nasdaq
论坛版主-副热带高压
论坛版主-副热带高压
  • 注册日期2011-08-26
  • 最后登录2017-05-22
  • 粉丝131
  • 关注9
  • 发帖数781
  • 来自
阅读:2436回复:0

[气象讯息]一位预报员对东太平洋极端厄尔尼诺的看法

楼主#
更多 发布于:2015-08-30 21:15
原文网址:https://www.climate.gov/news-features/blogs/enso/one-forecaster%E2%80%99s-view-extreme-el-ni%C3%B1o-eastern-pacific

One forecaster’s view on extreme El Nino in the eastern Pacific
一位预报员对东太平洋极端厄尔尼诺的看法

This is a guest post by Ken Takahashi, who is a research scientist at the Instituto Geofísico del Perú (IGP) and currently leads the national scientific committee ENFEN, which issues the official El Nino forecasts in Peru. This post does not necessarily reflect the views of IGP, ENFEN or NOAA.
这是一篇来自Ken Takahashi的客座文章,他是秘鲁地球物理研究所(IGP)的一名研究科学家,目前领导着国家科学委员会ENFEN,也是秘鲁发布厄尔尼诺预报的官方机构。这篇文章并不代表IGP,ENFEN或NOAA的观点。

El Nino was first identified by fisherman in the late 19th century off the coasts of Peru and Ecuador (Carranza, 1892; Carrillo, 1893). Unusually high Pacific Ocean temperatures depressed the region’s fisheries, and intense rainfall led to coastal flooding.  The most extreme El Nino events, in terms of the surface warming in the eastern and central Pacific, occurred during 1982-1983 and 1997-1998. During these two events, Piura, a city in the coastal desert in northern Peru, experienced annual rainfall amounts equivalent to the other 40 rainiest years combined! The economic loss due to extreme weather in Peru during those events is estimated as 7% and 4.5% of its GDP, respectively (CAF, 2000).
厄尔尼诺最先是在19世纪末由秘鲁和厄瓜多尔沿岸的渔民发现。太平洋反常高的海温打击了当地的渔业,猛烈的降雨引发沿海洪水。从中东太平洋表层增暖的角度来看,最极端的两次厄尔尼诺发生在1982-1983年和1997-1998年。在这两次厄尔尼诺事件中,皮乌拉,秘鲁北部沿岸沙漠中的一座城市,所记录得的年降雨量相当于其余40个雨量最多的年份之和!这两次事件带来的极端天气给秘鲁带来的经济损失分别是GDP的7%和4.5%。

The desire to help society prepare for those kinds of disruptions has led to great scientific advances in understanding El Nino. Still, one of the most frustrating things about El Nino for forecasters is why it doesn’t have the same impacts in the same places every time. In the past decade, the scientific community began to focus research on the diversity or flavors of El Nino and La Nina (the cold phase) as a possible explanation for the variability of impacts.
为帮助社会为此类极端天气做好准备,厄尔尼诺方面的研究有了巨大的进步。然而,最为预报员所困惑的事实之一是,为何每次厄尔尼诺在相同的地方却有着不同的影响呢?在过去十年中,学界开始关注厄尔尼诺和拉尼娜(对应的冷相位)的差异或特点,以试图解释不同事件影响的差异。

图片:1.ENSO_flavors_large.jpg


【图1】
Pattern of sea surface temperature deviation from average (°C) associated with a unit value of the C index (top) and the E index (bottom), based on Takahashi et al., 2011. The Nino 3.4 and 1+2 regions are indicated as dashed boxes. Most El Nino events can be described as a combination of these two patterns. Image from Ken Takahashi.
和中太指数(上)及东太指数(下)相关联的海表温度距平模式。Nino3.4区和1+2区由框线标出。大多数厄尔尼诺事件可以由这两种模式的组合来描述。图片来自Ken Takahashi。


In particular, they've focused on where along the equator the surface warming is largest, which does affect how El Nino and La Nina impact the global climate (Larkin & Harrison, 2005). Especially in Peru, El Nino can lead to very different rainfall impacts depending on whether the warming occurs in the eastern (wetter) or central Pacific (drier) (Lavado-Casimiro & Espinoza, 2014).
学者尤其侧重关注沿赤道的表层海水哪一部分增暖最明显,这影响了厄尔尼诺和拉尼娜对全球气候产生影响的方式。尤其在秘鲁,厄尔尼诺能产生差异很大的降雨影响,这取决于增暖是发生在东太平洋(更湿)还是中太平洋(更干)。

There are many different ways of classifying El Nino, but it is most common to measure it using sea surface temperature (SST) anomalies (departures from average conditions).  In order to classify the different types of El Nino, however, we need at least two indices or time series (Trenberth and Stepaniak, 2001). Some colleagues and I introduced an E index and a C index (data here), which isolate the SST changes in the Eastern and Central Pacific, respectively, that are unique to each region (Takahashi et al, 2011).
对厄尔尼诺进行分类有很多种方法,最常用的测量标准是表层海温距平。为了分类不同种类的厄尔尼诺,我们需要至少两个指数或时间序列。我和我的同事引进了东太指数和中太指数,将东太和中太的海温变化分离开来,互不相同。

How different are the extreme El Nino events from the regular ones?
极端厄尔尼诺和普通厄尔尼诺有何区别?

Usually the SST warming in the central and eastern Pacific overlap, or correlate, during El Nino. But during the two “extreme” El Nino events (1997-98 and 1982-83), the warming in the east, near the coast of South America was much stronger than the warming farther west in the central Pacific (as can be seen in the left panel below).
在厄尔尼诺中,中太和东太的海温增暖通常是重叠的,或者相关。但是在两次极端厄尔尼诺事件中(1997-98和1982-83),东太靠近南美海岸部分的增暖远比中太部分要强(可见下图)


图片:2.EasternCentral_scatterplot_large.png


【图2】
December-February average eastern and central Pacific sea surface temperature deviations from average: (left) Nino 1+2 (east Pacific; on x-axis) and Nino 3.4 (central Pacific; on y-axis); and (right): E (east Pacific; x-axis) and C (central Pacific; y-axis) departures from average. The year corresponding to December is indicated. Extraordinary El Nino events are indicated in red, while other eastern Pacific and central Pacific El Nino events are in green and blue, respectively. Gray indicates non-El Nino years. In both graphs, the dotted lines are an attempt to summarize the relationships shown by the dots, and the abrupt change of the slope of the dotted line highlights the uniquely different behavior shown by the 1982 and 1997 cases, and to a much smaller extent the 1972 case.
十二月至二月的中东太平洋平均海温距平:(左图)Nino 1+2(东太平洋;x轴)和Nino 3.4(中太平洋;y轴);(右图)东太指数(x轴)和中太指数(y轴)距平。对应年份已标出。极端厄尔尼诺事件用红色标出,其他东太厄尔尼诺和中太厄尔尼诺分别用绿色和蓝色标出。灰色意为非厄尔尼诺年。在两幅图中,虚线用来总结点之间的关系,其突兀的斜率变化展示着1982年、1997年还有稍不明显的1972年异常的表现。


In fact, the values of the central Pacific Nino 3.4 index were only slightly greater than those of the 1972-1973 event, but the values were around 3 times greater in the eastern Pacific Nino 1+2 region. These geographic differences are also clearly depicted using the E and C indices (right panel), with very high E values during the two extreme El Nino events. This difference in central versus eastern Pacific warming during extreme events compared to regular ones is also evident in monthly C and E index values (see graphs below).
事实上,1982年和1997年的中太平洋Nino3.4指数仅比1972-1973厄尔尼诺事件稍高一点,然而东太平洋Nino1+2指数确实其将近三倍多。这些区域差异在中太指数和东太指数上被清晰的描述了出来(右图),两次极端厄尔尼诺事件的东太指数非常高。极端事件里中东太增暖的差异在中太和东太月指数上亦显著异于普通事件。(见下图)

图片:3.EasternCentral_timeseries_large.png


【图3】
(top) Central Pacific (C) and (bottom) eastern Pacific (E) monthly SST indices during selected El Nino events and the current year. The estimated values for August 1-19, 2015, are indicated with an open circle. Graph by Ken Takahashi.
一些厄尔尼诺年份和今年的中太平洋(上)和东太平洋(下)月海表温度指数。空心圆为2015年8月1日至19日的估计值。图片来自Ken Takahashi。


We found that this is because, once the normally cooler eastern Pacific warms enough for heavy precipitating storms, El Nino shifts to a faster gear: the Walker circulation shifts dramatically towards the eastern Pacific and the processes that lead to El Nino growth strengthen threefold (Takahashi & Dewitte, 2015).
这是因为,东太平洋通常更冷,一旦其增暖得足够形成强降水风暴,厄尔尼诺将会加速:沃克环流大幅移向东太,这一过程导致厄尔尼诺发展速度提高三倍。

Predicting extreme El Nino this year
预测今年的极端厄尔尼诺

If the physics of extreme El Nino events are different, then they should sometimes be analyzed separately from the rest; this also makes sense considering their large societal importance. Of great urgency this year: Are our scientific understanding and models good enough for the prediction of an extreme El Nino?
如果极端厄尔尼诺的机制是不同的,那么在分析时它应该与其他厄尔尼诺分开来。考虑到其巨大社会影响,这么做也有一定意义。对于今年最为紧要的就是:我们的科学知识和模型是否已经足够胜任预报一次极端厄尔尼诺?

Although climate models provide objective predictions, models are far from perfect. They have common errors (particularly large in the eastern Pacific) and misrepresentation of slower changes in SST (decadal or 10-year timescales) or SST trends (2). By considering a collection of different models, or a multi-model ensemble (3), we hope that the errors cancel out among the different models. However, there are errors common to all models, such as the warm and rainy tendency in the cold and dry southeastern Pacific.
尽管气候数值提供了客观预报,数值离完美还有很远的距离。它们有着共同的误差(尤其在东太平洋),无法反映海表温度的缓慢变化(十年的时间尺度上)或变化趋势(脚注2)。考虑到我们有一系列不同的数值,或者多数值系集(脚注3),我们期望能通过不同数值来减少误差。然而,所有数值都犯一些相同的错误,例如会将干冷的东南太平洋预报的更温暖潮湿。

And we know that the models have a harder time making accurate predictions in the eastern Pacific. In particular, the models do not predict large enough SST anomalies in the far eastern Pacific during the extreme El Nino events (Takahashi et al, 2014). Even so, many models are predicting a strong El Nino in the central and eastern Pacific this year, similar to (or stronger than) 1972-1973, 1982-1983, and 1997-1998.
我们已经知道,数值在东太平洋的预测更不准确,特别是数值难以预报极端厄尔尼诺事件的远东太平洋海温距平。尽管如此,许多数值今年都预报出一次强厄尔尼诺,类似于(或强于)1972-1973,1982-1983以及1997-1998.

In addition to models, forecasters have other tools available, such as observational predictors and ideas based on physical common sense. The limitation in this case is the small number of events, with only two well-observed extremes, coupled with the fact that one El Nino is never a perfect mirror image of another El Nino, not even the extremes.
除了数值,预报员还有其他可用的工具,例如可观测到的预测指标以及基于物理常识的想法。这些工具的局限性在于过去可参照的厄尔尼诺事件较为稀少,极端厄尔尼诺只有两次被良好的记录下来,何况世上没有两次完全相同的厄尔尼诺,更不要提极端事件了。

This year the ocean has accumulated a substantial amount of heat, a necessary condition for El Nino, but this does not tell us whether El Nino will be extreme or not in the eastern Pacific (Takahashi & Dewitte, 2015). Again, an extreme El Nino is a very different beast from the others in terms of impacts on weather and wildlife in the coastal regions of northern Peru and Ecuador, so El Nino strength is not just a detail.
今年海洋已经积蓄了大量的热量,这是厄尔尼诺的必要条件,但这并不能告诉我们东太平洋会否出现极端厄尔尼诺。前面已经说过,从其对秘鲁北部和厄瓜多尔沿岸地区的天气和生物的影响来看,极端厄尔尼诺是非常不同的怪兽,所以厄尔尼诺的强度并不是一个简单的指标。

One feature we found potentially useful is that if the trade (easterly) winds in the central Pacific become very weak around August, this allows the eastern Pacific to warm up a few months later, possibly enough to trigger strongly enhanced precipitation that could help El Nino become extreme (Takahashi & Dewitte, 2015). This did not happen in 1972, which is perhaps why that El Nino did not become as extreme.
一个很可能有用的特征是,如果八月前后的中太平洋信风(东风)变得非常弱,那么东太平洋将在接下来的几个月增暖,足够引发更强的降水引领厄尔尼诺达到极端强度。这在1972年并未发生,也可能是当年厄尔尼诺没有达到极端强度的缘故。

图片:4.SSTA_windstress_maps_large.png


【图4】
Difference from average sea surface temperature (colors) and difference from average of surface wind stress (arrows showing direction and strength by the length of the arrow line) in August 1982 (top) and January 1983 (bottom). The red box outlines the averaging region for the wind stress predictor for judging the probability of occurrence of an extreme condition in the Eastern Pacific 5 months later in January. Images adapted from Ken Takahashi.
1982年8月(上)和1983年1月(下)的海温距平(填色)和表面风压距平(箭头表示方向,长度表示强弱)。红框标出了风压预测指标区域,用于判断5个月后东太平洋的极端情况的可能性。图片改编自Ken Takahashi。


This year we are putting this tool to the test. So far, the trade winds in August have not weakened as much as in 1997 but more than in 1982, indicating the probability of an extreme El Nino in 2015-2016.  However, the eastern Pacific (E index) has been tracking the substantially weaker 1972 event and it would have to surge upwards, as in 1982, to become extreme (Fig. 3b). A quite different outcome could be that E keeps following 1972, remaining below the extreme threshold, while the central Pacific continues to warm into perhaps a larger version of the 2009-2010 El Nino (see bottom graph of Figure 3).
今年我们将对这个工具进行一次检验。当前,八月信风大幅减弱,虽不及1997年的程度但超过了1982年,预示着2015-2016年一次极端厄尔尼诺的可能性。然而,东太平洋(东太指数)需要再增暖一些才能像1982年一样达到极端水平,而当前指数正沿着弱很多的1972年的指数趋势走(图3下)。如果指数沿1972年水平继续发展下去,东太平洋将低于极端程度门槛,而中太平洋可能会进一步增暖成为2009-2010厄尔尼诺的增强版,其结果将会大不同(见图3下图)。


图片:5.windstress_scatterplot_large.png


【图5】
Predicted departure from average westerly wind stress (see footnote 1) in August (x-axis) vs. the eastern Pacific warming (E) in the following January (y-axis). Observations are in red, while the CM2.1 model ensemble forecasts (repeated model runs with different starting conditions) are grey, with their 10%, 50%, and 90% percentiles shown by the black sloping curves to summarize the positions of most of the gray dots. Adapted from Takahashi & Dewitte (2015).
八月西风风压(见脚注1)距平和翌年一月东太平洋增暖的分布关系。观测值为红点,CM2.1模式集合预测(数值以不同的开始条件重复运行)为灰点,以及总结灰点位置的10、50和90百分位斜线。改编自2015年Takahashi和Dewitte的论文。


As you can see, the chance of an extreme El Nino in the eastern Pacific is not straightforward to assess (5). Several factors will affect such a estimation. This year’s El Nino is already different from anything seen before. Furthermore, the rules of how the climate system works do not stay the same throughout time (e.g. climate change may affect El Nino), so statistical relationships found in a previous period might not be valid anymore.
可以发现的是,东太平洋极端厄尔尼诺的几率无法直接得出(脚注5)。好几个因素将会影响到我们的预计。今年的厄尔尼诺不同于过去任何一年。进一步地说,气候系统运作方式会随着时间而改变(例如气候变化将会影响厄尔尼诺),因此靠上一个周期得到的统计学关系可能不再有任何意义。

Also, it is possible that random factors outside of the El Nino system could go against El Nino to keep it below the extreme threshold. Although several climate models are predicting a very strong El Nino, due to their common errors, we cannot fully trust them. Perhaps the only reliable rule is that El Nino can surprise us, and this year could be yet another example.
而且,一些厄尔尼诺系统外的随机因素也可能阻止厄尔尼诺达到极端强度。尽管好几个气候数值预报有非常强的厄尔尼诺,由于这些数值的通病,我们不能完全信任它们。也许唯一可信的是厄尔尼诺可能会在我们意料之外,今年可能成为完全另一个模式。

Anthony Barnston, lead reviewer

Footnotes
脚注

(1) The wind stress is based on the wind speed squared. Here, we are talking about the departure from average of the westerly wind stress. When the trades winds (winds from the east) become weaker, as they do during an El Nino event, the departure from average of the westerly wind becomes positive (because weaker trade winds mean stronger westerly winds, even if the actual wind is still from the east, but less strong than average). Then we square that departure from average. For example, if the westerly wind is usually -9 miles per hour, and now it is only -2 miles per hour, then the departure from average of the westerly wind is +7 miles per hour. And the departure from average of the westerly wind stress is the square of 7 miles per hour, which is 47 miles per hour.
(1)风压基于风速的平方。在这里我们讨论的是西风风压的距平。当信风(东风)变得微弱,就像它们在厄尔尼诺事件中表现的那样,西风距平将为正值(因为弱信风意味着强西风,尽管实际风向仍来自东方,但不如平均值强)。然后我们将距平值平方。举个例子,如果西风风速通常为-9mph,现在有-2mph,那么西风距平即为+7mph。西风风压距平是+7mph的平方,西风风压距平则为7的平方,即47。(译注:原文如此)

(2) Changes in the entire North Pacific plus tropical Pacific on an approximately 10-year time scale, known as decadal variability, can change the backdrop behind El Nino and La Nina and encourage one of these at the expense of the other. As it turns out, much of the advances in El Nino science took place during a warm Pacific decadal phase, but we have been in a cold phase since approximately the year 1999 (although there are hints that we might be switching back to warm; we need to wait another year or two to make sure). Which decadal phase we are in can subtly, but noticeably, affect the strength of El Nino or La Nina, and our prediction models may not adequately take this decadal variability into account.
(2)整个北太平洋和赤道太平洋区域的大约十年时间尺度上的海温变化,通常被称为十年际变化,能够改变厄尔尼诺和拉尼娜的海温背景,倾向于发展其中一个而抑制另一个。厄尔尼诺相关科学的进步大多发生在太平洋暖相位中,但我们在大约1999年起进入了冷相位(尽管目前有可能摆回暖相位的迹象,但是我们还需要一到两年来确定)。我们处于哪一个十年相位将温和地影响到厄尔尼诺或拉尼娜的强度,而我们的预测数值并未足够考虑到这些十年际变化。

(3) A multi-model ensemble refers to the use of more than one model to make a forecast of deviations from average of climate or of sea surface temperature. Because each single model has its own biases or peculiarities, averaging the forecasts of several models tends to cancel these out and deliver a forecast having fewer specific biases. If several models have common biases, however, using more than one model does not help as much.
(3)多数值集合指用不止一个数值来做出气候或海温距平预报。由于每一个数值有其偏差或缺陷,平均几个数值的预报能在一定程度上将误差消除掉,得出有更少偏差的预报。如果几个数值都有共同的偏差,那么多个数值集合无法发挥足够的作用。

(4) The 1953-54 El Nino (leftmost green dot in both panels) had its largest warming in the eastern Pacific around mid-1953, but in DJF the eastern Pacific became relatively cool.
(4)1953-54厄尔尼诺(两幅图中最左侧的绿点)在1953年年中东太平洋增暖最大,然而在12月-2月东太平洋变得相对更冷。

(5) Despite the large uncertainties in the eastern Pacific, Peru’s ENFEN will produce an estimate of the probabilities of the various strengths of El Nino, including the extreme type, later this week.
(5)尽管东太平洋有非常大的不确定性,秘鲁的ENFEN机构将会预估不同强度的厄尔尼诺的可能性,包括极端强度,预报将在本周末推出。

References

CAF, 2000: El Fenómeno El Nino 1997-1998. Memoria, Retos y Soluciones, V. Perú. Perú. Corporación Andina de Fomento, http://scioteca.caf.com/handle/123456789/676

Capotondi A, Wittenberg AT, Newman M, Lorenzo ED, Yu JY, Braconnot P, Cole P, Dewitte B, Giese B, Guilyardi E, Jin FF, Karnauskas K, Kirtman B, Lee T, Schneider N, Xue Y, Yeh SW, 2015: Understanding ENSO Diversity. Bull. Amer. Meteor. Soc., 96, 921–938, doi:10.1175/BAMS-D-13-00117.1

Carranza, 1892: Contra-corriente maritima, observada en Paita y Pacasmayo. Boletín de la Sociedad Geográfica de Lima,  1, 344-345.

Carrillo, C.N., 1893: Hidrografía oceánica. Boletín de la Sociedad Geográfica de Lima, 2, 72-110.

Larkin NK, Harrison DE, 2005: Global seasonal temperature and precipitation anomalies during El Nino autumn and winter. Geophys. Res. Lett., doi: 10.1029/2005GL022860

Lavado-Casimiro, W and Espinoza J.C., 2014: Impact of El Nino and La Nina events on rainfall in Peru (1965-2007). Revista Brasileira de Meteorologia, doi:10.1590/S0102-77862014000200003

Takahashi K, Montecinos A, Goubanova K, Dewitte B, 2011: ENSO regimes: Reinterpreting the canonical and Modoki El Nin?o. Geophys. Res. Lett., doi: 10.1029/2011GL047364

Takahashi, K., Dewitte, B, 2015: Strong and moderate nonlinear El Nino regimes. Climate Dynamics, doi: 10.1007/s00382-015-2665-3.

Takahashi, K., Martínez, R., Montecinos, A., Dewitte, B., Gutiérrez, D., Rodríguez-Rubio, E., (2014), White Paper #8a Regional applications of observations in the eastern Pacific: Western South America, Report of the Tropical Pacific Observing System 2020 Workshop (TPOS 2020), La Jolla, United States, 27-30 January 2014, GCOS report 184, vol. 2, 171-205. http://www.wmo.int/pages/prog/gcos/Publications/gcos-184_II.pdf

Trenberth, K. E., Stepaniak, D. P., 2001: Indices of El Nino evolution. J. Climate, doi:10.1175/1520-0442(2001)014<1697:LIOENO>2.0.CO;2
[nasdaq于2015-08-30 21:24编辑了帖子]
4条评分, 好评度 +5 金钱 +2 威望 +5 贡献值 +3
  • 风王2012
    好评度 5
    优秀翻译
    2015-08-31 11:25
  • 小范老师
    金钱 +2
    优秀帖
    2015-08-30 21:48
  • wangruizhi
    贡献值 +3
    克神出手【拜
    2015-08-30 21:21
  • wangruizhi
    威望 +5
    克神出手【拜
    2015-08-30 21:21
喜欢4 评分4
游客

返回顶部