Knowledge Resource Center for Ecological Environment in Arid Area
| 太湖流域典型土地利用坡面土壤水分时空变化与运动机制研究 | |
| 其他题名 | Study on the spatio-temporal distribution and movement mechanism of soil moisture under two typical land-use hillslopes in Taihu Lake basin |
| 徐飞 | |
| 出版年 | 2015 |
| 学位类型 | 硕士 |
| 导师 | 朱青 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 土壤水分是陆地表面水文过程和能量转换的一个关键变量，而在分布上往往存在高度的时空变异。在西北干旱区土壤水分是限制植被生长的关键因子，因此其时空分布研究得到广泛开展，而在太湖流域，由于面临着严重的生态环境问题，有关土壤水分的研究多集中在水体富营养化方面，鉴于土壤水分运动与营养盐输移密切相关，进行土壤水分时空分布研究有助于土壤中N、P等营养元素迁移的形成机制分析。此外，随着经济社会的发展，太湖流域土地利用方式发生重大转变，在丘陵区主要表现为林地转化农用地，引起农业投入的增加进一步加剧了该区的水环境问题，因此，开展基于典型土地利用类型的土壤水分时空分布及其应用研究显得尤为重要。 本文以太湖流域西部的丘陵山区为研究对象，以“现象—机制—应用”为主线，在基于不同土地利用类型土壤水分长时间序列定位观测的基础上，开展了土壤水分的分布运动研究。首先利用经典统计和地统计方法分析研究区茶园、林地两种典型土地利用类型不同深度（10cm和30cm）土壤水分的时空变异，并通过决策树和典范对应分析识别出土壤水分空间分布主控因子。在此基础上以分析出的主控因子作为辅助变量，进行土壤水分的协同克吕格插值，揭示土壤水分的空间分布特征及其形成机制。最后基于四种常用的土壤转换函数（Rowls、HYPRES、Vereeckn、Rosetta）得出土壤的田间持水量、饱和含水量等基本水力参数，通过土壤水分含量与其相减获取土壤中自由水的空间分布信息，以判断土壤水分运动的活跃区域。论文主要得出以下结论： 1、土壤水分存在一定的时空变异性，并且因土地利用类型和土壤深度的差异而不同。土壤水分的时间序列显示有明显的季节波动，与降雨量的时间格局一致，最高值达到最低值的2倍以上。研究区土壤水分变异系数（CV）值变化范围为19.8-55.2%，在空间上主要表现出中等变异性，茶园10cm和30cm两个深度CV的平均值分别为39.1%和46.6%，大于林地对应深度的28.4%和31.8%。此外，地统计结果表明，土壤水分有较好空间自相关性，由它引起的空间变异占总变异的比值超过了84%，此外，林地的基台值显著高于茶园，表明空间变异更大。 2、影响土壤水分分布的主控因子复杂，并且因土地利用类型、土壤深度、土壤干湿状况的不同有很大差异。干旱季节，两种土地利用类型土壤水分的主控因子较为相近，高程、坡度、地形湿度指数（TWI）等地形因子对土壤水分的影响最大，同时土层厚度也有一定影响。到了湿润季节，茶园10cm深度土壤水分依然主要受高程、坡度等地形因子的影响，30cm深度土壤水分则主要受土壤性质尤其是土壤质地的影响，而林地地形因子和土壤性质对两个深度土壤水分的影响都很大，且30cm深度土壤水分的影响因素比10cm深度更复杂。 3、采用辅助变量的协同克吕格插值精度要好于普通克吕格，两种插值方式精度的优劣受土地利用类型的影响较小，主要受土壤深度以及土壤干湿状况的影响。两种土地利用类型10cm深度普通克吕格和协同克吕格RMSE值之差多小于0.01m3/m3，而在30cm深度两者之差超过了0.03m3/m3，表明30cm深度更适合使用协同克吕格。对于土壤干湿状况的影响，在茶园当土壤水分含量低于0.2m3/m3时，10cm深度两种差值方法的精度几乎相同，30cm深度协同克吕格RMSE值仅比普通克吕格稍低，而林地土壤水分也存在相应的阈值，但数值变为比茶园更大的0.25m3/m3。 4、土壤转换函数是基于大数据样本建立的经验函数，因而可以较好的预测土壤田间持水量，其中茶园在0.27-0.31m3/m3范围内波动，略低于林地的0.28-0.34m3/m3。在此基础上得出的土壤水分运动的活跃程度主要受土地利用类型、土壤深度、土壤干湿状况的作用，表现为除个别日期外，土壤中自由水含量以及存在的面积比在夏秋季节高于春季和冬季，林地高于茶园，30cm深度略高于10cm深度。当土壤水分含量达到一定阈值之后，土壤中自由水存在的面积比与之成显著的正相关（R2大于0.96），茶园该阈值为0.15m3/m3，自由水面积比平均土壤水分含量成指数增长，而林地阈值增大为0.20m3/m3，相关关系变为对数增长。 总而言之，太湖流域丘陵区不同土地利用类型土壤水分的时空变异及控制因素都存在较大差异，且利用土壤水分主控因子结论有助于提高土壤水分空间预测的精度，此外，土壤水分运动在不同土地利用类型以及不同季节也有显著差异。上述结论可以为太湖流域丘陵区农业水肥管理、水环境保护等提供依据。 |
| 英文摘要 | Soil moisture, a spatio-temporal variable, is very importantfor understanding surface hydrologic processes and energy balance. In arid area of northwest China,soil moisture is a key factor limiting the growth of vegetation. In this case there are many studies on the spatio-temporal variability of soil moisture. However in the Taihu Lake Basin little research was conducted to study the spatio-temporal distribution of soil moisture. This kind of research can help understand the transport of nitrogen and phosphorus in the vadose zone. In addition, with the development of ecology and society, landuse has changed greatly in Taihu Lake basin (e.g., the conversion of forests to agricultural land in hilly area). This may increase the agricultural investment, whichwill further exacerbate the problem of water pollutions in the area.Therefore, it is very criticalto study the spatio-temporal distribution of soil moisture and its application, with consideration of different land-use types. In this paper, the study area was located in the hilly regions of west Taihu Lake Basin.Based on long-term observations of soil moisture under different land use types, the distribution and movementof soil moisture was explored. Firstly, the spatio-temporal variability of soil moisture under two typical land use types at 10- and 30-cm depths was analyzed by using classical statistics andgeostatistics. Secondly, the classification and regression tree (CART) and canonical correspondence analysis (CCA)approaches were applied to quantitative detect the relationships between soil moisture and environmental factors. Then, soil moisture was interpolatedby using Co-kriging method. Finally, soil field capacity, saturated water content and other hydraulic parameters were obtainedby four traditional pedotransfer functions (i.e., Rowls, HYPRES, Vereecken and Rosetta).Theirspatial distributions were computed byRaster Calculator in ArcGIS to determine hot spots of soil moisture movement. The main conclusions are listed as follows: 1.Soil moisture has spatio-temporal variability, which was influenced byland use and soil depth. Soil moisture was found to have a significant seasonal change, which was corresponded with the temporal pattern of rainfall,The highest value of soil moisturewas 2 times largerthan the lowest value. The CV values of soil moisture range from 19.8 to 55.2%, showing a moderate variability.The mean CV values of soil moisture in the tea garden (TG) at 10- and 30-cm depths were 39.1% and 46.6% respectively. In forest land, the mean CV values were 28.4 % and 31.8% for 10- and 30-cm depths respectively. The value of SD at 10-cm depthwas less than thatat 30-cm depth. In addition, geostatistical analysisindicatesstrong spatial autocorrelationofsoil moisture, accounting for more than 84% of the total variation, and the sill value in forest land (FL) is higher than in TG, which shows a higher spatial variability of soil moisture in FL. 2.Influencing factors of soil moisture varied with land use, soil depth and soil moisture conditions. Under dry conditions, soil moisture inboth TG and FL was mainly affected by topographic indices such as elevation, slopeand topographic wetness index (TWI).Under wet conditions, soil moisture at 10-cm depthwas still influenced by topographic indices in TG.Howeversoil moisturewas mainly affected by soil properties (e.g., soil texture)at 30-cm depth, It is noted that in forest land both soil properties and topographic factors influenced soil moisture at 10- and 30-cm depth. 3.Co-kriging was found to have better performance than kriging in predicting spatial distribution of soil moisture. These two interpolation methodswere affected less by land use,but influenced largely by soil depth and soil moisture conditions. In both TGand FL,the differencebetween the RMSE value of Ordinary Kriging and that of Co-Kriging was larger than 1 for 10-cm depth.However, the difference between the RMSE value of kriging |
| 中文关键词 | 土壤水分 ; 时空变异 ; 主控因子 ; 协同克吕格 ; 土壤转换函数 ; 太湖流域 |
| 英文关键词 | Soil moisture Spatio-temporal variability Main controlling factors Co-Kriging Pedotransfer functions TaiHu Lake basin |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 自然地理学 |
| 来源机构 | 中国科学院南京地理与湖泊研究所 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/287560 |
| 推荐引用方式 GB/T 7714 | 徐飞. 太湖流域典型土地利用坡面土壤水分时空变化与运动机制研究[D]. 中国科学院大学,2015. |
| 条目包含的文件 | 条目无相关文件。 | |||||
| 个性服务 |
| 推荐该条目 |
| 保存到收藏夹 |
| 导出为Endnote文件 |
| 谷歌学术 |
| 谷歌学术中相似的文章 |
| [徐飞]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [徐飞]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [徐飞]的文章 |
| 相关权益政策 |
| 暂无数据 |
| 收藏/分享 |
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。
