Knowledge Resource Center for Ecological Environment in Arid Area
| 青藏高原地区积雪与植被NPP变化响应遥感诊断 | |
| 其他题名 | Study on Interaction Remote Sensing Diagnosis Between Snow and NPP in Tibetan Plateau |
| 尹航 | |
| 出版年 | 2018 |
| 学位类型 | 博士 |
| 导师 | 曹春香 |
| 学位授予单位 | 中国科学院大学 |
| 中文摘要 | 青藏高原平均海拔在4000米以上，独特的地理条件造成其漫长的冬季有大量积雪覆盖，积雪产生的水热条件是植被生长的重要影响因素。青藏高原植被类型中高寒草甸、草原、灌丛所占面积比例最大，而植被净初级生产力（Net Primary Productivity，NPP）是草地生态系统最重要的参数之一。青藏高原属于全球变化的敏感区域，因此，青藏高原的生态系统更容易受到全球变化的影响，并进一步导致高原植被各植被参数发生变化。过去几十年青藏高原的气候发生了显著变化（年均温上升1.4℃），导致高原积雪与高寒植被都发生了明显变化。积雪作为植被生长重要的影响因素，研究青藏高原植被NPP与积雪的响应关系及响应机制对于了解NPP变化趋势具有重要的意义。本研究的主要内容是在环境健康遥感诊断的交叉学科框架下，针对作为生态系统健康指数之一的植被净初级生生产力，以时间序列遥感数据为基础，在宏观上对1983年到2012年青藏高原地区植被NPP和积雪相应关系进行遥感诊断。 本文的主要研究内容如下：（1）1983-2012年青藏高原地区长时间序列NPP与积雪数据集制作本研究基于长时间序列GIMMS3g-NDVI与气象再分析数据，针对青藏高原地区各植被类型最大光能利用率，应用CASA模型制作了1983-2012年植被NPP数据集，通过与实测生物量数据的对比，结果显示模型具有较好的估算精度。以长时间序列LTDR-AVH09C1与被动微波数据集，应用多元多时相数据融合模型制作了1983-2012年青藏高原积雪数据集，该数据集包含雪深与融雪期两个参数，以MODIS积雪产品与实测积雪数据为对比数据进行了精度评价，结果显示该数据集具有较高的反演精度。（2）1983-2012年青藏高原植被NPP和积雪变化时空演变分析研究首先分析了NPP和积雪参数的月均、年均分布的时空特征，以及近三十年的变化趋势；研究结果显示青藏高原植被NPP在8月份达到峰值，在11月至4月处于最低值，空间分布上呈现明显的由东北向西南逐渐增加的趋势，对比高程数据呈现明显的随海拔增加而减少的趋势；高原的植被NPP在时间序列上呈现总体增加的趋势，祁连山南部及念青唐古拉山北部的高寒草甸苔原地区呈现增加趋势较为显著。青藏高原的积雪覆盖在每年9月达到谷值，在1月与3月达到峰值，空间分布上在主要集中于唐古拉山、念青唐古拉山及昆仑山地区，随高程增加呈现增加的趋势，藏南谷地与川西地区呈现较少积雪分布的特征，时间序列上1998年冬季为积雪覆盖最多的一年，青藏高原积雪覆盖总体呈现增加的趋势，但并不显著，空间差异明显；青藏高原地区融雪期普遍呈现提前的趋势，其中昆仑山南麓地区融雪期提前趋势最为显著。（3）积雪与植被NPP相关性分析以及多种要素控制下的积雪与植被NPP相关性空间分异研究本研究区分了两种积雪参数和其他气候因素对青藏高原植被显著变化的影响，通过基于像元的线性相关分析分别提取了两种积雪参数影响青藏高原植被NPP显著变化的空间分布，并结合植被类型数据、地形数据、生态分区分析了1983年至2012年青藏高原植被NPP与积雪相关性的空间分异；植被生长期NPP对积雪的响应关系受到气候因素、地形因素和生态地理分区的影响。不同高程面的NPP与雪深、融雪期呈现不同的相关性特征，总体趋势呈现负相关，无论对于植被生长季的初期、中期还是后期，NPP与SD都在大部分高程面呈现显著的负相关，对于全年的NPP而言，仅当高程大于4600m是呈现显著负相关当降水量较高时，植被NPP与融雪期的正相关性维持在较高的水平，当温度较高时，植被NPP与雪深正相关性较高，当温度较低时，植被NPP与雪深、融雪期的负相关较为显著；当风速较大时，植被NPP与雪深、融雪期的负相关较为显著。以各生态分区为基础单元，高原山地高寒区（HIB1）、高原高寒荒漠区（HID1）、高原温带荒漠区（HIID2）植被NPP与积雪存在显著相关性，其他生态分区相关性不显著。（4）积雪对植被NPP响应关系遥感诊断首先，本研究针对生长期的不同阶段分析了单一要素与植被NPP的相关性，研究结果显示，青藏高原NPP与温度显著相关的区域随植被生长期变化而减少，并且是由西向东呈现面积逐渐缩小的趋势，高原腹地的高寒草甸苔原地区呈现显著的正相关；与降水相关性显著的区域则呈现随生长期的改变而增加的趋势，通过检验的相关性区域空间分布主要集中于藏南谷地地区及祁连山南部；NPP与融雪期相关性显著的区域面积呈现随植被生长而增加的趋势，空间分布主要集中于藏南谷地及川西等地区；植被NPP与冬季雪深在生长季初期仅在某些特定区域呈现显著的正相关，该区域主要集中在藏南谷地的小片区域，其他区域相关性不明显。其次，本研究基于归一化多元线性回归法以像元为单元分析对NPP影响的主导因素，研究选取了植被生长季的平均气温作为温度指标，在水分指标的选取上，我们通过比较降水量对NPP空间分布的影响，选取植被生长季土壤湿度数据作为积雪对植被影响的中间参数，选取上一个冬季的积雪融雪期与积雪深度作为积雪参数。通过分析NPP与四个气候因子（两条响应路径）的归一化多元线性回归分析来研究其后因子对植被变化的共同作用及各因子对植被变化影响的强弱关系。最终得到影响1983年到2012年青藏高原地区植被NPP变化的主导因子分布结果；结果显示无论哪条路径温度对于植被NPP的变化的影响贡献率最高区域面积最大，而降水和土壤水作为主要因素的响应区域则随植被生长期逐渐增加。由于在各影响要素中，积雪对植被NPP的影响并不显著，因此，本研究应用路径分析法诊断积雪对植被NPP间接响应关系，研究根据积雪对植被NPP响应机理确定了两条路径（影响植被的水热条件）：积雪-融雪期-植被生长以及积雪-土壤水-植被生长，结合路径分析法诊断两条路径下积雪对植被影响的主要路径以及其空间异质性，结果显示对于高原山地高寒区（HIB1），高原宽谷高寒区（HIC1），高山深谷区（HIIAB1）生态分区，雪深-融雪期-NPP的影像路径在植被生长的初期影响较为明显，对于HIB1与HIIAB1两个生态分区，雪深-土壤水-NPP在植被生长的后期作用较为明显；在其他生态分区，两条影响路径都不明显。 |
| 英文摘要 | The average elevation of the Tibetan Plateau is about 4 000 meters, covered with a large amount of snow in winter, and the hydrothermal condition caused by snow cover is an important factor for the vegetation growth. Alpine meadows, grasslands and shrubs lie the largest proportion among all the vegetation on the Tibetan Plateau. The Net Primary Productivity (NPP) is one of the most important parameters of grassland ecosystem.The Tibetan Plateau is one of the sensitive regions of global change. Small fluctuations in climate change will produce strong responses in plateau ecosystems, leading to changes in the pattern, process and function of plateau vegetation. Over the past few decades, great changes have taken place in the climate of the Tibetan Plateau (The average temperature rises 1.4℃ annually.), resulting in significant changes in snow cover and alpine vegetation on the plateau. Therefore, under the background of such climate change, it is of great significance to study vegetation productivity status on the Tibetan Plateau and analyze the response relationship and response mechanism of snow cover to NPP changes.The main content of this study is based on the subject idea of remote sensing for environmental health diagnosis. In view of the net primary productivity —— one of the indicators of ecosystem health, the advantages of time-series remote sensing data are exerted. NPP and snow parameters were quantitatively and qualitatively analyzed. Based on this, the influence of snow cover and other climatic factors on the changes of vegetation NPP was quantitatively analyzed, and the contribution rate of NPP changes in each snow cover parameter was analyzed. The main conclusions of this thesis include the following four aspects:(1) NPP and snow cover datasets from 1982 to 2012 in the Qinghai-Tibet Plateau. In this study, the long-time datasets——GIMMS3g-NDVI and LTDR-AVH09C1 were used as the basic data. CASA model and multivariate multi-temporal data were used to extract the NPP and snow information. The NPP dataset was simulated by solar radiation data and meteorological data to drive the CASA model to estimate the vegetation NPP. Compared with the measured biomass data, the results show a better precision. AVH09C1 from LTDR was first used to make snow products based on multi-temporal empirical model. The datasets further extracted two snow cover parameters of winter snow depth (SD) and snow melting period (SCM), and were verified by the comparison with MODIS data. (2) The monitoring of vegetation changes in NPP and snow cover on the Tibetan Plateau from 1982 to 2012. The spatio-temporal features of monthly average and annual average distribution of NPP and snow cover were analyzed. The results showed that the NPP of the vegetation on the Tibetan Plateau reached the maximum in August and minimum from November to April and spatial distribution showed an obviously increasing trend from the northeast to the southwest, contrast to elevation trend. The time series showed an overall increasing trend especially in the northern Qilian mountain and the northern part of the Nyainqentanglory. The snow cover reaches the lowest in September every year and the highest in January and March. The spatial distribution mainly concentrates on Tanggula, Nyainqentanglha and Kunlun Mountains as to the variation trend of elevation. The distribution of less snow occured in the southern Tibetan Plateau and western Sichuan. The snow cover in 1998 is the largest of all years, the overall snow cover on the Tibetan Plateau showed an increasing trend, but not significantly and heterogenicity is obvious. (3) Study on the relationship between single climate factors and vegetation NPP response. The effects of snow parameters and other climatic factors on the significant changes of vegetation on the Tibetan Plateau are distinguished. Two snow parameters are extracted respectively to the spatial distribution of significant changes of NPP on the Tibetan Plateau. Combined with vegetation type classifications and topographic data, the spatial variability of the driving force for the significant changes of NPP on the Tibetan Plateau from 1982 to 2012 was analyzed. The results showed that the areas of significant correlation between NPP and temperature on the Tibetan Plateau decreased with the vegetation growth period, and gradually decreased from west to east. Significant positive correlation was found in the alpine meadow tundra in the plateau hinterland. The areas of significant correlation between NPP and precipitation on the Tibetan Plateau increased with the vegetation growth period. The spatial distribution of the correlation area that passed the test mainly concentrated in southern Tibet Valley area and southern Qilian Mountains. The area of significant correlation between NPP and soil moisture showedan increasing trend with the growth of vegetation. The spatial distribution of NPP is mainly concentrated in the areas of southern Tibet Valley and western Sichuan. The NPP of vegetation and winter snow depth are only in some specific areas showing a significant positive correlation, the region mainly concentrated in the southern Tibetan valley Region, other regions of the correlation is not obvious; vegetation growing on snow NPP response relationship is affected by climatic factors, topography and ecological factors geographic partitions. There was no significant correlation between NPP and snow cover in areas with annual average temperature of less than -9 ℃ and annual precipitation of less than 150 mm. NPP had a significant negative correlation with snow in areas with temperature between -8℃ to 0℃. The correlation between NPP and snow is gradually weakened with an increasing temperature. The NPP with annual precipitation of 200-500 mm is significantly affected by the change of snow, while the area with annual precipitation of more than 500 mm is weakly affected by the snow. In terms of ecological zonation, the NPP of HIC1, HIC2 and HID1 showed a slight postponement trend. The NPP of other regions showed different degrees of advancement. The variation of snow varies greatly in different ecological units; NPP in the semi-arid subtropical zone of the plateau is obviously more sensitive to the change of snow cover than other regions. From the perspective of vertical terrain, the NPP and snowmelt gradually delays and the snow depth gradually increases with the altitude increasing. The NPP within the range of 4 700-5 400 meters was significantly affected by snow cover. (4) Remote sensing diagnosis of NPP responses of vegetation to snow was driven by multi-factors based on path analysis. In this study, the average temperature of vegetation growing season was chosen as the temperature index. By comparing the effects of precipitation on the spatial distribution of NPP, we selected the soil moisture (SM) data of vegetation growing season as the intermediate parameter of the impact of snow cover on vegetation. We select the last winter snow melting date and snow depth as snow parameters. By analyzing the normalized multiple linear regression analysis of NPP and four climatic factors (two response paths), the synergistic effect of subsequent factors on vegetation changes and the relationship between the various factors and vegetation changes were studied. Finally, the distribution of the dominant factors influencing the NPP changes of the vegetation on the Tibetan Plateau from 1982 to 2012 were obtained. The results showed that the temperature-related area with the highest contribution to the change of vegetation NPP was the largest in any path, while the response area with precipitation and soil water as the main factors increased gradually during the vegetation growth period. Furthermore, the path analysis method was used to diagnose the effect of snow cover on vegetation NPP based on remote sensing data and related meteorological data. Based on the two pathways of snow cover on vegetation NPP: snow - SCM- vegetation growth and snow cover -SM - vegetation growth, the main pathways affected by snow cover on vegetation were diagnosed by path analysis. The results showed that the impact of snow-soil moisture-NPP response of HIC1 and HIIC2 eco-zoning was more obvious while the more significant impact of snow-thaw-NPP response path was found in HIIAB1 ecological zoning. No matter which path of other eco-zoning response is not obvious. |
| 中文关键词 | 遥感诊断 ; 青藏高原 ; NPP ; 生态分区 ; 路径分析法 |
| 英文关键词 | Remote Sensing Diagnoses Tibetan Plateau NPP ecological zoning path analysis method |
| 语种 | 中文 |
| 国家 | 中国 |
| 来源学科分类 | 地图学与地理信息系统 |
| 来源机构 | 中国科学院遥感与数字地球研究所 |
| 资源类型 | 学位论文 |
| 条目标识符 | http://119.78.100.177/qdio/handle/2XILL650/288205 |
| 推荐引用方式 GB/T 7714 | 尹航. 青藏高原地区积雪与植被NPP变化响应遥感诊断[D]. 中国科学院大学,2018. |
| 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 |
| [尹航]的文章 |
| 百度学术 |
| 百度学术中相似的文章 |
| [尹航]的文章 |
| 必应学术 |
| 必应学术中相似的文章 |
| [尹航]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。
