【作者】 荣丽；

【导师】 李贤伟；

【作者基本信息】 四川农业大学 ， 森林培育， 2009， 博士

【摘要】 细根分解是森林生态系统养分和能量循环的重要环节,是贯穿淋溶粉碎等物理作用以及土壤生物参与的生物化学作用在一起的复杂过程。尽管细根埋藏于地下,给细根分解的研究带来挑战,但仍然引起了人们的广泛关注。本文选择了在四川盆地西缘洪雅县内具有显著生态、经济和社会效益的光皮桦(Betula luminifera)-扁穗牛鞭草(Hemarthria compressa)复合模式(HN)、光皮桦人工林(H)、扁穗牛鞭草牧草地(NC)、柳杉(Cryptameria fortunei)人工林林(LS)为研究对象,采用原状土芯法系统研究了这四种退耕模式的细根分解速率、影响分解的主要化学成分、养分释放及其对土壤养分的影响、与细根分解相关的主要土壤微生物功能群、土壤酶的动态,并通过室内升温模拟与野外试验相结合,研究了细根分解对土壤活性有机碳的影响以及土壤活性有机碳对升温的响应。结果表明:(1)在1a的研究期间内,几种退耕模式细根(草根)分解速率符合Olson单指数分解模型。分解1a后HN、H、NC、LS模式细根(草根)干重损失率分别为65.49%、60.55%、73.32%、57.51%;NC比其他三种模式细根分解都快,其次是HN、H,LS分解最慢。年分解常数(k)分别为1.06、0.93、1.32、0.86/a。四种退耕模式细根分解分为快-慢两个阶段,前90d分解较快,以后表现出缓慢趋势。在整个研究期间,各模式细根分解速率与木质素绝对含量相关性不显著(P＞0.05),但分解90d以后,H、LS模式细根分解速率与木质素绝对含量呈显著负相关(P＜0.05),HN模式细根分解速率与木质素含量呈较弱的负相关关系(P=0.087)。在分解较慢的阶段,木质素绝对含量影响退耕模式林木细根分解最大的指标,但扁穗牛鞭草草地除外。整个研究期间,各模式细根分解过程中的C/N与细根分解速率呈极显著负相关(P＜0.01),是影响退耕模式细根分解的最大质量指标。(2)四种模式下细根(草根)的C、K、Mg表现为净释放,C损失符合单指数模型,其决定系数R~2＞0.95(P＜0.01)。HN、NC模式的N、Ca表现为净释放,H、LS模式有富集现象,P在四种模式中都有富集。养分释放量大小为C＞K＞P＞N＞Mg＞Ca,但是H与LS模式的C释放量却小于K的释放量,HN模式细根的P释放与N释放量相差不大,H模式细根的P释放量要小于N释放量。分解1a后,C释放量大小顺序为:NC(80.31%)＞HN(78.71%)＞H(69.5%)＞LS(60.53%)。HN模式的N、P释放量都大于其他模式。N初始含量最高的HN具有最高的N释放率,N初始含量最低的LS具最低的N释放速率,其释放量大小顺序为HN(70.12%)＞H(56.49%)＞NC(46.67%)＞LS(21.86%);P释放量大小为HN(69.56%)＞NC(67.42)＞H(52.89%)＞LS(45.67%);K释放量大小为NC(93.52%)＞HN(89.20%)＞H(85.97%)＞LS(82.64%)。(3)分解1后,HN、H、NC、LS模式土壤有机质含量是分解试验开始前土壤有机质含量的1.44、1.15、1.25、1.14倍;HN模式下的土壤有机质含量分别是H、NC、LS模式下的1.34、1.29、1.48倍。土壤速效K与细根(草根)K释放量呈显著正相关关系;HN模式的细根N、P释放量与土壤全N、全P含量显著相关,LS模式下的细根P释放量也与土壤全P显著相关(P＜0.05)。各退耕模式细根K的释放量对土壤速效K的贡献较大,其贡献率达到70%以上,另外HN模式下细根N、P释放量对土壤全N、全P的贡献率也分别为68.45%和77.90%,LS模式下细根P释放量对土壤全P的贡献率为73.18%。与其他模式相比,光皮桦-扁穗牛鞭草复合模式细根分解过程中N、P释放量最高,且对土壤全N和全P的贡献较大,这对于维持林地土壤肥力是有益的。(4)在研究期间,HN模式的好气固氮细菌、氨化细菌、硝化细菌、纤维素分解菌数量都显著大于其他三种模式(P＜0.01),表明光皮桦-扁穗牛鞭草复合模式比其他模式更能促进了细根、草根含N、C化合物的分解和碳素循环,有利于提高土壤氮素水平以及土壤肥力促进林木生长;四种模式固氮细菌与氨化细菌数量呈显著正相关关系,与C/N呈显著负相关关系,氨化细菌数量与硝化细菌数量呈显著正相关,而这种养分途径却对于林草复合生态系统以及人工针叶林生态系统的养分动态具有重大意义,特别是林草复合模式实现自肥能力一个不可缺少的生态过程。HN、LS模式固氮菌数量与细根、草根全N含量呈显著正相关(P＜0.05)表明生物自生固氮是细根分解期间的重要过程。(5)HN模式下的土壤脲酶、蔗糖酶、酸性磷酸酶都在整个研究期间其活性较大,其中脲酶、蔗糖酶活性的增量比其他模式多,其次是NC模式,NC模式的酸性磷酸酶活性增加得最多,为初始值的1.52倍;整个研究期间LS模式下土壤脲酶、酸性磷酸酶活性最小,显著低于其他模式(P＜0.05)。以上结果表明了HN模式下与土壤肥力有关的酶活性比其他模式大,从土壤酶的角度也说明了光皮桦-扁穗牛鞭草复合模式比其他人工林供肥、保肥能力强,有利于地力的维持。HN模式下的土壤脲酶、蔗糖酶活性与细根分解速率呈显著正相关,NC模式下的土壤脲酶、土壤酸性磷酸酶活性与草根分解速率呈显著正相关,LS模式下的土壤脲酶、土壤多酚氧化酶活性与细根分解速率呈正相关(P＜0.05);土壤脲酶活性与细根C/N、纤维素绝对含量呈显著负相关关系(P＜0.05),但NC模式下的土壤脲酶活性与草根C/N相关性不显著,H模式下的土壤脲酶与纤维素绝对含量相关性也不显著;四种模式土壤脲酶活性与固氮细菌或与真菌数量呈显著相关关系,HN模式下的土壤蔗糖酶活性与细菌和纤维素分解菌数量正相关,另外H与NC模式下的土壤酸性磷酸酶还分别与细菌和纤维素分解菌数量正相关(P＜0.05)。由分析得出,不同模式下受细根(草根)分解速率及其化学成分影响的土壤酶种类不同,但光皮桦-扁穗牛鞭草复合模式、柳杉人工林下的土壤脲酶都受到细根分解速率、细根C/N以及细根纤维素绝对含量的影响,并且还受到固氮细菌或与真菌数量的影响。(6)在野外分解试验中,HN模式下的土壤微生物量碳(MBC)、水溶性有机碳(WSOC)、易氧化碳(ROC)、总有机碳(TOC)都大于其他三种模式,室内分解试验表明添加光皮桦细根和扁穗牛鞭草草根混合根的处理大于其他添加单一细根的处理,两者表现出一致的结果。野外试验表明,HN、H、NC、LS模式下的土壤微生物量碳对土壤总有机碳的贡献分别是1.2-3.3%、0.7-1.5%、0.8-2.2%、0.5-0.8%;HN模式的ROC/TOC大于H、NC、LS模式,表明与其他人工林相比,HN土壤有机碳活性大、易转化;各模式土壤易氧化碳含量与土壤总有机碳呈极显著相关关系(P＜0.05),说明土壤总有机碳的高低决定了易氧化碳的丰缺。室内模拟分解试验表明,添加细根的种类、培养温度和培养时间均对土壤MBC、WSOC、ROC、TOC的含量产生影响。120d后添加细根的处理其土壤MBC、WSOC、ROC、TOC含量都显著大于对照(P＜0.05);温度对MBC、WSOC含量的影响强度依次为:30℃＞20℃＞10℃,对ROC和TOC含量的影响强度为:20℃＞30℃＞10℃(P＜0.05)。添加外源细根会增加土壤MBC、WSOC、ROC、TOC含量:30℃的温度对MBC与WSOC含量的影响强度最大,而20℃对ROC和TOC含量的影响强度最大,但在考虑到野外试验结果时,应当根据细根化学成分变化的过程确定室内模拟培养时间,本研究建议室内模拟培养的时间最好延长至180d。综合分析表明,与其他人工林相比,光皮桦-扁穗牛鞭草复合模式中由于扁穗牛鞭草的植入改善了林下土壤的微环境,促进了林木细根分解,增加土壤有机质以及养分的积累、增加土壤微生物数量和土壤酶活性以及提高了土壤活性有机碳的含量,有利于该模式土壤养分循环以及土壤肥力的维持。更多还原

【Abstract】 Fine root decomposition is one of the major pathways of nutrient and energy cycling in forest ecosystem,and also,a complex process including physical（leaching and fragmentation） and biochemical processes.Estimating rates of root decay is challenging because roots are hidden from view.Despite this challenge,there is a large and growing database on roots decomposition from both tropical and temperate forests.However,much less attention was paid to subtropical forests,particularly the plantations founded in the process of converting farmland to forest（grass）.This study was carried out in the key scientific and experimental demonstration station of the sloping land conversion program for converting cropland to forest（grassland） of Sichuan Agricultural University,located at Liujiang town,Hongya county in southwest edge of the Sichuan basin in China.Fine root decomposition rate and its influencing factor,nutrient release and the impact on soil nutrient,major microbial functional groups（aerobic azotobacter,ammonifiers,nitrifiers, cellulolytic bacteria） and soil enzymes in relation to fine roots and grass roots decomposition were analysed in birch（Betula luminifera）-grass（Hemarthria compressa） compound model（HN）,which is a successful forest and grass compound pattern in the process of converting farmland to forest（grass） is beneficial to the development of environment and economy,birch（Betula luminifera） plantation（H）,grassland（Hemarthria compressa）（NC）,cedar（Cryptomeriafortunei） plantation（LS） using intact core method for one year.Moreover,soil active organic carbon dynamic was also investigated using intact core method and a 120-day laboratory incubation in soil microcosm for the four plantations. After studies,the following results have been obtained.（1） The decomposition rate pattern showed a good fit for a single-exponential model and the decomposition constant（k） for the four plantations were 1.06、0.93、1.32、0.86 year^-1 respectively,with an annual loss of 65.49%、60.55%、73.32%、57.51%during one year study.Decomposition pattern of fine root（grass root） was characterized by two distinct phases of weight loss in all four plantations.The initial rapid rate decay up to 90 d, and then was characterized by slower mass loss.After 90 d lignin content of fine root in H and LS was negatively correlated with fine root decomposition rate（P＜0.05）.As for HN, its lignin content of mixed roots showed a poorly correlation with fine roots decomposition rate（P=0.087）.Lignin content controlled the decay rate at the slower phase in birch-grass compound model,birch plantation and cedar plantation except for grassland.C/N of roots was negative significantly correlated with fine roots decomposition rate for all plantations （P＜0.01）,and was the main factor of controlling decay rate during the one year experiment.（2） Net C,K Mg release from decomposing roots occurred in the course of experiment in all four plantations.The dynamics of C loss from decomposing roots in four plantations fitted to a single exponential model with R²＞0.95（P＜0.01）.Net N and Ca released in HN and NC,but immobilization was observed in H and LS.P was immobilized in the all plantations and showed accumulation during the course of experiment.A large amount of C and nutrients released from decomposing fine root（grass root） in four plantations,and showed this order:C＞K＞P=N＞Mg＞Ca（HN）,K＞C＞P＞N＞Mg＞Ca（H and LS）,C＞K＞P＞N＞Mg＞Ca（NC）.After one year,the order of C release was NC（80.31%）＞HN（78.71%）＞H（69.5%）＞LS（60.53%）.N and P release in HN were higher than other plantations and the ranked order were HN（70.12%）＞H（56.49%）＞NC（46.67%）＞LS（21.86%）,HN（69.56%）＞NC（67.42）＞H（52.89%）＞LS（45.67%）, respectively.All the decomposing roots released most their initial K content during the period of decomposition and the order was NC（93.52%）＞HN（89.20%）＞H（85.97%）＞LS（82.64%）.（3） One year later soil organic matter content in FIN,H,NC and LS were 1.44,1.15, 1.25,1.14 times as much as in the initial content for the four plantation.Moreover soil organic matter content in HN was 1.34,1.29,1.48 times than that of H,NC and LS.Soil available K content was positive correlated with K release from decomposing roots（P＜0.05）.K release from decomposing root had the largest contributions to soil available K and accounted for more than 70%of the variance during the period of decomposition.N and P release from decomposing roots were positive correlated with soil total N and total P content in HN,and accounted for 68.45%and 77.90%of the variance,respectively. Positive correlation was found in LS for P release and soil total P content（P＜0.05）.P release from decomposing root in LS accounted for 73.18%of the variance for soil total P content.It was concluded that the greater amount of N and P release from decomposing root and the larger contribution to soil total N and P in HN plantation indicated that the faster turnover of nutrients,which in turn would be beneficial for maintenance of soil fertility.（4） Aerobic azotobacter,ammonifiers,nitrifiers and cellulolytic bacteria numbers in HN model showed the largest in the four plantations（P＜0.01）.Aerobic azotobacter in four models showed significant positive correlation with ammonifiers and negative correlation with C/N ratio,and ammonifiers also had positive correlation with nitrifiers.Aerobic azotobacter in HN and LS models showed positive correlation with relative nitrogen content of fine roots and grass roots.Our results showed that fine roots and grass roots decomposition were strongly influenced by synergic interaction among microbial functional groups.Moreover also suggested that biological nitrogen fixation may have an important role in self-sustaining capability for forest-grass composite ecosystem.（5） Soil urease,sucrase and acid phosphatase activities in HN were the highest in four plantations.Soil urease and acid phosphatase activities in LS were the lowest in the four plantations（P＜0.05）.In view of soil enzyme,birch-grass plantation ecosystem with high soil enzyme activities should be more nutrient supply,thus,the improvement of soil fertility would be expected for the composite ecosystem.Soil urease activity in HN,NC and LS was positive correlation to root decomposition rates,and also sucrase activity in HN,soil aicd phosphatase activity in NC,soil polyphenoloxidase in LS was positive correlation to root decomposition rates（P＜0.05）. Soil urease activity was negative correlation to C/N and absolute content of cellulose in decomposing roots.However,the correlation between soil urease and C/N in NC and the correlation between absolute content of cellulose and soil urease in LS were not significant. The correlation between soil urease activity and the number of aerobic azotobacter and soil fungi was significant.Moreover,soil surcease activity was significant positive correlation with cellulolytic bacteria numbers in HN,and also soil icd phosphatase activity in H and NC was significant positive correlation with bacteria and ceilulolytic bacteria numbers（P＜0.05）.It concluded that soil enzyme activity was strongly affected by root mass loss, chemistry component in root and soil microorganism. （6） In the field experiment,soil microbial biomass carbon（MBC）,water soluble organic carbon（WSOC）,readily oxidizable carbon（ROC）,total organic matter（TOC） content in HN were higher than other plantations.It was consistent with the result from laboratory microcosm experiment,in which these indices in treatment with the mixture of birch fine root and grass（Hemarthria compressa） root were higher than the other treatments with single birch fine root,single grass root,and single cedar fine root.MBC in HN contributed more C to soil organic C（1.2-3.3%） than H,NC and LS plantation（0.7-1.5%,0.8-2.2%, 0.5-0.8%,respectively）.ROC/TOC in HN was also the highest in the four plantations.In addition,ROC content was positive correlated with TOC content in four plantations（P＜0.05）.It indicated soil organic matter in HN was much easier transformed compared to other three plantations,and the content of ROC depended on the content of TOC.In the laboratory microcosm experiment,fine root source,incubation time and incubation temperature all influenced MBC,WSOC,ROC,TOC.After 120 d,soil MBC、WSOC、ROC、TOC content in the treatment with fine roots were higher than in the control treatment with no root added（P＜0.05）.The effect of incubation temperature on MBC and WSOC in four treatments showed this order:30℃＞20℃＞10℃,while the order of ROC and TOC was 20℃＞30℃＞10℃（P＜0.05）.Considering the results from the field experiment and variation of chemistry component,the incubation time in the microcosm experiment should be delayed.In our study the appropriate incubation time was 180 d.The results mentioned above implied that soil micro-environment under the birch （Betula luminifera）-grass（Hemarthria compressa） model was changed as for introduction of grass,in which fine root decomposition rate could be promoted.Furthermore soil organic matter and soil nutrients,soil microorganism and soil enzyme activity,soil active organic matter should be accumulated and increased,thus,improvement of soil fertility would be expected for forest-grass compound model.更多还原