[Stoves] USEPA: Absence of evidence is evidence of absence (re: Ron on WHO webinar)

[Traveller]

This too is a long post. "Take aways" for stovers at the end.

I have recently been diagnosed with many illnesses associated with PM2.5
<https://www3.epa.gov/airquality/particlepollution/>- aggravated asthma,
decreased lung function, and irritation of the airways, coughing, and
difficulty breathing.

Since I am a candidate for premature mortality, I decided to poke around
EPA web.

-------------------------

I began with Carbon Pollution Emission Guidelines for Existing Stationary
Sources: Electric Utility Generating Units
<https://www.federalregister.gov/articles/2014/06/18/2014-13726/carbon-pollution-emission-guidelines-for-existing-stationary-sources-electric-utility-generating>
-
A proposed rule, 6/18/14, as part of the US Clean Power Plan (which I
otherwise generally like, science or no science).

"These models assume that all fine particles, regardless of their chemical
composition, are equally potent in causing premature mortality because the
scientific evidence is not yet sufficient to allow differentiation of
effect estimates by particle type."

Whoa! This is justification by assumption!!

I have serious doubts about whether air quality monitoring stations in
developing countries such as India adequately capture all types of
pollutants and are usable for representative outdoor air pollution
exposures, and whether their design and use across the world is appropriate
for local conditions.

But this makes me wonder where EPA's justification for ambient
concentration standards and achievement strategies comes from. Ultimately
its economics are in doubt - in cost and benefit terms - though I grant
once the standards are set, economics may change.

It gets worse:

"In this analysis, the EPA assumes *that the health impact function for
fine particles is without a threshold*. This is based on the conclusions of
EPA's *Integrated Science Assessment for Particulate Matter,* [344]
<https://www.federalregister.gov/articles/2014/06/18/2014-13726/carbon-pollution-emission-guidelines-for-existing-stationary-sources-electric-utility-generating#footnote-344>which
evaluated the substantial body of published scientific literature,
reflecting thousands of epidemiology, toxicology, and clinical studies that
documents the association between elevated PM 2.5 *concentrations* and adverse
health effects, including increased premature mortality. This assessment,
which was twice reviewed by the EPA's independent Science Advisory Board,
concluded that the scientific literature consistently finds that a
no-threshold model most adequately portrays the PM-mortality
*concentration-response
relationship*." (Emphasis added.)


What is "the relationship" between PM of any type and size (2.5 and below)
and how are these correlations arrived at across different times,
populations with different health conditions and genetic makeup?

What is being assumed constant in all the statistical models that is hugely
variable across populations and over time?

Inasmuch as PM2.5 concentrations may have declined in many areas over the
last few decades when these studies were done and will continue to do so,
why is EPA continuing with some invariable relationship, even while
admitting that it is "less confident" (code for diffident, nervous) in
doing so?

I quoted Burnett (2014) in an earlier post, "we are not suggesting that
there is convincing evidence that PM mortality and ALRI risk is zero below
any specific concentration based on biological considerations (Brook et al.
2010). *Absence of such evidence from epidemiologic studies does not
necessarily imply evidence of the absence *of such a counterfactual
concentration. We thus take the conservative approach and
set a positive counterfactual concentration. However, our approach can
be adapted to a
different counterfactual if new evidence supporting a positive association
at lower concentrations becomes available."

How can it be simultaneously the case that there is no convincing evidence
of a threshold (onetime, annual, lifetime) for PM exposure for any
particular population AND that the "no-threshold model most adequately
portrays the PM-mortality concentration-response relationship."?

Besides, note that EPA discusses only *"concentration-response"
relationship; we have moved far away from "exposure-response relationships.*

Have data, will kill.

Assume emissions, assume concentrations (in the developing world), and
assume "concentration-response" relationship. Voila! Another paper that
EPA's next round of Integrated Assessment for PM will review. The
Assessment will then be reviewed, maybe twice, by EPA's independent (how
can anything be EPA's and independent?) advisory board?

"Second-hand" science Ronald Fisher warned about is as good as "second
hand" second hand smoke.

Literature review. Meta-analysis. These are tools of deception.

We have all joined Alice.

----------

"In general, we are more confident in the magnitude of the risks we
estimate from simulated PM 2.5 concentrations that coincide with the bulk
of the observed PM concentrations in the epidemiological studies that are
used to estimate the benefits. Likewise, we are less confident in the risk
we estimate from simulated PM 2.5 concentrations that fall below the bulk
of the observed data in these studies.

For this analysis, *policy-specific air quality data are not available,**
<https://www.federalregister.gov/articles/2014/06/18/2014-13726/carbon-pollution-emission-guidelines-for-existing-stationary-sources-electric-utility-generating#footnote-345>and
thus, we are unable to estimate the percentage of premature mortality
associated with this specific rule's emission reductions at each PM 2.5 *
*level*."

How reliable, pray tell, are EPA's multi-decadal forecasts for PM2.5
(including dust, pollen, open burning, forest fires, some of which are
"natural") in the first place? Can everything be controlled?

EPA says
<https://www3.epa.gov/airquality/particlepollution/actions.html#dec12> that
by 2020, even without this Clean Power Plan, 99% of US counties (with
monitors; who cares when there is no monitoring) will meet EPA's tightened
2012 standard of PM2.5.

So, EPA knowingly computes health benefits and monetizes them that it is
"less confident" about.

----------------

It gets curiouser and curiouser. What *are* the new (2012) EPA standards
<https://www3.epa.gov/airquality/particlepollution/2012/decfsstandards.pdf>?


"An area will meet the standard if the *three-year average of its annual
average* PM2.5 concentration (at each monitoring site in the area) is less
than or equal to 12.0 µg/m3 ." and " An area meets the 24-hour standard if
the *98th percentile of 24-hour PM2.5 concentrations in one year, averaged
over three years*, is less than or equal to 35 μg/m3 ." The PM10 standards,
correspondingly are, "An area meets the 24-hour PM10 standard if it does
not exceed the 150 µg/m3 level more than once per year on average over a
three-year period."


So, all you have is an average of averages at monitoring sites, particular
locations heights that may or may not correspond to human exposures. No
matter; concentrations, not exposures, kill. By EPA fiat.

In developing countries, the spotty air quality monitors do not even
measure arguably more toxic pathogens from open wastes, chemical spills,
rural fires and dust storms (except when they are transported to these holy
shrines of air pollution religion.)

If EPA's concentration standard has fuzzy justification, what's the point
of source emission standards? Just air basin modeling?

How many 30-year forecasts for air quality and premature mortality have
been made in the last forty years, and what is the record so far?

Oops, the rates and composition of incidence of disease have changed and
arguably exposures have changed in a way different from concentrations, so
all of this is forward-looking promises of a securities trader.

Sub-prime regulation based on concocting false assurances?

-------------------------
>From what I see here as jiggery-pokery, I am relieved. I could die any time
but probably not from PM2.5 concentrations in Washington, DC.
I wonder if some scientists would call this bankruptcy of science in the
service of regulatory zeal, climate change mania?

Dick Schmalensee and Bob Stavins, in their 2013 paper The SO2 Allowance
Trading System:The Ironic History of a Grand Policy Experiment
<http://web.mit.edu/ceepr/www/publications/reprints/Reprint_248_WC.pdf>*
concluded: "The central purpose of the SO2 allowance trading program was to
reduce the the acidification of forest and aquatic ecosystems.. The goal of
reducing SO2 emissions was met and exceeded. However, it turns out that the
ecological benefits of the program have been relatively small, largely
because it takes much longer than thought to reverse the acidification of
ecosystems .. On the other hand, other completely unanticipated benefits of
the program have been massive."

My cynical re-reading of history - which I observed first and second-hand -
is that the central drivers of that grand policy experiment were political
- a) ideological opposition to coal, no matter what the merits; lb)
Washington power struggle between coal states of Appalachia and Illinois
Basin versus coal states of the West; and, c) power struggles between gas
and coal interests, groups within the coal interests, and all the ancillary
enterprises.

Schmalensee and Stavins (who might have been in the White House CEQ at the
time of the "acid rain" debates and the "allowance trading" compromise
compared to EPA's earlier obsession with NSPS) further write

"..but instead with human health impacts of reduced levels of airborne fine
sulfate particles less than 2.5 micrometers in diameter (PM ter (PM2.5),
particles which derive from SO2 emissions. Epidemiological evidence of the
harmful human health effects of these fine particulates mounted rapidly in
the decade he decade after the CAAA was enacted .... human health benefits
of the program may have exceeded annual costs by a factor of more than
fifty! With its mandated 50 percent cut in SO2 emissions, the government
did what turned out to be the right thing for the wrong reason."


"Right thing"? These are all imputed benefits by assumption loaded on
assumption. There is no evidence that actual health costs - disease
treatment, lost days of work, even death - declined compared to some
baseline estimate prior to 1995. Besides, the environment changed - air and
water quality - as did the populace - aging, immigrants, internal
migration, new housing stock.

When your religion says that concentrations kill - damn exposures - then
you only need to assume that emissions anywhere change concentrations
everywhere, and voila, you have saved people!

Burnett (2014) have already told us the limits of epidemiological evidence,
and deployment of fancier models with more restrictive assumptions in order
to come up with "relative risk"; further explorations are required.

--------------
Now I am beginning to wonder whether the "epidemiological evidence" was
simply "evidence by assumption" and "changed circumstances" that *required
assuming that PM2.5 exposure was equally toxic no matter what species*.

Because, as the "acid rain" amendments to the Clean Air Act (1990)
required, SO2 emissions fell. A further boost was provided by higher
utilization factors of the US nuclear power fleet. Presumably, local
concentrations also fell.

With lower sulfates in the air, *did EPA scientists need to assume
equitoxicity in order to keep using old data on concentration-response
"relationship"*?

Then, if equitoxicity was not enough in predicting premature mortality and
valuation of statistical lives lost - EPA now recommends about $9 million
as Value of Statistical Life ($7.4 million in 2006$
<https://www.epa.gov/environmental-economics/mortality-risk-valuation>; I
inflate to the present) -* did it need to also assume "no threshold" to
raise mortality?. *

Killing by assumption is lucrative. Evidence of absence is a convenient
untruth.

-------------
EPA has learnt a lesson from the acid rain program: promises to "save
lives" sell better than promises to restore ecosystems. (I can make a model
with synergies to contradict this; ready to write a grant proposal to EPA?)

"Premature mortality avoided" is the yardstick by which EPA justifies its
regulatory reach.

By EPA's calculation, climate change action has annual benefit of 69,000
deaths averted by 2100.
<http://www.cnsnews.com/news/article/penny-starr/epa-chief-fighting-climate-change-could-prevent-69000-deaths-pollution-2100>
That could be worth $7 trillion a year by 2100. (In 2100$, that is. I
assume VSL will rise to $100 million in 2100, with inflation and higher
productivities, i.e., higher GDP.)

(I don't know how VSL factors in YLL. My premature death - on the basis of
average life expectancy for a Gujarati male born in 1950s and properly
vaccinated - will have fewer YLL than that compared to GBD's new universal
health target - all deaths before age 86 are "premature". I am willing to
buy DALY reduction.)

The Bush EPA failed to convince the US Supreme Court
<https://www.supremecourt.gov/opinions/06pdf/05-1120.pdf> - save Antonin
Scalia, who has suffered a premature death, and John Roberts (still around)
- that CO2 molecules are not weapons of mass destruction.

Then the Obama EPA made an "endangerment finding
<https://www.gpo.gov/fdsys/pkg/FR-2009-12-15/pdf/E9-29537.pdf>" with
respect to climate change, that "the air pollution is the combined mix of
six key directly-emitted, long-lived and well-mixed greenhouse gases
(henceforth ‘‘well-mixed greenhouse gases’’), which together, constitute
the root cause of human-induced climate change and the resulting impacts on
public health and welfare."

However, I could find only one specific reference to premature mortality -
to do with groundlevel ozone, which is of course a potent warmer. There's
much on temperature-related mortality - unclear evidence on net morbidity
(heat v. cold) - and reference to aero-allergens/temperature (no basis for
judgment). The primary culprit is ozone (whose precursors are sometimes
coemitted with PM, as is the case with inefficient solid fuels use).

It is interesting to see how the concept of "contribution" is seen in legal
terms. (I see "attribution" analogous to "contribution" but I haven't done
the legal research for that.) EPA says,

"In upholding EPA’s PM2.5 attainment and nonattainment designation
decisions, the DC Circuit analyzed CAA section 107(d), which requires EPA
to designate an area as nonattainment if it ‘‘contributes to ambient air
quality in a nearby area’’ not attaining the national ambient air quality
standards. Id. at 35. The court noted that it had previously held that the
term ‘‘contributes’’ is ambiguous in the context of CAA language. See EDF
v. EPA, 82 F.3d 451, 459 (DC Cir. 1996). ‘‘[A]mbiguities in statutes within
an agency’s jurisdiction to administer are delegations of authority to the
agency to fill the statutory gap in reasonable fashion.’’ 571 F.3d at 35
(citing Nat’s Cable & Telecomms. Ass’c v. Brand X Internet Servs, 545 U.S.
967, 980 (2005)). *The court then proceeded to consider and reject
petitioners’ argument that the verb ‘‘contributes’’ in CAA section 107(d)
necessarily connotes a significant causal relationship*. S*pecifically, the
DC Circuit again noted that the term is ambiguous, leaving it to EPA to
interpret in a reasonable manner. In the context of this discussion, the
court noted that ‘‘a contribution may simply exacerbate a problem rather
than cause it ** * * ’’ 571 F.3d at 39. This is consistent with the DC
Circuit’s decision in Bluewater Network v. EPA, 370 F.3d 1 (DC Cir. 2004),
in which the court noted that the term contribute in CAA section 213(a)(3)
‘‘[s]tanding alone, * * * has no inherent connotation as to the magnitude
or importance of the relevant ‘share’ in the effect; certainly it does not
incorporate any ‘significance’ requirement.’’ 370 F.3d at 13. The court
found that the bare ‘‘contribute’’ language invests the Administrator with
discretion to exercise judgment regarding what constitutes a sufficient
contribution for the purpose of making an endangerment finding. Id. at 14.
" From here <https://www.gpo.gov/fdsys/pkg/FR-2009-12-15/pdf/E9-29537.pdf>.

Contribution does not necessarily connote a causal relationship.

I suspect the same is true of attribution; not necessarily a causal
relationship.

GACC/IEA propaganda to the contrary. Relying on state of the law so far.
(Like emissions, law changes.)

-----------------------------------
*Lessons for us stovers: *

1. Do not get too anxious that "dirty cooking" kills. What kills whom how
is never that clear to doctors, and the GBD statements about "attributable
to risk factors" are allegations in evolution.

2. There is no evidentiary basis to assume equitoxicity and "no threshold"
for PM2.5. At least, not for the air pollution concentrations and
composition found in poor households for whom we debate "clean cooking."

3. Just because EPA has shifted from "exposure-response" to
"concentration-response" does not mean everybody everywhere has to fall for
the emissions -> concentrations -> death theology. At least, no promises of
"saving lives" are worth a statistical US life.

4. Insist on a long-term concentration profile, even if exposures aren't
measured, instead of the EPA/ISO gobbledygook about emission RATES (whether
lab or field tests). Concentration data have to include all air pollutants
from all sources in specific environments, not this WHO gaming with stove
emissions.

5. Then there're additional problems of keeping all risk factors unchanged
(my prior post), confounding bias, and the diversity among humans according
to their genetic basis, physical environments (agroclimatic zones, dwelling
type, transport).

6. Beware of US politics. EPA is becoming an Empire Promotion Alliance.

I am afraid what Cecil calls "fundamental foolishness" is a diversion from
more relevant research and planning issue - air quality and exposure
management strategies of the conventional variety. We cannot pretend that
the last 60 years of air management - science, economics, spatial planning,
buildings design - is worthless and "new source voluntary performance
targets" will save our sisters, daughters, grand daughters.

Not that emission rates are entirely irrelevant.

In air quality strategies, reduction in emission rates - by source type,
stationary or mobile, small or large - is but one of the means of lowering
concentrations to standards and presumably dosage and hence damage. For
large stationary sources with continuous emissions, usually a daily average
is specified for some pollutants (e.g. SO2). But with small sources like
household cookstoves - which can vary enormously in fuel quality and
operating practices - a per minute max or a 15-minute max would seem to be
an appropriate metric, along with some hourly average max.

I remember the controversies on SO2 or low-level ionizing radiation, or
other pollutants from sources large and small, and how the averaging times,
maximum one-time and lifetime exposures were incorporated in the standards
for equipment design or operations. But all that was real epidemiology of
cause and effect. With all this cookstoves stuff, all you have is
"premature deaths", a statistical artifact for allocating across other
statistical artifacts called "risk factors". A premature death is not a
death, and a risk factor is not a disease. Risk factors certainly
contribute to diseases and deaths, the causality links for which are
sometimes quite vague, at least debatable.

Nikhil




------------------------------------------------------------------------
Nikhil Desai
(US +1) 202 568 5831
*Skype: nikhildesai888*
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.bioenergylists.org/pipermail/stoves_lists.bioenergylists.org/attachments/20160910/830609b0/attachment.html>