An Study On Response Of Soil Respiration To Degraded Ecosystem In Mixed Wood Zone

Soil respiration is key for the global carbon cycle.Forest soil respiration is an important part of the soil respiration in terrestrial ecosystems,and its variation will bring far-reaching influence on the balance of global carbon.Precise measurement of soil respiration and the contribution of each part is the most basal and exigent problem in global change research.In this thesis,we researched the degraded ecosystems in mixed wood zone of Changbai Mountain that include original broadleaved korean pine forest(K),poplar-birch forest(Y), fraxinus and juglans mandshurica forest(S),mongolican oak forest(M),bare land(L) by fixing position measurment in field and analyzing in laboratory.Trenching-plot and infrared gas exchange analyzer(Li-6400P,LI-COR InC.) approaches were used to measure the total soil respiration(R_T) and soil respiration components accurately and consecutively.Major conclusions are summarized as follows:(1)The diel variation patterns of R_T in different degraded stages are nearly same,and are all single modal curve,but the time of maximums and minimums R_T and the variation ranges of R_T is different.Compared with control plot(K),maximums and minimums R_T of secondary forests in different degraded stages appeare earlier.The coefficients variation of the more serious degraded stages are larger than that of other plots significantly,which results in that the more serious degraded plots are affected easily by external environment.The seasonal variation patterns of R_T in different degraded stages show clearly single modal curve, and the maximums R_T of different plots appear in July or August.The diel variation patterns of soil respiration components in different degraded stages are similar to the diel variation patterns of R_T.The seasonal variation patterns of root respiration(R_a) in different plots show accordant single modal curve basically,and the maximums appear in July or August,but maximums of different plots donot arise simultaneity.The variation patterns of heterotrophic respiration(R_h) in growing season are similar to the seasonal variation patterns of R_T,but the variation range of the former is smaller than the latter.(2)As a complex ecological process,soil respiration receives the comprehensive influence of non-biological factors(air and soil temperature,soil water content,soil physical and chemical property) and biological factors(vegetation type,root system).Soil respiration could be better simulated by an exponential function(R=a*e^b*T) with soil temperature as the driving variable that can explain 44.3%⁹1.5%of soil respiration.The relationship of soil water content with soil respiration is more complex,and can be expressed by a linear model (R=a* W+b) that can account for 21.4%⁶1.1%of soil respiration.There are great changes on soil physical and chemical property since forest ecosystems were influenced,which is an important reason that soil respiration rates of different degraded stages and their variation patterns have changed.The seasonal variation patterns of R_a in different plots are in the same tendency with the contribution rate variation of R_a whose variation will bring far-reaching influence on the variation of soil respiration.Vegetation community composition,type and magnitude in different degraded stages change in the different way,and interact with surrounding environment and soil,then form vegetation community-soil-habitat and microclimate characteristics of different succession stages,accordingly affect soil respiration.(3)Since forest ecosystems degenerate,there are some changes in temperature sensitivity of soil respiration.The temperature sensitivity coefficients(Q₁₀) of R_T and R_A for every plots are in sequence of K＞Y＞M＞S＞L.The Q₁₀ values of litter respiration(R_L) are in sequence of K＞Y＞M＞L＞S.The Q₁₀ values of S,M and L are 2.09,3.15,1.75 respectively.R_A is more sensitive to temperature than R_h.(4)Nonlinear two-factor model could simulate soil respiration and can explain 44.3%⁹1.5%of R_T,which is better than linear two-factor model and single factor (temperature) model.According to nonlinear two-factor model and monthly mean of correlative environment factors,we can estimate the amount of CO₂ release for every plots during the growing season.The amounts of CO₂ release in Y and M are about 1.4 and 1.3 times than in K,and the amounts of CO₂ release in S and L account for 88%and 78.%of that in K.The contribution rate range of R_A in different plots is 21.26%⁵3.46%in which the rate in K is largest(50.41%).The mean contribution rates of R_L in Y and M are 35.03%and 32.14%respectively,which are both larger than that in K(26.78%).