Wednesday 7 January 2015

Detecting Climate Change and attributing extremes

Coinciding with the celebration in Lima of the XX Conference on Climate Change (COP 20), the WMO published a press release about the behavior of the first ten months of 2014. Among other interesting information, 2014 was on it's track to be the hottest (or almost) year ever recorded

Evolution of the mean annual temperature anomaly from the average of the period 1961-1990 (WMO). The presence of a El Niño or La Niña phenomenon is colored (El Niño favors a warmer year). 2014 has been a neutral year. Source: NOAA
The press release also stated that this is largely due to the high temperatures of the ocean surface layers:

Evolution of the heat content in the oceans in the layer 0-700 meters Source: NOAA

Also, the Met. Office published a report which basically comes to the same conclusions also presenting some very interesting graphics:

Ranking of the most warm and cold years. Source: Met. Office



All these are the results of a huge research effort for what has been called the "detection of climate change": determine whether the series of observations of different atmospheric variables, particularly temperature and precipitation show a trend in one way or in another from a statistical point of view. The results are here and they are very significant.

However, what concerns to the general public and to the media is to determine if this warming is responsible for adverse weather events and if we can blame anthropogenic greenhouse gas emissions.
Thus, whenever a significant weather event occurs, and specially if you have had an adverse effect on the population, the question most repeated by journalists is if this event is related to anthropogenic climate change. Meteorologists answers have always been: An isolated phenomenon can never be scientifically attributed to climate change because it can be within the natural variability of the atmosphere. And if that had anything to do, it could not be known in which exact proportion. They normally add: it would be the continued trend, increase or decrease of this type of phenomena which  could be related to climate change.
The journalist agrees that this is the correct scientific answer but ... that is not a newspaper headline as it doesn't point climate change as responsible of the specific adverse meteorological situation that affected us. And this is what the general public is interested in. But things are now changing.

Aware of both the importance of the matter (its value in communication and public awareness) and its own scientific interest, the scientific community intensified its efforts from  2009-2010 to try to answer the million dollar question. This is how many "attribution" projects started seeking to establish wether a particular phenomenon can be attributed to natural variability or not.

A first approach is purely statistical and is based on the study of series of precipitation and temperatures. It looks to variability over years and looks to what extent the episode of study is statistically consistent with the variability or is way out of the natural variability. If this were so, it would follow that possibly some new forcing has contributed to its generation. Naturally, for this type of work, long statistical series are needed and the conclusion that can be obtained about the "weirdness" of the episode does not conclude that its cause is the anthropogenic action.

A much more "direct" method is based on the use of climate models, in which the type of situation or episode is simulated either leaving the atmospheric dynamics "on their own" without any external conditioning or applying a "forcing" through the introduction of current concentrations of greenhouse gases in the atmosphere. This way we can see if the situation can be explained as a result of this natural variability or if it could only happen when concentrations of such gases are on scene.

In this context, the American Meteorological Society (AMS) launched an interesting initiative in 2012: to publish every year a special edition of its famous Bulletin with compilation of articles about the attribution of different adverse situations occurred during the previous year in different parts of the world. On the first issue of the year 2012, some very interesting findings were collected from 2010 and 2011 such as:

- Texas Heat Wave: It was related to "La Niña". Now it is twenty times more likely during "La Niña" years than it was in the 60s.

- Very cold December 2010 in the UK: Chances of occurrence have been halved as a result of anthropogenic climate change.

- Very warm November 2011 in the UK: Its occurrence is now sixty times more likely then in the 60s.

- Severe flooding in Thailand: No demonstrable relation to global warming.

If we go to the 2013 report, we also find interesting information such as:

- Heat waves in Australia, Europe, China, Japan and Korea: The anthropogenic influence increased their likelihood of occurrence or intensity.

- Cold Spring in the UK: The anthropogenic reduces its likelihood and intensity.

- Heavy rainfall in central and southern Europe: No anthropogenic influence was found or at least remains unclear.

- Heavy snowfall in the area of the Pyrenees: Unclear anthropogenic influence.

However, the "problem" of these reports is that they publish their results several months later, once the media interest has already gone and thus the chance of a major social and political awareness has disappeared... This is why the Met Office report we already talked about is so important. They have initiated a process (Christidis et al., 2014) to quickly determine the possible allocation of a specific temperature record to anthropogenic climate change. In fact, the Met Office affirms that the current average global temperatures would be very unlikely to occur without human influence on climate.

Anyway, what's really important is that a way has been stablished to study the influence of climate change on global extreme weather events. Besides its great scientific scope, this represents an important step forward in raising awareness about a phenomenon wich is already irreversible but could still be partially mitigated.

This way, meteorologists can start answering the million dollar question in a more accurate way.

Sunday 28 December 2014

Sandy and Climate Change

Two years ago, hurricane Sandy terrified the United States. More than 110 dead in the world's first economy and in the same area that one year earlier suffered the consequences of Irene (8 deaths then). Flooding in streets, tunnels and subway, and blackouts paralyzed New York and parts of New Jersey, affecting 24 American states on that fateful October 29th. It even paralyzed Mitt Romney and Barack Obama campaigns.
Beyond the first world, Sandy devastated some parts of the Caribbean before landing on the US: Haiti (54 dead), Guatemala (52 dead and 20 missing, coinciding with an earthquake), Cuba, Dominican Republic (2 deaths), Jamaica and Bahamas (here without any dead fortunately).
The presidential campaign itself (Obama and Romney mentioned it) and the press in general (it's global warming, stupid) suggested the relationship between Sandy and global warming. But is there really a relation in this kind of events?

First of all, why was Sandy so devastating? For the simple reason that it was a huge hurricane. It started as a tropical storm originated in the warm Caribbean in front of Venezuela. As it moved northward, Sandy was fed by the unusually warm waters of the Atlantic. Gradually, it was increasing the intensity of showers and thunderstorms until Sandy ran into a very cold front coming from Canada and became a hurricane superstorm with 1900 km in diameter and winds over 120 km / h (up to 280 km / h in the storm center).
But is this normal? Well... Normally these storms dissipate on the Atlantic Ocean but in this case, a rare high pressure system over Greenland (due to an abnormal warming), lead Sandy to the east coast of the US. This way, due to the collision between a warm storm with a cold system over the US, Sandy became a very strange monster: a snowing tropical hurricane. If we add the contribution of the full moon during those days (nothing to do with werewolves), which raises tides above average, we already have our extreme scenario.

Destruction caused by Hurricane Sandy in Breezy Point, New York. Photograph: Julie Hau
Great, but ... this is related or not to global warming? For me, the answer is somewhere between "something" and "pretty much". It is very difficult to establish cause-effect relationships in meteorology. The climate is a chaotic system by definition; multivariable, nonlinear and interdependent. Applying a mechanistic reasoning "A implies B" in meteorology is not correct.
Although, basically, that would not matter if we believe what George Lakoff (the famous linguist author of "Do not think of an elephant" (what are you thinking now?)) points: Through the term "systemic causality" Lakoff explains how sometimes we put the entire guilt on elements affecting something only partially, like if they were totally responsible for its whole. For example, Lakoff cites "smoking causes lung cancer". Well, it's smoking and more things right? It doesn't matter: smoking systemically causes lung cancer. For Lakoff, the same happens with climate change and Sandy and I think it makes sense.

Being objective, in meteorology, frequency is even more important than intensity. So will there be more extreme weather events due to climate change? The answer is "YES, definitely". On a previous post, we already discussed this expected increase in extreme weather frequencies. Andrew Cuomo, the Governor of New York: "We have a 100-year flood every two years now". And studies seem to prove him right. Grinsted et al. 2012 published in Nature an article where it was shown that the number of major storms has increased in frequency since 1923; indeed, they detected that extreme events are about twice more likely during warm years than during cold ones.
Now think about this 2012 summer and the episode of abnormal Arctic Ice loss. In principle, these warmer freshwater could alter the flow of the jet stream around the northern hemisphere leading to an increase of water temperature in the North Atlantic water. Will this have any relation with Irene, Sandy and the massive winter storms that have been hitting the US Northeast in recent years?

There is also an expected sea level rise due to global warming. Lin et al. 2012 indicated that the combined effects of this new climate and an increase of 1 meter in the sea levels could mean that 100 years ago extreme events occur between every 3 and 20 years ... in New York !!! In other words, we should expect a "Sandy" every 3-20 years and that doesn't mean that the next will come in 19 years...
Besides, the American northwest coast is full of small bays, inlets, lowlands etc which means that it is very vulnerable.


Can you imagine the economic and environmental costs of protecting all this area with dikes? They did something similar in Louisiana to try to contain the Bp's oil spill in 2009. Setting up a (terrible) sand dike of around 70 kilometers long costed $360 million. The cost is low to save human lives but ... can we not spend it on something else? Because if you don't care about weather, let's talk about money.


Wednesday 17 December 2014

Lightning in a changing climate

When speaking of the possible consequences of climate change,  the increase of the sea level, the intensity and frequency increase of hurricanes, ocean acidification and many other effects (which are enough to worry about) come to our mind. 
However, according to Romps et al, now we have to also worry about an increase of lightnings. On a paper titled "Projected increase in lightning strikes in the United States due to global warming" and published on the November issue of Science, they concluded that global warming will also result in an exponential increase in the production of lightnings during storms, one of the main causes of wildfires. 
These findings are not new: Price et al. 1994, also studied how global warming would increase the number of lightning-caused wildfires finding "a possible increase in lightning-caused fires in the US of 44% [...] with a 78% increase in the area burned by these fires".


Image: wleog.org
According to Romps et al. findings, climate change is causing a warmer atmosphere which holds more moisture, and this is one of the key ingredients to trigger lightnings.
As they explain in their article in Science, current lightning predictions are based essentially on the thickness of the storm clouds (higher clouds generate more lightnings). However, these models do not account other physical factors as the amount of moisture in the air or the potential of a storm cloud to generate air updraft. Therefore, they incorporated these factors in their prediction model.

What does this new study predict?
According to the developed equations, lightnings would increase exponentially, this means that in the next century it is estimated that the frequency of lightning increases by a 50% in the US territory. (combined with the expected longer and more frequent heat waves (Meehl and Tebaldi, 2004), this becomes a ticking time bomb).
Romps et al, explain as well that when lightning occurs a large amounts of ozone (which in excess is a powerful greenhouse gas) is released and this fact becomes a negative feedback for the climate system.
However, if we want to look in the bright side, lightnings also produce nitrogen oxides which indirectly reduce levels of methane, another important greenhouse gas.

Together with the increasing seismic activity, this is one of the oddest climate change effects I have seen. What do you think?
Btw, here you have Bill McGuire's (UCL geologist) book about how climate change could trigger earthquakes, tsunamis and volcanoes.


Tuesday 16 December 2014

WELL DONE PHILIPPINES!



Hi there! Sorry for my disappearance, the end of the term was quite stressful! I'm finally back and I will catch up with the latest things that have been going on during the last two weeks.
I will start with very good news (which is kind of unusual in my blog): Filipinos are a great example of how tragedy can deliver great lessons for the future if we are ready to learn from it.


Satellite view of typhoon Hagupit. Image: NOAA


Typhoon Hagupit (known as Ruby in the Philippines), made landfall last December 6th with winds up to 200km/h. 
The balance of this storm was 19 deaths in Philippines. One year after the powerful typhoon Haiyan, which left more than 7,000 dead in November last year, Filipinos have learned many lessons.

It is true that Hagupit was a slightly less intense storm, and, unlike Haiyan, it did not produce the storm surge of 5.2 meters high which devastated the city of Tacloban, but I think thatit is fair to say that this time the government was much better prepared. Early warning systems and preventive evacuations served to greatly minimize human losses.

At least 1.7 million people managed to evacuate their homes and find refuge in over 5000 evacuation centers. Shelter, food, water, sanitation and hygiene, health and logistics were among the main priorities.

According to the United Nations, the typhoon was estimated to affect 2.9 million people in nine different regions of the Philippines. The official death toll rose up to 19 (including at least seven children) and 916 injured.

As floods recede, more than half of the evacuees returned home in the last days. To date, 788,000 people remain in 3,0000 evacuation centers. Packages of food, water and hygiene kits were distributed to more than 95,000 victims. Meanwhile, government officers are planning to meet with NGOs to coordinate reconstruction efforts and humanitarian aid.

One of the evacuation centers. Image: Reuters

Philippine authorities, with the support of NGOs, continue to assess the damage and needs while focusing efforts in the provinces of Northern Samar, Eastern Samar, and Samar. They have also announced their goal to provide shelter assistance to affected communities until January 2015.

All national roads are already operational, but six bridges remain closed in the affected regions. Power outages continue in 12 provinces, while telecommunications coverage was restored to more than 74%.

The UN Office for Disaster Risk Reduction (UNISDR) congratulated the Philippines and said that this reaction demonstrates the leadership role that Asia plays in reducing mortality and economic losses from natural disasters.

The head of UNISDR Margaretha Wahlström said that the Philippine government has made an excellent job putting into practice the lessons learned from Typhoon Haiyan. 
She added: “We have been telling this story since the Indian Ocean tsunami ten years ago that nations and communities have the power to reduce their losses if they are well-organized, understand the nature of risk and develop the capacity to deliver early warnings and evacuate groups at risk ahead of the disaster event.” Now it is a demonstrated fact.

WELL DONE PHILIPPINES!



Btw, here you can listen to Mary Ann Lucille Sering (commissioner of the Philippines’ Climate Change Commission and lead climate official for the Philippines at the Lima conference) talk about adaptation and mitigation in Philippines


Monday 1 December 2014

The huge economical cost of coastal flooding in 2050

The effects of climate change are enough now-a-days to start assess the economical cost it will cause to humankind in the future.
Climate change, together with population growth in some areas of the world and economical growth, could lead to a big increase of flooding risk in big coastal cities (up to nine times today's risk by 2050).

A paper published in August 2013 in Nature (Hallegate et al. 2013) titled "Future flood losses in major coastal cities" has estimated this cost and is part of a bigger work carried out by the OECD (Organisation for Economic Co-operation and Development) in order to assess flooding risk (particularly due to storms) as a consequence of climate change and urban development from a social and economic point of view.

The new study has been partially based on previous OECD works, which classified the world's port cities on a scale based on their present and future vulnerability to storms taking into account the number of inhabitants and the value of the goods which could be affected by flooding.
Hallegate et al, have calculated the present and future losses due to flooding in 136 of the biggest coastal cities in the world taking into account the existing flood defenses (the city of Amsterdam is an interesting example with and extremely high exposure but small economic average annual losses due to it's extraordinary defenses).
The average losses in 2005 were quantified to be around 6.000 million dollars and could increase up to the mind-bending number of 52.000 million per year by 2050.
According to this metric, most of the more vulnerable cities in the world  are part of industrialized countries such as: Guangzhou and Shenzhen in China, Miami, Tampa-St. Petersburg, Boston, New York and New Orleans in the US, Bombay in India, Nagoya and Osaka-Kobe in Japan and Vancouver in Canada.
Due to their high wealth and low protection level, three US cities (Miami, New York and New Orleans) explain 31% of total economical losses in the 136 cities. Adding the Chinese city of Guangzhou, the four cities explain 43%.

The economic cost in dollars is a way of assessing the risk but a more complete way of assessing it is to see these economical losses as a percentage of the city's total wealth.
According to this criteria, Guangzhou (China), Guayaquil (Ecuador), Ho Chi Minh (Vietnam) and Abidjan (Ivory Coast) are within the most vulnerable cities in the world.





To estimate future climate change impact, a sea level rise between 0.2m and 0.4m by 2050 is assumed.
In addition to this, about a quarter of these 136 cities are located in river deltas, and therefore more exposed to river floods, as well as decreases of soil on land or promotion of coastal seabed, due to the local subsidence, particularly in cases where overexploitation of groundwater accelerates natural processes.

In my opinion, one of the most important findings of this study is that even a "moderate" sea level rise would cause substantial increases in losses if adaptation measures are not taken because the defenses against flooding are designed to past conditions.
Doing nothing is not an option, as it could cause losses exceeding one trillion or more per year, according to the study. 
Therefore, coastal cities will have to improve its flood action plan, including the construction of good defenses, at an estimated cost of around 50,000 million dollars per year to 136 cities.
The report makes the point that significant increases in the flooding risk in cities which are not vulnerable today may occur.
The five cities with the highest estimated risk increase are Alexandria in Egypt, Barranquilla in Colombia, Naples in Italy, Sapporo in Japan, and Santo Domingo in the Dominican Republic. 





Wednesday 26 November 2014

Fake and Overused Weather Photos

Just came across THIS on Facebook and couldn't resist the temptation to share it with you.
You have my word: I will never ever use this pictures!!!

The Ridiculously Resilient Ridge

Welcome again to my blog,

as last week's post was talking about the anomalous cold weather in eastern US, I thought that talking about California's drought would make sense. As I said on my last post, US weather is more bipolar than ever!

California is the first US economy and half of the country's fruit is produced in it's central valley. This study by the University of California, Davis (Sacramento) assesses the massive economic impact of the drought (around 2.200 million dollars)

The official state of emergency was declared in January 2014, in order to address the problem and try to find ways to conserve and use water more efficiently. Serious problems related to water use had already begun: farms can not provide water for livestock, thousands of workers in agriculture are loosing their jobs and severe wildfires are taking place, just to name a few.

The gravity of California's drought is so great that it is already changing the geography of the state: one of it's most important rivers is not reaching the Pacific Ocean anymore.

Just look at a couple of pictures: 



Satellite images showing the snow cover over the Rocky Mountains in January 2013 vs January 2014
California's Governor Jerry Brown encourages individuals and companies to put to work strategies to try to save water and reduce consumption to 20% or more. While many of these initial measures are directed to residents, the agricultural sector accounts for up to 80% of total water consumption in California, according to MacDonald, 2010. The restrictions on water use are usually applied voluntarily, but in certain areas they have become mandatory.

In early February 2014, a heavy rainfall fell over Sonoma. 2013 was the driest year on record and because of this, soil was very poor in moisture.
When soils are lacking moisture so severely, they harden and crack, becoming less permeable to water and this results in heavy rainfall not being absorbed, causing flash floods.
As the drought deepens, flash floods may become more frequent, bringing new challenges to farmers in the area (including the cannabis industry).

As if it were not enough with these, Amos et al., 2014 link drought with higher earthquake probability.

This is the third consecutive year of drought in California. Now comes the one thousand dollar question... Can we attribute this to climate change?

Let's first check the science behind the event.


During the last three years, a vast anticyclone hast remained over the northeastern Pacific Ocean (the so called "Ridiculously Resilient Ridge" I love the nickname!) preventing the winter storms from reaching California and as a result, bringing extreme temperatures and massive drought to this area.
This is also a consequence of the Rossby waves we talked about in the last post.
In this report, the Stanford researchers Noah Diffenbaugh and David Swain (the video stars) analyze the impact of climate change in this particular event trough a combination of GCMs (Global Circulation Models) and statistical methods which have allowed them to determine whether the formation and resilience of such an anticyclone had been possible in the absence of the effects of Climate Change.
They concluded that the atmospheric conditions driving to this situation in California are very likely related to human-caused Climate Change. In other words, human-caused climate change has dramatically increased the likelihood of extremely high atmospheric pressure over the North Pacific Ocean which has lead to the drought in California.