美国专家澄清：没说误报损失跟地震同样惨重

          美国专家澄清：没说误报损失跟地震同样惨重

                  作者：直言了  2008-05-25

    许多报刊媒体现在还流传着新华社05月20日发表的署名报道《美地震专家认为短期临震预报是世界性难题》。该报道的采访对象美国地震学家露西尔·琼斯博士、美国USGS（美国地质勘探局）的南加州课题首席科学家。应本人核实查询要求，琼斯博士把接受新华社记者采访的回答全部原文给本人寄了一份。通过对比可以清楚看到，该报道大部分内容与原文不符，不少内容是把中国媒体和地震官员的话塞到美国专家琼斯博士的嘴里，给读者社会造成了严重误导。

    详细对比说明需要很多文字，这里就说些特别突出的例子。

    报道说：“对地震的精确预测，不仅要预报出准确的时间和地点，还应该预测出震级。琼斯认为，目前世界上还没有这样的技术。” 

    事实是：琼斯博士原本结论并非是报道说的一般化结论，而是有条件的结论。正因为是有条件的，所以，在回答问题的时候，琼斯博士对记者问题做了个鲜明的反问：“如何回答要看您所谓的‘预报’指的是什么”。

    在回答的时候，琼斯博士很明确地说明：如果所谓地震精确预报指的是时间，那么，在地震的起始和结束之间没有关联、且震级信息在震前不存在的条件下，预报地震几乎是不可能的。实际上，在前震、主震和余震彼此相关的情况下，琼斯博士所在的USGS可以做出足够准确的地点和震级的预测预报（用琼斯博士自己的话说，“我们（USGS）做得很好”），但还没能做到极短时间内临震的足够准确的时间预测预报。对比原文和报道，可以明显看到，那篇报道把有条件的结论搞成一般化的无条件结论了。客气说，那种报道是夸张；不客气说，那是篡改。

    报道说：“从另外一个角度说，即便这些异常迹象可以作为预报地震的参数，地震监测人员也不能仅仅依靠某一个单独的异常事件做出地震预报，因其可能只对应极小的发震几率。一旦误报地震，损失往往同样惨重。”

    事实上，在琼斯博士的原文里，那些文字根本就没出现。琼斯博士的原文表达的是完全相反的立场，即：不做预报而发生地震、比误报带来的伤亡损失和安全威胁要大得多，因此，很需要加强预报工作及相关科研。这个立场，在本月22日USGS发布的南加州地震模拟预报中做了充分表达。该预报假设2008年11月23日上午10点、在南加州发生M=7.8级以上的大地震，阐明了那将给当地带来什么样的灾难和损失、说明了如何应对来减少伤害和损失。那篇模拟预报清楚地表明：模拟预报的地震可能根本不会发生、或发生了也是情况大不相同，但是，这个预报所提供的经验教训，可以应用于许多其它地震事件、有助于应对许多将来可能发生的地震事件。

    报道说：“在中国以及世界其他一些国家，都有不少地震震前预兆的民间说法，比如动物的异常行为、奇特天象出现等。琼斯说，这些说法到目前为止，都缺乏十分确凿的科学依据。”

    事实：这样的报道篡改了琼斯博士的答复本意。“在中国以及世界其他一些国家，都有不少地震震前预兆的民间说法，比如动物的异常行为、奇特天象出现等”这句话，不是琼斯博士所言，而是新华社记者提问所言。在记者提问原文里，也没有“世界其他一些国家”和“民间说法”的字样。比较关注的读者都知道，那言论来自中国地震官员，而所谓“民间说法”不过是“民科”的一种说法，是“官本位”学术制度下实行关门主义、学术小圈子垄断化和封杀不同见解的借口，是用“民间科学”作为“不科学”甚至“伪科学”的代名词来对不同见解做政治化的封杀。

    相关问答原文如下：

    记者问：In China, people sometimes will predict earthquake or other natural disasters according to some unusal event, for example, the unusal behaviors of animals, some strange astronomical phenomena, etc. Is there any scitific evidence or research to these?

    琼斯答：Not really. Every effort to prove these (like the Collaboratory for the Study of Earthquake Predictability) has shown there is no solid signal.

    就此把动物表现异常一般化来预报地震的做法，琼斯博士表示的立场也是美国地质学界一般化的立场，而不是有具体针对性的立场。譬如，琼斯博士所提到的实验，是30多年前做的一般化实验，试图证明动物异常行为和自然灾害之间的一般化关系，且实验对象大都是离人类社会很近的动物，结果以失败告终。然而，就有具体针对性的地震预报而言，美国地质学界并不否认动物异常行为和地震之间存在某种震前关联。譬如，他们很关注日本地震学界的相关科研发展；又譬如，对中国的海城地震预报，美国甚至全球地质学界多数科学家都承认它的各个方面的成功（当然，有少数专家坚持认为那是侥幸和心理学的结果）。

笔者议论：

    琼斯博士的回答，反映了西方学界“老科学”和“新科学”的冲突。琼斯博士的见解属于牛顿物理时代开始到19世纪末期的地质学和地理学。而美国宇航局（NASA）的见解则属于20世纪开始的现代科学和实现宇航以来的当代科学，它所开展的地震科研成果，十年进展胜过地质学一个世纪。

    具体说，美国宇航局汇集相关领域科研而提出了一个理论：地壳运动造成地震，但又是什么造成和诱发了地壳运动而发生地震的呢？必须寻找和观测造成地壳运动的动力和诱因。地核能量释放是个关键因素，而星球引力作用也是个重要的诱因，因此，除了古典科学所阐述的地壳运动观测手段之外，还必须有卫星手段，必须有电磁、光学、天文、天象、以及现代物理学等现代科技手段，就可以做到地质学难以做到的“入地”观测。

    基于这样的理解，美国宇航局开展的地震观测科研十年左右取得突飞猛进的进展，特别是在地震预报的地点和震级两个要素方面的进展十分突出。例如，2002年02月，该部门在美国国家科学院发布了地震观测研究报告；到2004年10月，在五级以上地震的16个事件中，有15个与研究报告期望值吻合。就是说，在临震预报准确率所要求的三大要素中，地点和震级的预测预报获得了突破、尚待突破的是时间预测预报。就此进展，美国宇航局说：目前，精确预报地震依然做不到，但是，朝着这个方向发展的大门已经打开了。

    对美国宇航局获得的突破性进展，美国地质观察部（USGS）并不掩饰自己的“争风吃醋”。譬如，琼斯博士在回答相关问题的时候，很痛快地对NASA科研进展表明了“不屑一顾”的看法：“NASA声明的那些成绩，除了他们自己，没人同意。我们USGS做的更好。”琼斯博士是位在美国学界享有盛誉的科学家、在国际学界也颇有名气，可表现却象是一个天真无邪的少女。也许正是这种天真无邪的儿童心态，造就了美国一批又一批世界水平的科学家。

    就地震预报而言，比较美国地质观察部（USGS）和美国宇航局（NASA），可以很明显地感到“老科学”和“新科学”的不同：前者基调是保守和悲观的、试图告诉您他们不能做什么；而后者基调是创新和乐观的，试图告诉您他们可能能做什么。美国是崇尚乐观、未来和创新的社会。也许这也是个缘故吧，USGS全国科研经费只获得到5000万美元的年度拨款，不及一个“财富百家”跨国公司R&D的开支；而后者NASA所获得的年度科研经费经常是以十亿或几十亿美元计算。

    在琼斯博士回答中，本是有上述内容的。可是，四川大地震发生后，某些地震官员为“地震不可预报”理论辩护说，若有个地震预报说明了发生地点、震级和时间，那就是谣言传言。对比美国宇航局获得的科研进展看，显然，那些官员的知识是过时和陈旧的知识。也许是为了迎合那些地震官员的辩护需要吧，在发表报道的时候，有关美国宇航局地震预报科研进展和琼斯博士的看法等内容，全都被砍掉了、一个字也不提相关科研新发展，很明显地是有意识地引导民众从现代科技倒退到19世纪及以前的科技时代。

    简而言之，许多媒体流行的新华社05月20日署名报道《美国认为临震预报是世界性难题》，是篇违背事实的新闻报道，是篇试图把某些中国官员编制的“地震不可预报”和“震前无预兆”等涉嫌推卸责任的舆论塞到美国专家嘴里、用美国专家的名义来误导读者社会的宣传新闻。后面是琼斯博士发来的问答采访问题的原文全文，公布出来以正视听。

    顺便说一下：就本人的核实查询，USGS的其他科学家也说明了看法，特别是针对所谓“预报”发表了学术见解。可以明显看出，中国的某些新闻媒体和地震官员混淆了关于地震工作的“预测”、“预报”和“警报”，并用“警报”概念替代“预测”和“预报”。这些讨论，即将陆续发帖探讨和供各位参考。

美国USGS首席科学家琼斯博士接受新华社采访所做回答的原文：

（发自：琼斯博士本人，2008年05月23日，美国加州时间14:53）。

THE TRAN SENT FROM DR. LUCILE JONES

To: zhiyanle 。

Subject: Re: 5/12 quake & fact-check.。

Date: Fri, 23 May 2008 14:53:18 -0700。

You asked for information about my Xinhua interview. Below is the email by which I responded to them.

Lucy  

----

Hi Summer （这里，Summer 是新华社记者提供的英文姓名）- 

1: ScienceNow reported that you are familiar with the Longmenshan fault, which has caused the Sichuan earthquake? Can you give what you know about the fault, and this major earthquake?

Is it an active fault from the long-time scale? will there be other major earthquakes in the next decades? or from continental scale, what's the situation about the India Plate and Eurasia plate?

The Longmenshan fault is part of the extensive system of faults in western China that is carrying China to the east - to get out of the way of India's continue move into Eurasia. I found a good general interest discussion of the India-Eurasia collision at http://www.shangri-la-river-expeditions.com/wchinageo/wchinageo.html。

Here is a geologist's picture of Asia:

This figure was created by Peter Molnar and Paul Tapponier in 1975 - and you can see the Long Men Shan thrusts labeled here. So this is a significant enough fault that it shows up at the continental scale. Movement on this fault pushes up the eastern edge of the Tibetan plateau and the fault is the eastern boundary of the plateau.

The Longmenshan is an active fault, moving many mm/yr - a recent study had 15 mm.yr. This makes it about half the speed of the San Andreas but that is still fast geologically.

2: Currently, predicting earthquakes is still elusive, what are factors affecting the prediction of earthquakes? however, is there any possibility to predict?

This answer depends on what you mean by predict. There is important information about the spatial distribution of faults - where the earthquakes will happen. At that level, we do an excellent job of predicting the spatial distribution of earthquakes and as I just said, the Longmenshan is a very active fault and was well understood before this earthquake.

But what you probably mean is predicting the time of large earthquakes. But since small earthquakes happen on the same fault and happen much more often, the prediction you want requires predicting the magnitude of a future earthquake as well as the time. The problem is that the magnitude is determined during the earthquake by how far the rupture that starts at the epicenter moves down the fault. If it starts and travels 100 meters it will be about magnitude 4. If it starts and travels 300 km it will be close to magnitude 8. But what stops the earthquake may not be connected to what makes the earthquake start and so information about the size may not be in the earth before the earthquake begins. If that is true, prediction as people want it is impossible.

2:As you know, is there any successfull instances of prediction? especially in the California where earthquaks often occurs.

The clearest prediction of a big earthquake was the 1975 Haicheng earthquake. A very complete evaluation of what happened in that earthquake was published in the Bulletin of the Seismological Society of America. If you call them, they could send you the paper.

Every earthquake makes another earthquake more likely. Mostly those earthquakes are near the first one and smaller and then we call them aftershocks. But about 5 percent of the time the triggered earthquake is bigger than the first one and we change the names and call the first one a foreshock and the triggered event is the mainshock. The triggering also dies off with space so sometimes the triggered event is a long way from the mainshock and is then called a "triggered event" instead of "aftershock." So after the first event happens, the probability of another event is much higher than it usually is.

In California, we have a website that shows that probability (http://pasadena.wr.usgs.gov/step/). Because of our building codes, predicting California earthquakes often doesn't save many lives and can cause a lot of deaths - if a lot of people jump on the freeway to get away, they may be more likely to die in a traffic accident than they would have in the earthquake. In China, more people die in the earthquake and are less likely to die in the prediction, so the same information may turn into a prediction in China but not in California.

3: I google-searched an article at Nasa website

http://www.nasa.gov/centers/goddard/news/topstory/2004/0929earthquake.html

It seemed that NASA's program accurately predicted the locations of 15 of California's 16 largest earthquakes this decade, what do you think of this prediction programs?

NASA claims this, but no one else agrees. The Southern California Earthquake Center (SCEC) compared this prediction to the National Seismic Hazard Maps and found the Maps did a better job than the NASA study. SCEC has started a new project called the Collaboratory for the Study of Earthquake Predictability to do these comparisons. (see http://www.cseptesting.org/)

4: why it is so difficult to predict earthquakes? can you explain the characteristics of earthquake activities?

5:Short-term prediction seems nearly impossible, for example with several days or even several hours. But maybe long-term prediction can be sort of possibility, what's the situation of long-term prediction worldwide?

See above for why it is so difficult. Also, there are many earthquakes every day - they are just usually small. "Long-term prediction" is usually based on fault history and that does OK. Short-term is difficult unless the earhtquake has a foreshock.

6: In China, people sometimes will predict earthquake or other natural disasters according to some unusal event, for example, the unusal behaviors of animals, some strange astronomical phenomena, etc. Is there any scitific evidence or research to these?

Not really. Every effort to prove these (like the Collaboratory for the Study of Earthquake Predictability) has shown there is no solid signal.

7: As for USGS, is there any techniques or programmes predicting or monitoring the earquakes? What's your valuable experience in working at South California, How is the situation about earthquake in United States?

We have the earthquake probability page using aftershocks (http://pasadena.wr.usgs.gov/step? Otherwise that is it. One reason is that the whole earthquake monitoring and research program in the United States is funded at ~$50 million for the whole country - the same value (not adjusted for inflation) that we had in 1990. #  

（原文完）

关于地壳运动造成地震的地壳漂移说：

http://www.geocities.com/zhiyanle2/pic08/usgs_india-asia_tectmap.jpg

http://photo13.hexun.com/p/2008/0527/204801/b_DA13062294232E43505CB5DCFA7656F6.jpg   

来源：光明网>>zhiyan_blog

来源网址：http://blog.gmw.cn/u/36000/archives/2008/17351.html

2008.6.29.下载