干湿交替过程三维土壤表观电导率与植被地上生物量之间的响应关系

Quantitative relationships among soil water, salt and dominant species biomass accumulation was established by the continuous laboratorial measurement and stratified soil samples in numerous previous studies. However, these simple analyzed methods can’t completely explain the response mechanisms of soil water and salt on dominant species biomass in the three dimensional soil body under the conditions of the complex land cover or dry-wet cycles in arid area. In this project, in order to reveal the effect of soil water and salt on the biomass accumulation of dominant species in arid area, 4 vegetation types of oasis-desert ecotone was selected as research object. Then, the multiple three-dimensional digital mapping techniques were firstly used to simulate the distribution characteristics of three-dimensional soil apparent electrical conductivity(SECa)(agency of soil water and salt) at different stages of dry-wet cycles. Secondly, based on the observed biomass in the different sampling sites, the single phase/interannual aboveground biomass(AB) data was calculated by the vegetation fractional coverage/TIMESET curvilinear integral-biomass correlation models. Lastly, the quantitative analyzed were used to reveal the influence of spatio-temporal variability of three-dimensional SECa on AB in different vegetation types areas. The purpose of this project is to improve our understanding of soil-vegetation physiological mechanism in the oasis-desert ecotone of arid area, and thus providing a potential scientific guidance in conserving the natural resources.

土壤分层采样和室内盆栽连续观测获取的土壤水盐-优势种生物量之间的定量关系，简化了干旱区在复杂植被覆盖/干湿交替过程条件下的三维土体水盐对植被生物量累积的影响机制。为了准确揭示水盐动态背景下植被生物量的积累策略，本项目在灌区外围，选择过渡带上的四种植被类型为对象，首先，利用多种土壤属性三维制图模型，模拟干湿交替过程不同阶段三维土壤表观电导率(Soil apparent electrical conductivity, SECa)(水盐指示因子)的分布特征；其次，基于植被盖度-/TIMESET曲线积分-生物量模型获取单期/年内地上生物量(Aboveground biomass, AB)；最后，定量分析不同覆被下三维SECa时空变异性对AB积累策略的影响过程。本项目旨在深化和完善干旱区土壤-植被知识库，为干旱区绿洲荒漠交错带上重要生物资源的保育提供科学指导依据。

揭示中国内陆干旱区土壤盐度空间分布与演变特征关系到区域水土资源的安全性，同时可为准确评估土壤资源利用潜力，缓解气候变化影响下的旱区水安全研究提供重要的科学依据。本课题针对旱区面积广袤观测资料稀缺性、土壤盐度空间异质性强烈等问题，提出综合采用遥感陆表定量反演方法与土壤景观模型、机器学习算法相结合来研究旱区绿洲及过渡带的土壤盐度空间分异特征与变化规律的理论实践体系，为绿洲土壤遥感调查提供新的思路，也为旱区水资源利用效率及土地退化评估提供基础数据支撑。.围绕本课题研究任务，共计发表论文13篇论文，均发表在Science of total environment、Remote sensing、Geoderma、Catena等高水平SCI期刊以及CSCD期刊上，共发表文章13篇，其中7篇SCI论文，2篇论文为第1标注，9篇第二标注。全面达成了项目设计的研究目标和相关指标。.（1）收集与土壤发生发育相关的遥感变量和其他矢量数据集，并以土壤调查数据为基础，结合土壤景观模型、机器学习算法、EM38电磁感应仪器和数字制图技术，首次获取新疆南部七个主要绿洲多个土壤深度的高精度土壤盐度空间分布图。研究成果与南京土壤所第二次全国土壤调查数据制备的数字土壤属性图进行对比得出，本研究结果无论在精度还是在空间分辨率上都优于当前常用的土壤数据集，Soilgrid和HWSD，可更好的表征新疆南部旱区绿洲土壤盐渍化的垂直分布特性。为旱区水土资源评价提供更准确的数据支撑。该系列成果已发表在土壤环境和遥感等领域SCI期刊。.（2）融合尺度效应于数字制图研究中，探寻绿洲灌区不同下垫面土壤盐度的最佳反演尺度，借助Discrete wavelet transform技术，计算农田及绿洲外围自然植被生长区土壤盐度变异规律，之后融合wavelet reconstruction技术，证明在集合多元环境变量之外，通过关联尺度效应进一步丰富了土壤数字制图在精度提升上的路径。最后通过Hilbert transform技术识别出植被和地形是土壤盐度多空间变异性的主要贡献因素。.本课题旱区绿洲土壤盐度遥感反演，模型模拟研究提供了新认识，具有重要的科学意义。