从太空看中国空气污染全景

从太空看中国空气污染全景 徐安琪 2012年2月20日 美国科研人员通过卫星数据对中国十年间的PM2.5污染浓度进行了推算。科研团队成员徐安琪揭示了研究结果。 “卫星空气质量监测能够弥补地面监测基站建设不足所造成空间断层及信息缺口。” 相关文章 北京“蓝天”不蓝，问题出在哪, 2011年12月12日 北京的“蓝天”掩不住健康风险 2011年12月5日 清理致命灰霾 2011年10月19日 从遥远的太空俯视地球，或许我们无法看清中国的长城，但是却可以将这个国家的空气污染尽收眼底。 最近，我们在耶鲁大学、哥伦比亚大学和巴特尔研究所的研究团队根据卫星数据对中国各省的可吸入颗粒物浓度进行了研究。尽管这些卫星数据还不尽完美，但是我们依然可以借此对过去十年里中国全境PM2.5污染物地表浓度年平均值进行初步估算。 PM2.5也就是空气中直径小于等于2.5微米的可吸入颗粒物。由于这种细小颗粒 物能够渗入人体肺部组织和血液，带来哮喘、癌症、心血管疾病等健康隐患，因此，近几个月来在中国引起了公共安全人士的广泛关注。 那么，如何利用卫星数据对PM2.5污染进行监测呢,简单说来，就是利用卫星搭载设备对气溶胶光学厚度(AOD)进行测定。气溶胶粒子通过吸收散射能够对光线产生消减作用。利用其这一特点就可以对颗粒污染物的地面指标进行监测。 目前有些研究发现，AOD与地面监测指标之间存在着一定的数学关系，并为此建立了相应的数学模型。当然，用大气垂直气柱内的气溶胶光学厚度来表征地表污染程度是一个很复杂的过程，这取决于气溶胶的垂直结构、组成、粒径、分布、水含量等多种因素。所以，地域差异和气候差异都会对PM2.5的测定造成影响(详细测定方法请参见该文)。 PM2.5浓度指标以各省市的平均大气污染暴露浓度表示。将全省分成一定的网格区域。各网格内根据卫星数据得出的PM2.5浓度乘以该网格内居民占全省人口总数的百分比，即计算所有网格区域的大气污染暴露浓度的人口加权平均值后便得到人口加权大气污染暴露水平。 也就是说，相对于郊区或人口稀少地区而言，人口稠密地区在全省PM2.5平均值中所占比重更大。该方法充分考虑了地广人稀的低污染地区以及人口稠密的高污染环境(或者反之)的情况。 因此，人口加权值更注重可吸入颗粒物对居民的实际影响。简而言之就是，通过这些数字可以了解中国某省普通居民某天所承受的一般空气质量情况。 那么，这些数据又让我们看到了什么呢, 正如下面的中国PM2.5变化趋势图所示，除了四省之外(不包括台湾)，全国各省、自治区及直辖市的PM2.5年均浓度都高于世界卫生组织(WHO)建议的水平。从图一中我们可以看到，2007年，中国31个省、自治区和直辖市中，大多数地区的人口加权细颗粒物污染浓度都超过了世卫组织每立方米10微克((/m3)的年均空气质量标准。 图一，2007年中国各省人口加权细颗粒物年均污染浓度(PM2.5)包括台湾 而通过图二和图三的时间序列我们可以了解到中国各地区PM2.5污染的变化趋势。其中，山东、河南两省的PM2.5污染最为严重。北京、上海、广东的PM2.5年均浓度虽然在过去三年有所下降，但是在过去九年里基本都维持在一个较为稳定的水平上。而且，也正如我们所料，西藏、内蒙等西部欠发达省份的细颗粒污染浓度最低。 图二，中国部分省市PM2.5年均浓度时间走势图(2001-2010)当然，任何模型都存在不确定因素，本研究也不例外。利用卫星进行AOD观测时，如果遇到积雪或沙漠等地表亮度较高的情况，其准确度就会受到影响。同时，这种方法也无法提供污染物的垂直分布信息(例如，它无法分辨距离地表不同高度的粒子)。就我们所使用的模型而言，其不确定性大约在?25%之间。也就是说，误差在每立方米6.7毫克左右。 当然，卫星测绘与地面数据之间也并非严丝合缝。地面实测数据更为准确。可是，卫星空气质量监测却能够弥补地面监测基站建设不足所造成空间断层及信息缺口。同时，卫星能够对空气质量进行持续稳定的监测，从而便于不同时期、不同地区之间进行比较。面对着高涨的舆情，这样的检测手段正是中国决策层所需要的。 表三，中国各省市自治区(包括台湾) PM2.5年均浓度时间数据表(2001-2010)点击看大图) ( 徐安琪，耶鲁大学森林与环境研究学院博士生，2012年环境绩效指标项目负责 人。 Chasing data in China’s provinces Angel Hsu February 21, 2012 The first independent review of provincial-level environmental data highlights a plethora of challenges to transparency in China, writes Angel Hsu, introducing the second of two new studies. “It proved difficult to assess the validity and reliability of official statistics, giving us concerns about how much they reflect the reality of on-the-ground conditions.” Related articles Seeing China’s pollution from space February 20, 2012 Beijing’s hazardous blue sky December 05, 2011 Clearing the haze October 19, 2011 China’s environmental data has created many international headlines in recent months, particularly its controversial air-quality measurements. While Deng Xiaoping urged the Chinese citizenry to “seek truth from facts”, China is still a long way from providing the environmental data and information that allows for just that. My colleagues and I recently released a study that provides a detailed analysis of provincial-level environmental data in China. We introduced a model framework for environmental performance indicators to assist the Chinese government in tracking progress toward policy goals, as well as recommendations for how the Chinese government can apply more aggressive performance metrics to environmental decision-making. In total, we looked at 32 indicators in 12 environmental policy categories (among them air pollution, water quality, climate change, biodiversity, agriculture and forestry). The data we reviewed and used to construct these indicators were all derived from official Chinese statistics. We concluded that the lack of clear policy targets for many environmental metrics in China, as well as concerns over data sources and transparency, hamper the government’s ability to effectively address pressing environmental issues at the provincial level. While the report elaborates these challenges in detail, a few of the main findings are summarised below. First, the existence of baseline environmental data is highly uneven. To develop performance indicators that evaluate the efficacy of environmental policies, baseline data are necessary to benchmark performance. Less than half the indicators evaluated had this. Baseline data were most prevalent for economic sustainability indicators (68%) and least prevalent for ecosystem vitality indicators (20%), while environmental health indicators were in the middle (42%). This pattern reflects the priorities of Chinese environmental policymaking in the past decade, which has emphasised pollution control and resource efficiency in the industrial sector. Second, difficulties in accessing raw data hinder data quality evaluation. Our report provides pilot indicators based on official statistics; however, we did not have the ability to independently evaluate those statistics. We found that official statistics for most indicators lack detailed information on data collection methods and monitoring systems, and in no instance were we able to obtain raw data from monitoring stations. Nor were we able to obtain data from third parties that might have been used to corroborate official statistics. For all these reasons, it proved difficult to assess the validity and reliability of the official statistics. These difficulties gave us concerns about how much the official statistics reflect the reality of on-the-ground conditions. Third, ongoing measurement systems are also highly uneven. Consistent measures, produced on a regular basis, following established methodologies, in a transparent and verifiable manner, are critical for environmental performance monitoring. In China, the measurement systems related to industrial efficiency are exemplary models. In this area, the published data meet the foundational requirements and, as a result, permit operational use of performance indicators in the five-year plans. The other measures generally fall short. For example, methodologies for ecosystem measures tend to change over time, making comparison problematic, and the metrics used to measure air and water quality are transformed in ways that make tracking performance difficult. And fourth, policy targets for the vast majority of candidate indicators are not easily identified. Overall, we were able to establish a basis for constructing a policy target for 21 of the 33 indicators we included in our framework — eight in the environmental health objective, seven in the ecosystem vitality objective and six in the economic sustainability objective. We considered 50 additional indicators but did not include them because of the lack of clear policy targets by which to gauge performance. However, the lack of properly specified policy targets is not unique to China. Similar challenges around goal-setting exist in many countries, especially in the developing world. A major theme underpinning all of these conclusions is the need for greater data and information transparency. Even though laws on the disclosure of environmental information came into effect on May 1, 2008, China still has a long way to go in terms of providing environmental data transparently. Researchers at Chinese NGO the Institute of Public and Environmental Affairs and the US-based Natural Resources Defense Council found that most big cities in China failed to publish adequate pollution information in 2011 in the third edition of the Pollution Information Transparency Index (PITI), released last month. Only 19 out of 113 cities received passing scores for information transparency. The authors concluded that environmental information disclosure is an “innovative system” in China that does not, so far, go beyond the initial stages. It is our belief that the value of both the PITI and our report, “Towards an Environmental Performance Index in China”, lies in being able to provide transparency to environmental data and results in China. Transparency and access to information are fundamental tenets of sound environmental policymaking. Greater transparency can stimulate research and policy for developing innovations that can only help China navigate the difficult path of sustainable development. Angel Hsu is a doctoral student at the Yale School of Forestry and Environmental Studies and project director for the 2012 Environmental . Performance Index Homepage image by Riskind Tags: Green_monitor Governance Print article Email this article to a friend Share