黑河中游土壤性质变异及其与植被分布的关系研究
文献类型：学位论文
| 作者 | 李丹凤 |
| 学位类别 | 博士后 |
| 答辩日期 | 2016-08 |
| 授予单位 | 中国科学院研究生院 |
| 授予地点 | 北京 |
| 导师 | 傅伯杰 |
| 关键词 | 土壤有机碳 Soil organic carbon 土壤水分特征曲线 Soil water characteristic curve 模型模拟 Model simulation 土壤水分 Soil moisture 植被分布 Vegetation distribution |
| 学位专业 | 生态学 |
| 中文摘要 | 在干旱区，土壤是荒漠植被的载体，土壤质量的高低，直接影响着荒漠生态系统的稳定性。了解土壤性质的时空变异性是理解景观尺度上土壤中各种过程的关键，研究土壤性质的时空变化对植物物种多样性分布格局的影响，对于揭示植物种群的保护和恢复机理具有重要意义。首先，土壤水分是干旱区植被生长的最大限制因子，其空间分布及动态变化限着定居的植物种类和数目，控制着植被的格局、多样性及演替，对生态系统的结构和功能起重要作用。另一方面，植被变化通过改变地表反照率、粗糙度、土壤温度及土壤水分等属性，直接影响土壤-植被-大气之间的水热交换通量，进一步影响植被自身的生长状况。其次，土壤中各种生物地球化学过程均涉及到土壤碳（包含有机碳（SOC）和无机碳（SIC））和土壤全氮（TN），二者的含量对土壤植物系统的相互作用有直接影响。再次，土壤剖面质地构型及结构对其持水性能强弱有重要影响。本报告针对黑河中游地区绿洲和荒漠镶嵌分布且相互影响的景观格局特征，以绿洲——荒漠过渡带的基本土壤属性和植被分布为研究对象，开展土壤含水率的连续定期监测、剖面不同深度土壤碳和全氮浓度以及土壤机械组成测定，结合植被调查，分析不同尺度土壤养分含量的变异性及其影响因素，开展基于土壤性质的土壤持水性 能模拟，研究土壤水分状况的时空间变化及其与植被生长的关系。得出以下主要结论： （1） 黑河中游地区不同景观单元，不同子区域，以及区域尺度上不同土壤层次之间在SOC，TC和TN的浓度以及单位面积的储量方面均存在差异性。土壤属性，植被指标以及农业管理措施等因素可以从不同程度上解释SOCC，TCC和TNC在不同尺度上的变异性。在景观尺度，农田土壤的SOCC，TCC和TNC在不同土层的SR值均较低，表明传统的耕作措施不利于改善农田的养分状况。土壤粉粒含量对林地SOCC，TCC和TNC的主导作用体现了防护林网和阻沙林带对于减弱风力侵蚀，增加细颗粒沉降量和提高土壤肥力的正面效果和重要性。灌木和草本的地上生物量，尤其是湿重，被认为是荒漠土壤中TCC 以及草地土壤中SOCC，TCC和TNC的主导指示因子。在子区域尺度，长期的耕作以及无机肥和农家肥的配合施用导致甘州区农田土壤中SOCD和TND高于临泽县和高台县农田。土壤质地较细和水分亏缺状态导致甘州区内距离黑河河岸较远处林地土壤的SOCD和TND高于距离河岸较近处的林地土壤。 （2） AP模型，MV模型和Rosetta程序对土壤水分特征曲线（SWCC）的预测效果优于AH模型。不同模型和不同质地类型之间，田间持水量，萎蔫系数和可利用水量（AW）的预测值与实测值均呈现不同程度的空间变异性。三属性值在同一质地类型内部不同土样之间亦存在较大的变异性，尤其是砂土。半物理—半经验模型（AP 模型）和概念性模型（MV 模型和AH 模型）由于具有较扎实的物理基础，并且模型计算过程中涉及和考虑了质地类型内部若干不同粒级组分，因此有望更加准确地预测SWCC。更进一步，由于MV 模型完全不依赖于实测SWCC数据，不受SWCC测定过程中的各种误差和不确定性的影响，对SWCC和AW的预测效果优于AP模型和Rosetta程序。 （3） 监测期内研究区年均降雨量仅104.7 mm，几乎全部集中在植被生长旺盛的6-8月份。样带内共出现48 种植被，主要为耐旱的灌木，小灌木和浅根系且耐盐碱的一年生或者多年生草本。荒漠、农田和草地的年均ET值分别为为311.6，579.8和419.2 mm，生育期年均ET占年ET值的比例分别为88.0%，90.3%和90.4%。三种景观单元的月累积ET值均随月累积降雨量的增加而增大，NDVI的季节性变化与植被的周期性生长趋势相吻合。监测期内三种景观单元0-1 m土层的SWS波动较明显，而1-2和2-3 m土层SWS值相对稳定，草地各层SWS 值最高，荒漠最低。剖面三层SWS 在空间上变异最强的为荒漠，草地SWS在空间上各点的差异最小。农田SWS的空间变异性随时间变化变异最弱，其次是荒漠。草地不同监测点的浅层地下水位受抽水灌溉及农田灌溉水深层渗漏补给的影响，导致SWS的空间变异性呈现较强的时间变化。在单独每个观测年内，不同深度的SWS与ET之间以及SWS与NDVI之间均呈现一定的线性正相关，该规律体现在不同景观单元之间以及同一景观内部不同样点之间。 估算区域土壤碳氮储量并分析不同空间尺度上其变异性及影响因素，有助于中游地区生态恢复，在对不同景观单元的土壤碳氮固存水平做出较准确评估的基础上，通过合理规划土地利用，发挥干旱区潜在的土壤碳汇功能。其次，具有扎实物理基础、计算过程严谨且结果可靠的概念性模型有望应用于在流域或区域尺度上对土壤持水性能进行准确估算的研究中，该应用前景将有利于众多水文模型在干旱区的应用。针对现有植被的生长状况及其水分适应策略展开研究，将有利于探讨现有植被格局的生态适应性，发展能够改善生态系统服务的供给能力，更加稳定且可持续的生态系统。 |
| 英文摘要 | Soil is the carrier of desert plants, and its quality directly affects the sustainability of desert ecosystem in arid land. Knowledge about the temporal and spatial variability of soil properties is the key to understand various processes in soil at the landscape scale. Furthermore, study on the influence of the temporal and spatial heterogeneity of soil properties on the distribution of plant species diversity is vital for illustrating the mechanism of vegetation restoration. First, soil moisture is the most important factor controlling the plant growth, and its spatial distribution and temporal variation limit the species, numbers and succession of plants, and play key roles on the structure and functions of ecosystem. On the other hand, the surface albedo, roughness, soil temperature and moisture can be considerably influenced by changes in plant species and distribution, which will affect the transfer of heat and moisture in the soil-plant-atmosphere continuum. Secondly, various biogeochemistry process all involve the carbon and nitrogen levels in soil, which have direct influence on the interaction in soil and plant system. Thirdly, soil water retention are profoundly affected by the soil textural profile and structure. This report focusing on the ecotone of desert and oasis, analyzed the multi-scale variability and influencing factors of soil carbon and nitrogen, estimated water retention of soil from particle size distribution and bulk density, and analyzed the relationship between soil water temporal and spatial variability and plant distribution in the middle reaches of the Heihe River basin. Main results and conclusions obtained were listed below: （1） The concentrations and densities of SOC, TC and TN differed among landscapes, sub-regions, and among soil layers at regional scale. Factors involving soil properties, vegetation condition and agricultural management practices exerted varying degrees in explaining the variations of SOCC, TCC and TNC at different scales. At landscape scale, conventional tillage may be inappropriate for nutrients improvement indicated by the low SRs of SOCC, TCC and TNC in cropland soil. The dominant role of silt particles on the SOCC, TCC and TNC in woodland indicates the importance of shelterbelts or windbreak to alleviate wind erosion, deposit fine particles and improve soil nutrient levels. Aboveground biomass of shrubs and herbs, especially fresh weight, was regarded as a dominant indicator of TCC in the desert and of SOCC, TCC and TNC in the grassland. In regard to sub-regional scale, longer-term cultivation and application of mineral fertilizer combined with farmyard manure led to relatively higher SOCD and TND in cropland of Ganzhou. Fine-textured soil and water deficiency contributed to the higher SOCD and TND in soils of woodland far from the river bank than woodland near the river bank in Ganzhou. （2） The AP model, MV model and Rosetta program showed better agreement with the experimental SWCCs than the AH model. Different degrees of spatial variability in filed capacity, wilting point and available water (AW) were observed among different methods and textural classes. Great variations in the three properties also existed within textural class, especially for sand soil. The semi-physical and conceptual models may be more promise by having robust physical basis and involving specified subclasses of textural data than the Rosetta program, which is based on neural network analysis and acts on broadly-defined soil types. Furthermore, the MV model is not prone to the uncertainty and errors of measurements because it is independent of experimental SWCC data, while the practices of the AP model and Rosetta program are database dependent. The MV model overall outperformed the AP model and Rosetta program in predicting SWCC and AW. （3） Annual mean rainfall was 104.7 mm during the monitoring period, and almost all of which occurred from June to August when the plants flourish. There were total 48 plant species in the rectangular transect, mainly species were draught-tolerant shrubs, undershrubs and salt-tolerant annual and perennial herbs with shallow roots. Annual actual ET in desert, cropland and grassland were 311.6, 579.8 and 419.2 mm, respectively, among which, 88.0%, 90.3% and 90.4% were occurred from April to October. The accumulative monthly ET increased with the increasing accumulative rainfall, positive linear relationship were observed between them. The seasonality of NDVI change coincided with the periodic growth of plants. The SWS in the 0-1 m soil layer fluctuated stronger than the 1-2 and 2-3 m layers for each of the three landscapes. Grassland had the highest SWS, while the lowest SWS was observed in desert. During the monitoring period, SWS in the three layers of desert had the strongest spatial variability, and SWS in the three layers of the grassland showed the lowest spatial variability. The temporal variation of the SWS spatial pattern in cropland was the lowest. Deep percolation of the irrigation water in cropland and the fluctuation of shallow groundwater would influence the soil moisture levels and contributed to the stronger temporal variation of the SWS spatial pattern in grassland. There was no obvious statistic trend between SWS and ET or NDVI during the total three years, but positive linear correlation was observed in either year. This trend existed among different landscapes and different sampling points in each landscape.Investigating the soil carbon and nitrogen stocks, their variability and influencing factors at multiple scales will benefit for the ecological restoration in the middle reaches. Based on the accurate estimation of soil carbon levels, it is potential to play the sink roles of soil to store carbon by reasonably planning the land use in arid regions. Conceptual models with robust physical basis, rigorous calculation processes and reliable prediction results is promising to estimate soil water retention at regional and watershed scales. This prospect will benefit the application of many hydrological models in arid land. Studies on the vegetation distribution, current status and its suitability mechanism of soil water are vital for developing more stable and sustainable ecosystem with improved ecosystem services provision. |
| 源URL | [http://ir.rcees.ac.cn/handle/311016/36874]  |
| 专题 | 生态环境研究中心_城市与区域生态国家重点实验室 |
| 推荐引用方式 GB/T 7714 | 李丹凤. 黑河中游土壤性质变异及其与植被分布的关系研究[D]. 北京. 中国科学院研究生院. 2016. |
入库方式： OAI收割
来源：生态环境研究中心
浏览0
下载0
收藏0
其他版本
除非特别说明，本系统中所有内容都受版权保护，并保留所有权利。