Research On The Atmospheric Deposition Characteristics And The Ecology Response To Acid Deposition In Shaoshan Subtropical Forest, Central-south China

In recent decades, the problem of acid deposition is intensified in China due to the consumption of large quantities of fuel as a result of rapid economic development, and the South China has become the third largest acid deposition zone after Europe and America, also showed the trend of continuous development, and Hunan province located at Central-south China is even more serious, acid rain frequency is long-term high. Forest ecosystems provides a great receptor surface to acid deposition, excess acid deposition may be affected it seriously, leading to soil acidification, forest decline and even so. So, study of acid deposition on forest ecosystems, the ecological response to acid deposition process and propose the controlling method to prevent forest decline is crucial. Therefore, this paper select the typical characteristics of Shaoshan subtropical forest as research object, studying the characteristics of atmospheric deposition, the effect to Shaoshan forest ecosystems by acid deposition and feedback mechanism, can accumulate a useful data of forest ecological responses under acid deposition for China even for the world, and at the same time, can provided a practical application consultation for acid deposition controlling and forest resource management in Shaoshan.Four criteria plots were set in Shaoshan mixed forest, where acid deposition,rainfall below canopy and stream water were monitored from January 2001 to December 2004. We studied the characteristics of atmospheric dry and wet deposition of Shaoshan, analyzed the ion source of bulk precipitation and aerosol by employing PMF method, also analyzed the temporal and spatial dynamic change characteristics of base cations, nutrient nitrogen and dissolved organic carbon in forest ecosystems, then we discussed the acid buffering effect to acid deposition from canopy, sub-canopy and litterfall layer. We evaluated the nitrogen saturation condition by using in-out budget theory and calculated the critical loads of nitrogen by simple mass balance (SMB) method, and used BP neural network to forecast the acid of precipitation and stream water. The research results showed that: (1) The precipitation acidity in Shaoshan is strong, the annual average pH values is below 5.0, acid rain frequency is very high, generally above 90%, the basic ions in precipitation is poised, the ratio of SO42-/NO3- is between 1.72 to 5.65, so precipitation type is still a sulfuric acid precipitation, but the ratio is gradually decreased, indicating that the contribution of nitrogen oxides is increased. The mean dry deposition flux of sulfur is 188 mg m^-2 month^-1 and 105.25 mg m^-2 month^-1 estimating by sedimentation rate method and the net flux method separately. The PMF analytical method of rainfall and aerosol in Shaoshan show that the most important source is soil dust, followed by agricultural sources, and some contribution is just come from traffic sources and long-range transport. (2) The canopy and sub-canopy in Shaoshan forest have shown a good buffering effect, vegetation is sensitive to acid rain, the buffering effect of the canopy is stronger than sub-canopy. The mean net leaching fluxes of base cations is 259.84 meq m^-2 a^-1 and 63.3 meq m^-2 a^-1 from canopy and sub-canopy respectively, which is higher than the H⁺ absorbing fluxes, with 165.9 meq m^-2 a^-1 and 40.9 meq m^-2 a^-1, because of weak acid leaching. The value of pH in litterfall leaching is higher than the average pH of surface soil layer, but because the base cations leaching from litterfall are not enough to counteract the organic acid produced by litterfall decomposition in spring, summer and autumn, so the pH of litterfall leaching is going down in these seasons except in winter, and litterfall in Shaoshan forest is a fine acid-buffering system in winter. (3) There is no evident influence between DOC concentration and bulk precipitation acidity, whether in canopy or in litterfall leaching. Some absorbing phenomenon of DOC has been found in sub-canopy, and accumulation of DOC mainly occurred in litterfall layer in Shaoshan forest, with net DOC fluxes of 11.08 g m^-2 a^-1. A posistive relationship between DOC concentration and pH in stream water is also found, but the output DOC concentration and flux by stream water is very low. (4)The input of inorganic nitrogen in Shaoshan is major in NH₄⁺-N, mean fluxes of inorganic nitrogen input is 3.136 g m^-2 a^-1, but luckily is that it is lower than critical loads of nitrogen. The output of inorganic nitrogen in Shaoshan is major in NO3--N, mean fluxes of inorganic nitrogen output is 0.22 g m^-2 a^-1, just as 7% of inorganic nitrogen input, that to say, more than 90% of inorganic nitrogen is reserved in forest. Though high inorganic nitrogen emission and deposition have been found, Shaoshan forest is nitrogen-limited and far from nitrogen saturation. (5) We simulated the rainfall and stream water acidity using BP neural network model, found that when we selected the concentrations of SO42-, NO3-, Ca2+, NH₄⁺ and combination of factors (Ca2++NH₄⁺)/(SO42-+NO3-) as input, and pH value of precipitation as output, the absolute forecast errors are within±0.4 pH units. We selected precipitation acidity and the acidity of the previous stream water as input, and selected the pH value of stream water as output, the predicted absolute errors are within±0.1 pH units, suggesting that the BP neural network model has high precision and can be used to predict acidity of rainfall and stream water in Shaoshan.