退耕还林（草）工程区植被动态变化规律及影响要素
文献类型：学位论文
| 作者 | 王浩 |
| 学位类别 | 博士 |
| 答辩日期 | 2016-05 |
| 授予单位 | 中国科学院研究生院 |
| 授予地点 | 北京 |
| 导师 | 刘国华 |
| 关键词 | 生态工程，退耕还林（草），气候变化，人类活动，植被动态 ecological conservation, Grain for Green Project, climatic change, human activities, vegetation activities |
| 其他题名 | Changing trends and driving force analysis of vegetation activity in the Grain for Green Program region, China |
| 学位专业 | 生态学 |
| 中文摘要 | 近年来，随着社会经济的高速发展，人们对自然资源的利用强度不断加大，中国正面临着诸如土地荒漠化，土壤侵蚀，水土流失等一系列环境问题。针对这些情况，政府相继开展了包括“三北防护林”，“退耕还林（草）”，“天然林保护”等六大生态工程，来改善环境、解决问题。但针对生态工程自身效果的争议却始终存在。一方面，大量的研究结果表明生态工程在改善土地退化，增加土壤碳储量，修复植被等工作中起到了应有的作用。另一方面，同样有研究指出，受到自然条件的影响，一些地区的生态工程效果并不尽如人意。总体来说，关于生态工程效益的争议，主要是围绕自然条件的变化和人为活动影响究竟孰轻孰重展开的。生态工程区的植被退化，究竟是不恰当的人类活动还是极端气象条件引发的，温度、降水的变化是否会对生态工程的正效益产生大的影响，未来在评估生态效益时，是否应该将气象条件作为指标纳入其中，这些都是有待解答的。本研究针对退耕还林（草）生态工程区，利用2000 年-2010 年NDVI 遥感数据和相应时间段的气象资料，分析NDVI 和气象因素的时空变化特征，并针对不同季节和年际时间尺度分析了植被活动与气候的相互关系，同时利用残差分析结果分离人类活动对NDVI 变化的影响，并依照土地利用类型确定不同土地利用变化的贡献，最终根据NPP 变化趋势，对退耕还林（草）工程实施效果做一总结。研究中取得的主要结论如下： （1）退耕还林（草）生态工程区自2000 年至2010 年以来，气温、降水均呈上升趋势，增温率为0.27℃/10yr，年均降水变化率为4.032mm/yr。各季节气温、降水变化有显著差异，春、夏、秋、冬气温变化的线性趋势分别为：-0.047℃/l0yr、0.146℃/10yr 、0.629℃/10yr 、0.35℃/l0yr ，降水变化率分别为2.354mm/yr、1.161mm/yr、0.431mm/yr、0.087mm/yr。从空间变化趋势来看，近11 年来工程区年均气温、降水分布差异较大，基本都为自北向南，自西北向东南逐渐增加。研究区72.18%的区域温度呈现上升的趋势，各季节除春季外，其它三季的大部分区域温度同样呈增加趋势。秋季温度增加区域所占比例最大，春季最少。降水方面，研究区总面积74.05%的区域降水呈现上升趋势。四季中降水均有所增加，春季降水增加区域所占比例最大，夏季最少。从整体上看，2000 年-2010 年退耕还林（草）工程区气候向暖湿型转变，四季中除春季转向冷湿型，其它季节也均退耕还林（草）工程区植被动态变化规律及影响要素向暖湿型转变。 （2）退耕还林（草）工程区2000 年-2010 年NDVI 变化整体呈现了上升的趋势。其中年际平均NDVI 变化率为0.0009/yr。各季节年均NDVI 变化中，除冬季外，其余季节都出现增加趋势。秋季增幅最大，大于春夏增幅之和。从空间统计结果来看，近11 年来工程区59.4%的区域NDVI 呈现了增加的趋势，NDVI明显增加的区域主要集中在黄土高原区域。而对于不同季节来说，工程区四季的大部分区域NDVI 都呈增加趋势。秋季NDVI 增加区域所占比例最大，夏季次之，春季最少。 （3）退耕还林（草）生态工程区2000 年-2010 年NDVI 和温度、降水相关性分析表明，研究区53.02%的区域NDVI 与温度呈现了负相关，这些区域主要集中在北部干旱半干旱区以及研究区西南部。与之对应的则是研究区70.64%的区域NDVI 和降水呈现正相关作用，这充分说明了水分对该区域内植被活动的重要性。而从季节角度看，各季节NDVI 和温度、降水的相关性表明，在易发生干旱的春、夏季节，降水对植被生长意义重大。而秋季生长季末期时，如能维持适宜的温度和降水，将会有效延长植被生长期，增强植被活力。总而言之，本研究的相关结果不仅表明气象因素的变化会对植被活动产生一定的影响，还证明了这种影响会随着年际和季节的差异而产生明显的时间和空间变异性。因此，在未来有关植被活动的相关研究中，应将季节性因素的影响纳入评价体系。 （4）残差分析结果表明，退耕还林（草）工程区中54.48%的区域经历正向人类活动的影响，其中显著正影响区域主要集中在黄土高原地区。整个NDVI增加区域中，只有11.72%的区域在经受人类活动负影响之后NDVI 依然保持了上升趋势，这些区域主要分布在干旱半干旱区内，NDVI 的变化和降水呈现明显的正相关效应。同时，NDVI 上升区域内近90%的区域都与人类活动的正效益有关，特别是黄土高原和西南地区的一些区域在气象条件相对不利的情形下依然保持了NDVI 的增长趋势，说明了人类活动在恢复植被，克服不利自然因素方面起到的巨大作用。整个NDVI 减少区域中，只有5.02%的区域在经受人类活动正影响之后NDVI 依然保持了下降趋势，这些区域都受到了类似干旱或霜冻等极端天气的影响。但超过90%的NDVI 下降区域都与人类活动的负效益有关，人类活动对植被生长的负面影响依然显著。 （5）2000 年-2010 年退耕还林（草）生态工程区各土地利用类型中，林地和人工用地都呈增长趋势，其相对于2000 年净增长百分比分别为0.77%和21.67%。在减少的土地利用类型中，田地的面积减少最大，其减少的面积占发生变化总面积的40.74%。总体来说，林地增加面积的主要来源是田地，这类转变的发生符合退耕还林等生态工程实施之后的结果，说明生态恢复取得了明显的效果。贡献率分析的结果说明，草地类对NDVI 变化的贡献最高，其次是田地和林地。而对于发生变化的土地利用类型来说，高贡献率排名前五位除了草地向林地转化外，其余四位都是田地和草地或林地的相互转化。NDVI 增加区域各土地利用类型贡献率统计结果表明，田地向草地和田地向林地的转变对该区域NDVI 变化有着显著影响。NDVI 下降部分的贡献率统计结果表明，各类有植被用地向人工用地的转变，草地和林地向田地的转变所造成的负向影响最为显著。可以看出，不论是在量变还是质变的程度上，农田，草地和林地之间的相互关系都是决定NDVI变化的重要因素。 （6）NPP 计算结果表明，退耕还林（草）工程区2000 年-2010 年NPP 呈现增加的趋势。其中，CASA 模型的计算结果明显高于MOD17 数据的分析结果。NPP 空间分布情况表明，CASA 模型对NPP 的估算结果更符合植被活动的变化趋势。研究区NPP 整体上呈现从西北向东南增加的趋势，CASA 模型中整个研究区65.96%的区域NPP 呈现增加的趋势，其中显著增加区域主要分布于东北平原，黄土高原和甘肃省南部等区域。MODIS 数据中NPP 增加区域占研究区总面积的65.55%，显著增加的区域为黄土高原地区，青海省东部以及横断山区等地区。根据贡献率分析结果可以看出，几种主要的植被类型中，林地类对NPP 变化的贡献最高，其次是田地和草地。对于发生变化的土地利用类型来说，贡献率排名前五位除了草地向林地转化外，其余四位都是田地、草地和林地的相互转化。 结合之前的研究结果，我们可以看出，林、草、田三种土地利用类型的相互转化，不论是在植被活动强度，还是在NPP 的变化上都会对整个植被生态系统产生显著的影响，这种影响甚至高于有植被的土地利用类型（林地，草地，田地）和无植被的土地利用类型（人工用地，其他类用地）之间的相互转化。综上所述，尽管气象条件能够影响NDVI 变化趋势，但人类活动是NDVI 变化中更为决定性的因素。生态恢复措施能够给当地环境带来各种收益，恰当的恢复措施能够克服不利气象条件带来的影响，维持植被的活力。退耕还林（草）生态工程的实施，使得大量田地转化为林地和草地，有效的促进了植被活动的增强和NPP 的增加，提高了当地的生态环境质量，生态恢复措施的效果得到了显著体现。因此在未来生态工程实施的过程中，应该注重和当地环境条件相结合，充分发挥正向人类活动的主导作用，选择适当的生态恢复措施，将生态恢复效果最大化。 |
| 英文摘要 | Along with the rapid economic development, the environmental problems of China have become increasingly serious during the past decades. To solve those problems, including desertification, soil erosion and sand storms, the Chinese government has initiated six ecological restoration programs including the “Three-North Shelterbelt Project”, the “Natural Forest Protect Project” and the “Grain for Green Project”. However, the debate about the effectiveness of the ecological restoration programs is still ongoing. On one hand, many studies found that the ecological restoration programs improved the land deterioration, increased the vegetation coverage and soil carbon storage successfully. On the other hand, several studies pointed out that limited by the climate change, the ecological restoration programs didn’t work well in some regions. In general, the debate over the effectiveness of the programs is mainly focused on recognizing a more dominating player, the climate or the human activities. Many issues still demand answers: whether the decrease of the vegetation activity in the program regions was caused by the climate change or the improper human activities; whether the climate can cause a big impact on the results of the programs; should climate change be considered when assessing the benefits of ecological restoration programs. To answer these questions, this study chose the Grain for Green Program (GGP) region as the study area. We used NDVI remote data and the climate data in the GGP region to identify the spatio-temporal patterns of the vegetation and the climate from 2000 to 2010. Targeting different seasonal and interannual time scale, we analyzed the correlation between the vegetation activity and the climate change. And based on the result of the residual analysis, we separated the effects of climate and human activities on the NDVI. The final objective of this study is to investigate the role of the human activities in changing vegetation growth by calculating the contribution rate of different land use types, and to estimate the effectiveness of the ecological restoration program with the changes of the net primary productivity (NPP) in study area. The main results are as follows: (1) Analysis reveals that the temperature and the precipitation in the GGP region increased respectively by 0.27℃/10yr and 4.032mm/yr from 2000 to 2010. Both the temperature and the precipitation for four seasons have their own trend that differs from that of the interannual scale. More specifically, the change rates of the temperature/precipitation for four seasons are -0.047°Cl0yr-1/2.354mmyr-1, 0.146°C l0yr-1/1.161mmyr-1, 0.629°Cl0yr-1/0.431mmyr-1 and 0.35°Cl0yr-1/0.087mmr-1,respectively. Climatic changes in the GGP region are not only reflected on the time scale, but also in the spatial scale. Basically, both of the temperature and the precipitation are increasing from north to south, and from northwest to southeast. 72.18% of the area shows an increasing trend for annual temperature. The seasonal temperature keeps increasing in most of the areas in the GGP region, except for spring. A total of 74.05% of the area in the GGP region represents an increasing trend for precipitation. And most area displays an increasing trend for precipitation in all seasons. Overall, the climate in the GGP region is becoming warmer and wetter, but in spring exclusively, the climate in the study area is becoming colder and wetter. ( 2) NDVI in the GGP region showed an increasing trend over the last 11 years with a rate of a change of 0.0009/yr. In spring, summer and autumn, NDVI shows an upward trend, but in winter, it decreases. The autumn NDVI has the greatest change rate, bigger than the sum of the changing rates of the spring and the summer. The spatial distribution of NDVI showed distinct characteristics in the GGP region from 2000 to 2010. 59.4% of the region showed an increasing trend for the NDVI, and the significant increasing area was mainly located in the Loess Plateau. The seasonal NDVI keeps increasing in most of the areas in the GGP region for all seasons, and autumn witnesses the largest increasing rate while spring, the lowest. (3) To further validat the correlation between the NDVI and the climate, we calculated the correlation coefficient between the NDVI and the climate variables (temperature and precipitation) for each pixel. The results show that, 53.02% of the study area showed a negative correlation between the NDVI and the temperature. These areas were mainly located in the arid and semiarid regions and the southwest of the study area. Contrarily, 70.64% of the study area witnessed a positive correlation between the NDVI and the precipitation, which indicate that precipitation plays an important role in determining the activities of the vegetation in the GGP region. On the seasonal scale, we found that precipitation is much more important than temperature for the vegetation growth in spring and summer, because high temperature will easily cause drought and reduce the vegetation activities. And in autumn, the results prove that maintaining the appropriate temperature and rainfall can enhance the photosynthesis of vegetation, and extend the growth process. In genera, the findings of this research not only underline the importance of climate variables for vegetation growth, but also suggest that the effects of seasonal climate variables on vegetation should be explored further in related researches in the future. (4) The result of residual analysis showed that, 54.48% of the GGP region experienced human-induced improvement. The significantly improved areas were mainly located in the Loess Plateau. In areas with an increasing NDVI, only 11.72% of the areas were led by the favorable climate, while human-induced degradation was found in these areas. Nearly 90% of areas with increasing NDVI are positively influenced by human activities, though some of them suffered bad climate condition. In areas with a decreasing NDVI, over 90% of them are negatively affected by human activities. Only in 5.02% of the areas, the extreme weather such as drought and frost, offset the benefit of human-induced improvement and made the NDVI decreased. As a result, although the climate can affect the NDVI changing trend, human activities can change the NDVI trend more effectively. Thus, human activities play a more important role in changing the vegetation activity. (5) The structure of land use in the study area changed greatly over the last 11 years. Forest and artificial land increased significantly in size. The net increase rate for these two land use types are 0.77% and 21.67% respectively. Meanwhile, the area of cropland decreased the most. Because of the GGP and other ecological restoration programs, large parts of the cropland were converted to the forest. Some major progress was made in ecological programs. The results of the contribution rate analysis showed that, grassland made the highest contribution to the change of the NDVI, and the cropland and forest are next. Different land use change types made different contribution to the NDVI change trend, and the analysis results show that, along with the transform from grassland to the forest, changes between the cropland and the grassland/forest contribute the most. In the NDVI increased areas, cropland converting to the grassland/forest has significant effects on the NDVI. And in the NDVI decreased areas, two main types of land transformation have the highest contribution rate: vegetation land use type transforming to artificial land, and grassland/forest to cropland. In general, the conversion between the cropland and the grassland/forest plays an important role in the vegetation change trend. (6) According to the analysis results, NPP in the GGP region kept increasing from 2000 to 2010. The NPP calculation result based on the CASA model is obviously higher than the result based on the MOD17 remote data. The spatial distribution of NPP indicated that the CASA model is a better way to estimate the NPP change trend. Basically, the NPP in the study area is increasing from northwest to southeast. Based on the CASA model, 65.96% of the areas showed an increasing trend for NPP, and the areas with major increasing NPP were mainly located in the Loess Plateau and the southern part of the Gansu province. Meanwhile, based on the MOD17 remote data, 65.55% of the areas showed an increasing trend for NPP, and the significant increasing areas were mainly located in the Loess Plateau, the eastern part of the Qinghai province and the Hengduan Moutains region. The results of the contribution rate analysis showed that, forests made the highest contribution to the change of the NPP, followed by cropland and grassland. Similar to the results of the NDVI, along with the transformation from grassland to the forest, the converting between the cropland and the grassland/forest contributes the most to the NPP change trend. Combining with the analysis results of the NDVI, we can find that the conversion between the forest/grassland and the cropland can easily cause significant effects on the vegetation activities and the NPP change trend. Sometimes, these effects are more powerful than the conversion between the vegetation land use types (forest, grassland and cropland) and the non-vegetation land use types (artificial land and others). In conclusion, though climate conditions can affect the trend of NDVI, human activities are far more important for NDVI changes. Ecological restoration programs can bring various benefits to local environment. The implementation of the GGP made a large part of the cropland converted to forest and grassland, and effectively increased the vegetation activities and the NPP in the program area. Proper human activities can maximize the benefits of the ecological restoration program and minimize the effects of the extreme weather conditions. Therefore, for future ecology projects, we should be more cautious about the measures of ecological restoration as to maximize the restoration result in face of extreme climate conditions. |
| 源URL | [http://ir.rcees.ac.cn/handle/311016/36971]  |
| 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 | 王浩. 退耕还林（草）工程区植被动态变化规律及影响要素[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。