Monday, May 28, 2018

Spurious Correlations in Climate Science

Naive statistics underlie many causal claims in climate "science"

You know who Charles Darwin is of course but you may not have heard of his mad cousin Francis Galton who did the math for Darwin’s theory of evolution. Two of the many procedures Sir Galton came up with to help him make sense of the data are still used today and are possibly the two most widely used tools in all of statistics. They are ordinary least squares (OLS) linear regression and OLS correlation. [Soon after these amazing mathematical innovations, Sir Galton retired from the evolution business and devoted the rest of his life to making the perfect cup of tea.]

Both of these statistics are measures of a linear relationship between two variables X and Y. Linear regression coefficient B of Y against X is a measure of how much Y changes on average for a unit change in X and the linear correlation R is a measure of how close the observed changes are to the average. The regression and correlation metrics are demonstrated below with data generated by Monte Carlo simulation used to control the degree of correlation.

In the HIGH (R=0.94) and VERY HIGH (R=0.98) correlation charts, linear regression tells us that on average, a unit change in X causes Y to change by about B=5 and this assessment is very consistent. The consistency in this case derives from a low variance of the regression coefficient implied by high correlation. The strong correlation also implies that the observed changes in Y for a unit increases in X is close the the average value of B=5 over the full span of the data and for any selected sub-span of the time series.

In the LOW (R=0.36) and MID (R=0.7) correlation charts, the regression coefficients are correspondingly less precise varying from B=1.8 to B=7.1 for LOW-R and B=3.5 to B=5.6 for MID-R in the five random estimates presented. The point here is that without a sufficient degree of correlation between the time series at the time scale of interest, though regression coefficients can be computed, the computed coefficients may have no interpretation. The weak correlations in these cases also imply that the observed changes in Y for a unit increases in X would be different in sub-spans of the time series. The so called “split-half” test, which compares the first half of the time series to the second half, may be used to examine the instability of the regression coefficient imposed by low correlation.

Correlation is a necessary but not sufficient evidence of causation. Although correlation may imply causation in controlled experiments, field data do not offer that interpretation. If Y is correlated  with X in field data, it may mean that X causes Y, or that Y causes X, or that a third variable Z causes both X and Y, or that the correlation is a fluke of the data without a causation interpretation. However, because correlation is a necessary condition for causation, the absence of correlation serves as evidence to refute a theory of causation.

An issue specific to the analysis of time series data is that the observed correlation in the source data must be separated into the portion that derives from shared long term trends (that has no interpretation at the time scale of interest) from the responsiveness of Y to changes in X at the time scale of interest. If this separation is not made, the correlation used in the evaluation may be, and often is spurious. An example of such a spurious correlation is shown in the graphic below. It was provided by the TylerVigen collection of  spurious correlations.

As is evident, the spurious correlation derives from a shared trend. The fluctuations around the trend at an appropriate time scale (whether annual or decadal) are clearly not correlated. The separation of these effects may be carried out using detrended correlation analysis. Briefly, the trend component is removed from both time series and the residuals are tested for the responsiveness of Y for changes in X at the appropriate time scale. The procedure and its motivation are described quite well in Alex Tolley’s Lecture  available on Youtube.

The motivation and procedure for detecting and removing such spurious correlations in time series data are described in a short paper available for download at this link: Spurious Correlations in Time Series Data . The abstract of this paper follows: Unrelated time series data can show spurious correlations by virtue of a shared drift in the long term trend. The spuriousness of such correlations is demonstrated with examples. The SP500 stock market index, GDP at current prices for the USA, and the number of homicides in England and Wales in the sample period 1968 to 2002 are used for this demonstration. Detrended analysis shows the expected result that at an annual time scale the GDP and SP500 series are related and that neither of these time series is related to the homicide series. Correlations between the source data and those between cumulative values show spurious correlations of the two financial time series with the homicide series.

It is for these reasons the argument that “the theory that X causes Y is supported by the data because X shows a rising trend and at the same time we see that Y has also been going up” is specious because for the data to be declared consistent with causation theory it must be shown that Y is responsive to X at the appropriate time scale when the spurious effect of the shared trend is removed. Some examples from climate science are presented in the papers below along with the URL to their download sites.

Are fossil fuel emissions since the Industrial Revolution causing atmospheric CO2 levels to rise? Responsiveness of Atmospheric CO2 to Fossil Fuel Emissions

Can sea level rise be attenuated by reducing or eliminating fossil fuel emissions? A Test of the Anthropogenic Sea Level Rise Hypothesis

Can ocean acidification be attenuated by reducing or eliminating fossil fuel emissions? An Empirical Study of Fossil Fuel Emissions and Ocean Acidification

Is surface temperature responsive to atmospheric CO2 levels? #1 Validity and Reliability of the Charney Climate Sensitivity Function

Is surface temperature responsive to atmospheric CO2 levels? #2 Uncertainty in Empirical Climate Sensitivity Estimates 1850-2017

Is surface temperature responsive to atmospheric CO2 levels? #3 The Charney Sensitivity of Homicides to Atmospheric CO2: A Parody

A further caution needed in regression and  correlation analysis of time series data arises when the source data are preprocessed prior to analysis. In most cases, the effective sample size of the preprocessed data is less than that of the source data because preprocessing involves using data values more than once. For example taking moving averages involves multiplicity in the use of the data that reduces the effective sample size (EFFN) and the effect of that on the degrees of freedom (DF) must be taken into account when carrying out hypothesis tests. The procedures and their rationale are described in this freely downloadable paper Illusory Statistical Power in Time Series Analysis.

Failure to correct for this effect on DF may result in a false sense of statistical power and faux rejection of the null in hypothesis tests as shown in this analysis of Kerry Emmanuel’s famous paper on what he called “increasing destructiveness” of North Atlantic hurricanes: Circular Reasoning in Climate Change Research.

An extreme case of the effect of preprocessing on degrees of freedom occurs when a time series of cumulative values is derived from the source data as in the famous Matthews paper on the proportionality of warming to cumulative emissions [Matthews, H. Damon, et al. “The proportionality of global warming to cumulative carbon emissions.” Nature 459.7248 (2009): 829].

It has been shown in these downloadable papers that the time series of cumulative values has an effective sample size of EFFN=2 and therefore there are no degrees of freedom and no statistical power.

Degrees of freedom lost in moving window preprocessing Effective Sample Size of the Cumulative Values of a Time Series

Degrees of freedom lost in a time series of the cumulative values of another time series #1 Limitations of the TCRE: Transient Climate Response to Cumulative Emissions

Degrees of freedom lost in a time series of the cumulative values of another time series #2 From Equilibrium Climate Sensitivity to Carbon Climate Response

Degrees of freedom lost in a time series of the cumulative values of another time series #3 The Spuriousness of Correlations between Cumulative Values


Extinction of Puffins premature

The Telegraph sent its science reporter up to the Farne Islands in Northumbria to write up a story it dramatically headlined ‘UK puffins may go the way of the dodo with fears of extinction in 50 years.’ It claimed:

    So far the news has been bleak. The puffins arrived four weeks later than usual and initial estimates suggest the number of breeding pairs has fallen by 12 per cent.  A combination of climate change, overfishing, plastic pollution and extreme weather has left the little seabirds struggling for survival.

This was followed up by BBC Radio 4’s Today programme and also by the Daily Mail.

But the story is #fakenews.

First, as one reader pointed out, puffin colony numbers go up and down all the time.

Second, as Paul Homewood notes, all that stuff about “climate change, overfishing, plastic pollution and extreme weather” is just pabulum. It’s like a catechism that true believers are required to chant – a ritual demonstration of faith, not the thorough examination of facts that you might hope for from a responsible specialist reporter.

Yes, there seems to be evidence that industrial fishing of sand eels has had a deleterious impact on puffin colonies. But definitely not climate change.

    If numbers rose from 3000 in 1939, to 55000 in 2003, when supposedly we have had global warming, how can it now be responsible for a decline?

Well indeed.

You might argue there’s no harm when the legacy media runs stories like this: all they’re trying to do is sex up a nice nature pic with a bit of attention-grabbing doom and gloom scribble underneath, and it all helps to raise awareness of environmental issues.

But I’d disagree.

#Fakenews environmental crisis stories like this, repeated day in day out, have a cumulative impact in generating precisely the kind of mass hysteria which has led to the great climate change scare.

People feel in their bones that something needs to be done – and urgently – because why would newspapers and the BBC be running this stuff if it weren’t true and a real problem?

Politicians, in turn, feel compelled to respond to this apparent crisis. The resultant policies are invariably disastrous.


Climate change should HELP Midwest corn production

Climate change and global warming put some forms of life at risk, but researchers found one instance that might not feel the heat – corn.

Contrary to previous analyses, research published by Michigan State University shows that projected changes in temperature and humidity will not lead to greater water use in corn. This means that while changes in temperatures and humidity trend as they have in the past 50 years, crop yields can not only survive – but thrive.

“There is a lot of optimism looking at the future for farmers, especially in the Midwest,” said Bruno Basso, lead author of the study and University Distinguished professor.

Basso and his colleague Joe Ritchie, co-author on the study, calculated how much energy crops receive from the sun and how it is converted to evaporative loss from the crop, known as evapotranspiration.

“Think of the energy balance like a bank account. There are additions and subtractions,” Basso said. “The energy coming from the sun is a known, measured quantity that adds to the bank account. The primary subtraction is liquid water from the crop, and soil using the solar energy to convert the water to vapor.”

The researchers used the energy balance to calculate the evaporative water loss for 2017, which set a world record yield of 542 bushels per acre. They found that the water loss was the same as it was for lower yielding crops because the energy balance was about the same.

The trend for the past 50 years of a slightly more humid environment decreases the energy for the crops’ water use.

“Our analysis, and that of other climate researchers, shows that the amount of water vapor in the air is gradually increasing in the summers because the daily low temperatures are getting gradually warmer, but the daily high temperatures are cooling – or staying the same – in many areas of the Midwest,” Basso said. “This causes more humidity and slightly decreases how much energy is used for evaporation.”

Basso also tested a water balance calculation on the crop models that, similar to the energy balance, has additions from rainfall and irrigation and subtractions from evaporation from the crop.

“A water balance is just like the bank account of an energy for crops,” Basso said. “There must be a balance to make crops ‘happy’ so that all the energy reaching the crop surface is evaporated.”

In the United States, as a result of improved hybrids and agronomic practices, corn production has steadily increased by an average of two bushels per acre every year for the past 40 years.

Basso explained that data from the National Corn Growers Association competition for high yields shows the potential for continued higher yields in the future. His findings support that climate change won’t hinder its production if the trend of the past 50 years continues into the next 50 years.

“The energy for evaporation is changing little, so if the number of days the crop grows and uses water is the same now and, in the future, the evaporation loss will be the same and slightly less,” Basso said. “In fact, the warmer temperatures allow the use of longer season hybrids that will make for even greater yield possibilities.”


The German wind energy market is threatened by a sharp downturn after years of continuous growth. Ten thousand jobs have already been cut last year

Düsseldorf: When Volker Malmen sat on stage in a hotel in Bremerhaven just over a week ago, the head of Orsted Germany could not help laughing when the moderator asked him which countries were important for the wind industry as a growth market. “Well, Germany probably not,” replied the managing director. And Malmen’s statement has weight in the industry – the Danish company is one of the world’s leading wind farm operators. The Orsted boss is not alone in his opinion.

According to a survey by the market research institute Windresearch, together with WindEnergy Hamburg, the largest wind fair in the world, the mood in the industry is basically positive – but in Germany the surveyed project planners, operators and manufacturers are not so optimistic about the situation, in some cases even consider it “very negative”. The survey is exclusively available to Handelsblatt.

Germany is the most important sales market for wind power in Europe; last year alone, more wind turbines were installed here than ever before. But while the global wind industry is booming, the German market is threatened by a sharp downturn after years of continuous growth.

In 2017, around 1800 new wind turbines with an output of 5330 megawatts were added in Germany, but in the worst case scenario it could only be 1100 megawatts in 2019. The German Wind Energy Association warns against the loss of thousands of jobs. Ten thousand jobs had already been cut last year.

On the one hand, the industry is facing enormous price erosion due to the worldwide reduction in subsidies and the switch to tenders. On the other hand, Germany is considered a particularly difficult market. Here, the demand for wind turbines had almost collapsed due to the auction system introduced only in 2017.

Over 1200 companies from all over the world were surveyed in the WindEnergy Trend Index 2018, both in the onshore and offshore wind sectors. In both cases, the interviewees assess the situation in Germany as significantly worse than for the rest of Europe, Asia or North America. The onshore industry seems to be particularly concerned, with 38 percent rating the current situation as negative to very negative.

Dirk Briese, Managing Director of windresearch, attributes this to the tenders. “In other countries such auction systems have existed for some time. In addition, the lowest results to date were achieved in Germany. And in the shortest time,” Briese explains.

The operators Orsted, EnBW, Vattenfall and also the Spanish energy supplier Iberdrola won part of the projects with zero-cent bids in two tender rounds. They then want to market their electricity completely without EEG compensation at the price traded on the stock exchange. To date a novelty in the wind industry. According to Briese, this is a very rapid and radical change.

Nevertheless, the majority of people believe that they have grown at this turn of time. They are already expecting a more optimistic mood in the offshore sector this year. It is hoped that the situation could brighten further from 2020. Only half of twenty percent currently believe that the German market is in a negative starting position. The prospects for onshore wind energy are not quite as bright. Here, 19 percent do not believe in an improvement in two years either. However, more than forty percent even assess the situation as positive to very positive in two years’ time.

One reason for this optimism could be further cost reductions in the construction of wind turbines. There’s still potential down here. All those surveyed agree that the biggest leap in offshore wind energy is imminent. Almost seventy percent of those surveyed estimate the chances of saving even more at high to very high levels.

Siemens offshore CEO Andreas Nauen recently called in an interview with the Handelsblatt newspaper for politicians to improve the general conditions in order to prevent Germany from falling behind. “If Germany wants to stay at the top and not fall behind other countries, something has to change. We hope that the expansion corridor for offshore wind energy will widen considerably,” said Nauen. According to the government’s targets, this is 15,000 megawatts (MW) by 2030, Nauen says, which is too little.

The operators Orsted, EnBW, Vattenfall and also the Spanish energy supplier Iberdrola won part of the projects with zero-cent bids in two tender rounds. They then want to market their electricity completely without EEG compensation at the price traded on the stock exchange. To date a novelty in the wind industry. According to Briese, this is a very rapid and radical change.

In the results of the second tender round for onshore wind energy last week, the tendered volume was not reached for the first time. Of 670 MW, only 111 bids with a volume of 604 MW were received.

Companies see markets such as China, India or Taiwan as more promising. The respondents see the best opportunities for the onshore sector from 2020 onwards in Asia, while in the offshore sector they see better conditions for Europe.

In principle, however, growth is increasingly shifting from Europe to Asia and is becoming smaller. This also became clear at the Windforce trade fair in Bremerhaven. The assessment of the industry experts: New projects are being implemented in Asia, while the German market is sluggish.


How 19 Wealthy Foundations Control The Anti-Fossil Fuel Agenda And Escape Scrutiny

Nearly 20 wealthy foundations funneled hundreds of millions of dollars between 2011 and 2015 into a network of environmental organizations to attack the fossil fuel industry, according to a new study published this week by Matthew Nisbet, Ph.D., a Professor of Communication Studies at Northeastern University.

The study, which only analyzed a subset of the organizations active on climate issues, provides a glimpse at the massive funding apparatus behind the anti-fossil fuel echo chamber – and the lack of scrutiny that this big money campaign has faced.

Energy In Depth has previously exposed how wealthy anti-fossil fuel foundations like the Rockefeller Brothers Fund and Rockefeller Family Fund are financing a wide range of activist groups.

These foundations have admitted to funding studies that attack the oil and gas industry, media outlets that provide favorable coverage of those studies, and activist groups to trumpet their anti-fossil fuel agenda online and in the press.

Nisbet’s paper notes that there has been relatively little scrutiny of the presumed independence of these voices in the media:

“When left‐of‐center and progressive foundations are covered in the U.S. press, coverage tends to be predominantly positive and uncritical, deepening a lack of public scrutiny relative to their philanthropic activities, successes, and failures.

“These grantmakers are also among the major patrons for academics and their work and are the main supporters of the rapidly growing nonprofit journalism sector. Many scholars and journalists, therefore, have reason to be cautious in their assessment (Reckhow, 2013).” (emphasis added)

Nisbet emphasizes how the 19 foundations examined in his study, including the Rockefeller Brothers Fund, Bloomberg Foundation, and Hewlett Foundation, have shaped the conversation on climate change, due to the immense concentration of grant money:

“Far from being passive supporters of actions to address climate change, major U.S. foundations for several decades have played an active role in defining a common roadmap for their grantees and partners.

“By framing the challenges, defining the priorities, and promoting specific ideas, philanthropists have actively shaped common ways of thinking that have bound together otherwise disconnected organizations and leaders into shared approaches and strategies.” (emphasis added)

Nisbet cited other literature that described the impact of the foundation approach as an “outsize megaphone, both actively shaping how people view social problems and championing specific methods through which these problems can be addressed.”

The largest of these foundations is the Energy Foundation, and the study describes how its size allows philanthropists to exert major control over the environmental community to focus work on its preferred policies:

“Launched in 1991, the Energy Foundation has been the main instrument that a network of influential U.S. philanthropists has used to define a portfolio of policy options, political strategies, and energy technologies to address climate change.

“Set up by way of large block grants from the Rockefeller Foundation, Pew Charitable Trusts, and MacArthur Foundation, and supported in later years by the Hewlett Foundation, Packard Foundation, and other funders, the principal function of the Energy Foundation has been to leverage money in a highly concentrated pattern on behalf of policies that shift markets, industry, and consumers in the direction of renewable energy technologies and energy efficiency practices.” (emphasis added)

The outsized influence exerted by a small number of wealthy foundations has led to group-think in the climate change conversation and has increased political polarization around the issue by focusing on divisive, anti-fossil fuel policies.

Nisbet cited a few specific strategies supported by the foundations that have had this impact:

“Yet related to these strategies, campaigns opposing the Keystone XL oil pipeline and natural gas fracking along with new causes related to racial, gender, and identity‐based justice have also likely contributed to deepening political polarization, serving as potent symbols for Republican donors and activists to rally voters around.

“These issues also divide liberal and centrist Democrats, and were a major point of contention during the Democratic primaries.” (emphasis added)

The efforts to support the agendas of these foundations is all-encompassing and includes communications, promotion of renewable technologies, and efforts to limit fossil fuel development.

In fact, the “Park Foundation and Rockefeller Brothers Fund…are notable for supporting strategies that directly target the fossil-fuel industry by way of communication, media, and mobilization campaigns,” Nisbet writes.

Meanwhile, the analysis of foundation spending shows alternative ways of reducing greenhouse gas emissions that contribute to climate change remain largely ignored or even attacked.

For example, the study found that $6,834,000 was spent specifically to “oppose, limit natural gas development,” even though increased use of natural gas has allowed for a significant reduction in greenhouse gas emissions in the United States

Emissions of carbon dioxide in the United States have decreased 13 percent from 2005-2016 while natural gas consumption increased 25 percent over the same period.

The study breaks down the attack on natural gas:

“Specific to natural gas fracking, $6.8 million was provided to restrict or ban drilling, $2.1 million to protect drinking water supplies; and $3.9 million for research on health and environmental impacts.

“To support efforts to ban/restrict fracking, Schmidt ($3.3 million), Hewlett ($1.5 million), Park ($1.1 million), and Heinz ($1 million) were the leading funders. Schmidt gave to a mix of national‐ and state‐based groups. Hewlett gave primarily to the Colorado Conservation Fund ($1.3 million).

“Park gave primarily to groups working in New York state, and Heinz to groups in Pennsylvania. Relative to protecting drinking water supplies, major funders included Heinz ($1 million) for efforts in Pennsylvania; and Park ($760,000) for work in North Carolina and New York.

“Major funders of research on fracking’s health and environmental impacts included Heinz ($2.7 million), Park ($780,000), and Schmidt ($390,000). These funds were given to a mix of universities and environmental groups.”

The foundations only granted $1.3 million to support work on carbon capture and sequestration and just $55,000 to promote “fossil fuel industry innovation to limit emissions.”

Nisbet concludes his paper by predicting the next steps in the anti-fossil fuel campaign and the public’s ability to hold these wealthy foundations and their political allies accountable:

“In coming years, as the endowments of major foundations continue to grow, providing philanthropists with ever greater resources, they are likely to play an even more active and strategic role in funding actions to address climate change in the United States and elsewhere.

“In 2017, the Hewlett Foundation, for example, announced it would spend $600 million over the next decade to combat the problem (Gunther, 2018). By framing the challenges and defining the solutions to climate change, as they did in the years following the defeat of the cap and trade bill, Hewlett and other major philanthropies are likely to deepen their ability to bind together organizations and leaders into shared approaches and strategies.

“In an era of political dysfunction and diminished public spending, many will look to philanthropy and their resources for answers. Yet in contrast to elected officials and government agendas, there are few channels to hold funders accountable for their decisions or to a shine a light on their actions…

“Financial support for efforts restricting fossil fuel development and for turning public opinion against the industry is also likely to expand. Examples include municipal lawsuits filed against fossil fuel companies to recover damages for climate change impacts; and decisions by states and cities to divest their pension plans of energy-related stocks.

“To aid these efforts, some funders will also deepen their support for journalistic investigations of the fossil fuel industry. Such strategies, however, are likely to intensify controversy over the ties between funders, advocacy groups, and journalists.” (emphasis added)




Preserving the graphics:  Most graphics on this site are hotlinked from elsewhere.  But hotlinked graphics sometimes have only a short life -- as little as a week in some cases.  After that they no longer come up.  From January 2011 on, therefore, I have posted a monthly copy of everything on this blog to a separate site where I can host text and graphics together -- which should make the graphics available even if they are no longer coming up on this site.  See  here or here


No comments: