IMHE OpenIR  > 山地表生过程与生态调控重点实验室
四川盆地农田土壤固碳潜力模型模拟
Alternative TitleSimulation of SichuanBasin Farmland Soil Organic Carbon Sequestration Potential
Language中文
罗勇
Thesis Advisor朱波
2014
Degree Grantor中国科学院研究生院
Place of Conferral北京
Degree Name博士
Degree Discipline土壤学
Keyword四川盆地 农田 土壤有机碳 固碳潜力 Dndc 尺度拓展 模型模拟
Abstract土壤有机碳库作为陆地生态系统碳库重要组成部分,在碳循环动态中有举足轻重的作用。农田土壤有机碳库作为陆地生态系统中最活跃和最有影响力的碳库之一,对其进行研究是目前全球变化领域热点,有助于为调整现有农田管理措施提供参考依据,从而制定有利于土壤碳积累的农田管理措施。 在过去几十年,土壤有机碳储量和分布格局已经得到广泛研究,对我国国家尺度的估算范围在50-185 Pg C 不等。其中中国表层土壤有机碳库38-40Pg。由于研究尺度过大和方法的不同导致差异巨大,且多是利用全国土壤二次调查数据,研究缺乏时效性。为了更好地估算土壤有机碳变化,迫切需要对不同尺度特别是中小尺度下土壤有机碳的精细研究。 四川盆地占西南地区面积比例8.58%,占全国面积比例1%,四川盆地的有机碳库对西南乃至全国都具有重要的作用。但目前四川盆地农田土壤有机碳库的相关研究结果较少。近年来土壤有机碳动态模型在全球农田生态系统碳循环研究中得到了较为广泛的应用。DNDC模型、RothC模型、CENTURY模型等有机碳动态模型得到了学者的验证。但是在四川盆地区域尺度上土壤有机碳动态模型的应用较少;且点位尺度模型拓展到区域乃至全国或全球尺度时,随着研究尺度增加,尺度增加造成的土壤异质性越大的问题。 因此,本文依据最新的农田土壤调查数据,对四川盆地典型农田土壤有机碳库现状进行了研究,并将实测采样点结合小尺度典型县DNDC模型模拟结果,利用地理加权回归方法,建立环境关系模型并予以验证,模拟得到了四川盆地现实农田土壤有机碳,同时构建的四川盆地农田土壤固碳潜力DNDC模拟所需的高度空间离散化的模拟单元,大大提高了DNDC模型模拟四川盆地农田土壤固碳潜力的可靠性,这在一定程度上解决了以往学者在区域尺度DNDC模拟研究中,最小单元难以细化导致的随着研究尺度增加,土壤异质性越大的难题。同时,估算了四川盆地农田土壤有机碳现状及潜力,并揭示其时空累积特征,对于正确认识并准确评估四川盆地农田碳库在全国或者全球碳库中的地位具有参考意义。 因此,本文以四川盆地农田土壤为对象,收集了四川省全国土壤二次调查数据,并于2012-2013年对川中丘陵和成都平原四个典型县进行了土壤采集,整合DNDC生物地球化学模型和典型县1:5万土壤数据库、四川省1:100万土壤数据库,以及气象、作物、耕作管理等 GIS 数据库开展研究。首先,利用1980年和2012年典型县土壤调查数据,分析了近30年来川中丘陵和成都平原农田土壤时空变化特征;其次,在点位和区域尺度上以实测数据对DNDC模型进行校正、验证逐级拓展。第三,以2012实测土壤有机碳和典型县DNDC模拟土壤有机碳为对象,探讨了环境地形因子与有机碳相关关系,建立了四川盆地农田土壤地理加权回归模型,估算了四川盆地农田土壤有机碳及其空间分布。第四,利用地理加权回归模拟结果,研究了近30年四川盆地有机碳的演变规律和空间分布。最后,通过设置4 种不同推荐管理措施情景(例如少(免)耕、秸秆还田和有机肥施用等),以2012年高度离散化的四川盆地农田土壤数据为基础,结合四川省农田土壤秸秆施肥调查数据,建立了四川盆地农田土壤固碳DNDC模拟数据库,分析了 2012~2042 年四川盆地的固碳潜力及空间格局。主要研究结论如下: 1. 川中丘陵区的盐亭县和长宁县农田耕层SOC含量都较低,全国土壤二次调查平均SOC密度分别为1.71和2.32 kg m-2,低于全国农田耕层平均有机碳密度;最新调查数据显示,盐亭县和长宁县农田土壤耕层平均SOC密度分别为2.66和3.08 kg m-2,都呈现上升趋势, 30年来单位面积固碳速率分别为327和236 kg ha-1 yr-1,是一个较强的碳汇,接近于全国平均单位面积的平均固碳速率316 kg ha-1 yr-1。成都平原的广汉市和大邑县农田SOC含量都较高,全国土壤二次调查平均SOC密度分别为3.84和4.00 kg m-2,最新调查农田土壤耕层平均SOC密度分别为4.56 kg m-2和5.01 kg m-2,均大于全国土壤二次普查耕层20cm水稻土平均SOC密度。广汉市和大邑县耕层土壤30年来耕层土壤单位面积固碳速率分别为215和316 kg ha-1 yr-1,是一个较强的碳汇。 2. 以中国科学院盐亭农业生态实验站实测数据驱动模型,对DNDC点位模型的作物生物量和土壤有机碳模拟验证结果表明,随着模拟时间的增加,干扰因素降低,小麦模拟产量与实测产量越接近,2012年MEE为-7.94%, 2012年玉米各处理模拟产量与实测产量模拟结果MEE为-8.87%。模型对2008土壤耕层有机碳模拟的误差范围在1.17-16.37%间,2012年误差范围为-6.85-17.92%。在各典型县区域尺度上的验证结果来看,成都平原和川中丘陵地区2012和2013年各土壤类型有机碳模拟平均值与实测平均值之间均达到极显著相关水平(P<0.01),在所有模拟值与实测值的统计比较中,相关系数都大于0.7。4个典型县有机碳密度模拟与实测MEE分别为4.41%、4.50%、4.41%、-2.92%。表明经校正后的 DNDC 模型在四川盆地农田土壤研究中适应性良好,可以较好表达点位和区域农田土壤有机碳动态。 3. 利用典型县部分实测数据对基于县域DNDC部分模拟结果和实测结果建立的回归克里格模型验证结果表明,有机质地理加权回归的ME高于回归模型预测结果的ME,其能够利用四川盆地环境地形因子,较准确地模拟2012年有机碳空间分布。 4.利用地理加权回归模型结果,得到2012年四川盆地农田耕层20cm土壤有机碳储量0.22Pg,平均有机碳密度3.67 kg m-2,其中水田耕层20cm有机碳密度4.15kg m-2,有机碳储量0.10Pg,旱地耕层20cm有机碳密度3.43 kg m-2,有机碳储量0.12Pg,近30年来四川盆地农田土壤有机碳总体呈上升趋势,平均固碳速率325 kg ha-1a-1,是一个较强的碳汇。 5. 四个不同农田管理处理DNDC模拟结果显示四川盆地未来30年农田土壤有机碳密度均有一定程度升高,其中A、B、C、D四个综合管理措施固碳潜力分别0.079 Pg、0.067、0.097Pg、0.022Pg。各处理潜在固碳速率C处理>A处理>B处理>D处理,分别为533 kg ha-1 yr-1、433 kg ha-1 yr-1、367 kg ha-1 yr-1、123 kg ha-1 yr-1,与韩冰等认为四川省未来固碳潜力趋势一致,但低于其估算的四川省农田平均固碳潜力718.15kg ha-1 yr-1,C处理是较优的农田增汇管理措施。
Other AbstractAs a main component of terrestrial ecosystem carbon stocks, soil carbon stocks play an important role on carbon cycle dynamics. The study of Soil organic carbon pool in farmland has become the hot of global change, which would provide the reference for adapting the existing farm management measures, and develop the farmland management measures in favor of soil carbon accumulation. The research targeted for soil organic carbon storage and its spatial pattern characteristics under typical county scale could provide a reference for regional carbon sequestration. In the past few decades, soil organic carbon storage and distribution patterns have been studied extensively, and the estimations from different studies indicated that the total SOC stocks of China’s soils ranged from 50 to 185 Pg. The topsoil organic carbon stocks of China was 38-40Pg. Since large-scale and different methods lead to a huge difference result, and most of them use the second national soil survey data which make the study lack of timeliness. In order to better estimate the soil organic carbon change, it’s necessary to research on soil organic carbon in different scales especially small scale. Sichuan basin accounted for 8.58% of southwest China, and 1% of China’s total area, so the organic carbon stocks of Sichuan Basin make an important role for the southwest and the country. But only a few relevant researches on soil organic carbon stocks of Sichuan basin farmland. In recent years, the dynamic model of soil organic carbon has been more widely used in the global carbon cycle farmland ecosystem research. The organic carbon dynamics models, such as DNDC model, RothC model and CENTURY model have been verified by scholars. But dynamic models of soil organic carbon are rarely applied in the Sichuan basin regional scale. When the point scale model extended to regional and national or global scale, scales increase soil heterogeneity that cause greater problems with the increasing of research scales. Therefore, based on the latest farmland soil survey data, the paper researched on soil organic carbon stocks in the typical Sichuan basin farmland. Combined with the measured sample points and the DNDC model in the typical county small-scale, using geographically weighted regression method to establish the environment relationship model, simulated the sequestration potential of soil organic carbon on farmland in the Sichuan basin. Build the High spatial discrete units which is required by DNDC and greatly improve the reliability of DNDC model simulating the sequestration potential of soil organic carbon on farmland in the Sichuan basin. To some extent, it solves the problems in DNDC’s simulation study that the smallest unit is difficult to refine which lead to the soil heterogeneity with the increasing of research scale in the area scale. Meanwhile, estimating the status and the sequestration potential of soil organic carbon on farmland in the Sichuan basin, and revealing its spatial and temporal accumulation characteristics have very important reference value for the correct understanding and accurately assessing the status of Sichuan basin’s soil organic carbon on farmland in the country or the world's carbon stocks. Therefore, the agricultural soils of Sichuan wereselected as the research object. We collected the second national soil survey data in Sichuan Province, gatheredsoilin the hilly areas of central Sichuan andfour typical counties of Chengdu Plain in 2012-2013, integrated DNDC biogeochemical model, as well as 1:50,000 soil database infour typical counties,1:1,000,000 soil database in Sichuan Province, GIS databases ( such as weather, crop, tillage management). First, temporal variation characteristics of agricultural soils in the hilly areas of central Sichuan and four typical counties of Chengdu Plainduring the recent 30 years were analyzed, based on soil survey data of typical countiesin 1980 and 2012. Second, practical meas
Document Type学位论文
Identifierhttp://ir.imde.ac.cn/handle/131551/7868
Collection山地表生过程与生态调控重点实验室
Recommended Citation
GB/T 7714
罗勇. 四川盆地农田土壤固碳潜力模型模拟[D]. 北京. 中国科学院研究生院,2014.
Files in This Item:
File Name/Size DocType Version Access License
罗勇:四川盆地农田土壤固碳潜力模型模拟.(5861KB) 开放获取CC BY-NC-SAView Application Full Text
Related Services
Recommend this item
Bookmark
Usage statistics
Export to Endnote
Google Scholar
Similar articles in Google Scholar
[罗勇]'s Articles
Baidu academic
Similar articles in Baidu academic
[罗勇]'s Articles
Bing Scholar
Similar articles in Bing Scholar
[罗勇]'s Articles
Terms of Use
No data!
Social Bookmark/Share
File name: 罗勇:四川盆地农田土壤固碳潜力模型模拟.pdf
Format: Adobe PDF
All comments (0)
No comment.
 

Items in the repository are protected by copyright, with all rights reserved, unless otherwise indicated.