. During the study period at IM stations in Latvia S042--S ion concentrations (Fig. 1) in BP, TF and SF showed significant negative linear trends (Tērauda and Nikodēmus, 2007). The most rapid decrease of S042--S ions concentration in SF compared with BP can be explained by a decrease of pollutant dry deposition and accumulation on stems of trees (Tērauda and Nikodēmus, 2007). The concentrations of S042--S in Rucava and Taurene increased in the order: BP<TF<SF. The S042- enrichment of TF and SF is primarily the result of wash-off of accumulated dry deposition (Lindberg and Lovett, 1992). The deposition of S042-S derived from BP and TF did not change during the monitoring period. The only statistically significant decreasing trend was found for S042--S deposition in BP at Taurene, where the annual mean S042--S of precipitation dropped from 9.3 kg/ha to 3.2 kg/ha per year. The SF volumes were very small, representing only 0.5% in Rucava and 0.4% in Taurene of the rainfall, and therefore were not considered further in deposition calculations. During the observation period, the concentrations of nitrate ions in precipitation both in Rucava and in Taurene IM stations showed decreasing trends. However, the decline in NO3-N in neither types of precipitation were significant, except in SF at Rucava (Tērauda and Nikodēmus, 2007). The deposition fluxes of nitrate 38 were significantly greater in Rucava compared to Taurene but the differences between BP and TF were not statistically significant. n - number of observations Figure 1. S042-S concentrations in bulk precipitation (BP), throughfall (TF) and stemflow (SF) at the Rucava (a) and Taurene (b) IM stations Similarly as in studies conducted in Finland and Estonia, the essential plant nutrients calcium, magnesium and potassium concentrations and deposition with precipitation did not show significant decreasing or increasing trends during the study period at both IM stations in Latvia. In general, mean concentrations of the base elements in BP, TF and SF during the study period were higher at Rucava IM station than at Taurene IM station.
39 The significant increase of cation concentrations in throughfall may be due to leaching or exchange processes and washing off of dry deposition from foliage, branches and stems (Pajuste, 2004). The deposition fluxes of base cations were significantly higher in TF (1.7 times for Ca, 2.1 times-Mg and 2.4 times-K higher at Rucava IM and 1.4 times-Ca, 1.6 times-Mg and 4.3 times-K higher at Taurene IM) than in BP. Transboundary air pollution from Western and Central Europe is the main reason why heavy metals contents in precipitation at Rucava IM are substantially higher than in precipitation at Taurene IM. Lead is often trapped by canopies so that TF contributes a lower Pb flux than than BP, but elements such as Zn show enrichment in throughfall (Bergkvist et al., 1989; Ceburnis and Steinnes, 2000). In SF waters, concentrations of Pb and Zn are the highest, and this indicates the wash-off of intercepted dry deposition (Ukonmaanaho, 2001). The trend analysis of heavy metal deposition with BP showed that deposition of Pb significantly declined both at Rucava and Taurene stations: during the period from 1994 - 2004 from 26.0 g/ha to 7.8 g/ha in Rucava and from 24.9 g/ha to 5.4 g/ha in Taurene. However, deposition of Zn showed a significant increasing trend (from 50.4 g/ha per year to 271.8 g/ha per year in Rucava and from 42.1 g/ha per year to 197.4 g/ha per year in Taurene). A similar trend is also seen in heavy metal deposition in TF. Throughout Europe (Renberg et al., 2000) Pb concentrations in BP, TF and SF have declined due to use of nonleaded gasoline (Ukonmaanaho, 2001). Zinc concentrations in BP and TF decreased but in SF significantly increased at Rucava IM. At Taurene IM station concentrations of Zn in precipitation did not show any substantial changes in the period of study. Data summary and analysis of Chemical composition of precipitation at the Rucava and Taurene IM stations has shown that quite big fluctuations of Chemical composition of precipitation occur in Latvian pine forests. At the same time, changes of separate parameters show clearly trends that in time may improve conditions of pine forests growth. Tendencies of changes and their intensity depend both on Chemical composition of precipitation, dry deposition of polluting substances, biological and Chemical processes in forest ecosystems. The forming of forest litterfall Chemical composition Input of substances in forest ecosystems is not only with precipitation, but also with forest litterfall (LF). The doctoral work reviews the role of soil and needle Chemical composition in forming litterfall Chemical composition and analyzes seasonal changes of litterfall Chemical composition. The results of soil analysis show that the concentrations of biogenic elements in upper soil horizons up to 40 cm deep were higher in the Rucava catchment.
However, in deeper soil layers (60 - 80 cm) higher concentrations of Ca, Mg and К and Ntot were found at the Taurene monitoring catchment. This can be explained by soil formation conditions. In a Costal lowland soils are formed on marine sediments, containing a relatively great admixture of small shells. However, at the Taurene IM catchment, soils are formed on glaciofiuvial sediments and differences of grading composition determine the distribution of Chemical element concentrations in soil layers. Concentrations of heavy metals were higher in the О horizon at the Rucava IM catchment, and gradually decreased with deeper layers, indicating heavy metal input to soil from the atmosphere. Concentrations of specific Chemical elements in needles differed in relation to needle age. The Ca concentration in the older needles (C+2) was considerably greater (mean 5.1 (± 0.8) g/kg at Rucava IM and 4.2 (± 2.0) g/kg at Taurene IM) than in the younger needles (C) (mean 3.3 (± 1.0) g/kg at Rucava and 2.2 (± 1.1) g/kg at Taurene). The Mg and Ntot concentrations did not change substantially with needle age. The К concentration was greater in the youngest needles (mean 6.3 (± 1.3) g/kg at Rucava and 5.7 (± 0.7) g/kg at Taurene) due to greater biological vitality and gradually decreased in older needles (mean 5.3 (± 0.6) g/kg at Rucava and 4.9 (± 0.7) g/kg at Taurene). Changes of heavy metals concentrations depending on needle age were not unequivocal. The distribution of Zn concentration depending on needle age showed no clear differences. The concentration of Pb depending on needles age increased in Rucava, but such a trend was not seen in Taurene. Fig. 2. Seasonal variation of litterfall production (± standard deviation) during the year in the Rucava and Taurene IM stations Studies in pine forests ecosystems of Latvia have shown that fluctuations of Chemical composition of needles during a longer period of time are very variable. |
See all the reviews