Welcome to DU! The truly grassroots left-of-center political community where regular people, not algorithms, drive the discussions and set the standards. Join the community: Create a free account Support DU (and get rid of ads!): Become a Star Member Latest Breaking News General Discussion The DU Lounge All Forums Issue Forums Culture Forums Alliance Forums Region Forums Support Forums Help & Search

NNadir

NNadir's Journal
NNadir's Journal
May 2, 2021

Well, folks, we "got there:" The first weekly average CO2 concentration over 420 ppm at Mauna Loa.

As I've indicated several times I somewhat obsessively keep a spreadsheet of the weekly data at the Mauna Loa Carbon Dioxide Observatory, which I use to do calculations to record the dying of our atmosphere, a triumph of fear, dogma and ignorance that did not have to be, but nonetheless is.

This week's reading is the first in the history of weekly average readings, going back, to 1975 posted by the Mauna Loa is the highest ever recorded at the Mauna Loa carbon dioxide observatory, 420.01 ppm.

Generally, each year, these measurements peak in late May or early June. We probably haven't seen the worst measurement of 2021 yet, despite all the enthusiasm for the idea that Covid lockdowns would slow climate change. If they did, it's imperceptible.



The figures for this past week:


Week beginning on April 25, 2021: 420.01 ppm
Weekly value from 1 year ago: 416.95 ppm
Weekly value from 10 years ago: 393.48 ppm
Last updated: May 2, 2021



Up-to-date weekly average CO2 at Mauna Loa


The increase in carbon dioxide concentrations when compared to the same week in 2020 is "only" 3.04 ppm. (If one keeps track as I do, there is a fair amount of statistical noise in these measurements, but the trends are consistent.) The highest weekly increase over 2020 this year, 2021, was 3.90 ppm, observed in the week beginning February 28, 2021.

In my spreadsheet, I keep records of the increases over 10 year periods, in this case, a comparison of the reading this past week, with the last week of May in 2011. Using Excel functions, I can sort them by values high to low and do a lot of other things. The value for the 10 year increase is the highest ever recorded, 26.53 ppm.

The 12 month running average for increases over a ten year period, week to week, 2021 to 2011, is 24.37 ppm, 2.44 ppm per year and rising.

If any of this troubles you, don't worry, be happy. Head on over to Benny Sovacool's paper in Environmental Science and Technology about how so called "renewable energy" could save as many lives as nuclear energy saves, and could prevent as much carbon dioxide emissions as nuclear energy has been doing consistently for decades if only...if only...if only...if only...

If only what, exactly?

Positive Externalities of Decarbonization: Quantifying the Full Potential of Avoided Deaths and Displaced Carbon Emissions from Renewable Energy and Nuclear Power (Benjamin K. Sovacool and Chukwuka G. Monyei, Environmental Science & Technology 2021 55 (8), 5258-5271.

I discussed "Benny boy's" paper last night, with my head about to explode, because I'm sick and tired of hearing what so called "renewable energy" could do. I've been hearing it for half a century, and here we are...here we are...

The anti-nuke squad tries to steal Jim Hansen's thunder and manages to look stupid again.

At least it does seem that "Benny Boy" finally may have conceded, after spending his entire adult life bashing nuclear energy, that it saves lives.

That doesn't stop him from repeating his cult mantras that so called "renewable energy" is the bestest bestest best yet, even if we have to put dredging mines to haul up metal ores from the seafloor to get materials to convert our last wilderness into industrial parks for wind turbines.

Guess what? So called "renewable energy" hasn't done anything. It isn't doing anything. It won't do anything, where "anything" is saving human lives from air pollution and addressing climate change.

It's not "we could hit 420 ppm concentrations of the dangerous fossil fuel waste carbon dioxide in the planetary atmosphere;" It's "we did hit 420 ppm concentrations of the dangerous fossil fuel waste carbon dioxide in the planetary atmosphere." We did this after half a century of cheering wildly for wind and solar energy.

The history of the reactionary attempt to go to a so called "renewable energy" powered world like the one abandoned by humanity in the 19th and early 20th century, this on a planet with 1/7th the population it supports now, is filled with conditional "could" statements. And it's not about money, because we've been throwing oodles of money at so called "renewable energy" for this entire century, on a scale of trillions of dollars:

The amount of money "invested" in so called "renewable energy" in the period between 2004 and 2018 is over 3.036 trillion dollars; dominated by solar and wind which soaked up 2.774 trillion dollars.
Source: UNEP/Bloomberg Global Investment in Renewable Energy, 2019

The result of spending all of this money has had no result with respect to addressing climate change, other than the result that as a result of this squandering on stuff about which we liked to talk but didn't work - that's the only word for it, "squandering" - we're at 420 ppm.

That's a fact.

Facts matter.

This isn't a popular statement, necessarily, among my fellow Democrats - and I've gotten in lots of trouble for making it even as I insist it is true - but the reality is that opposing nuclear energy is rapidly morphing into a crime against humanity.

Have a nice Sunday afternoon.
May 2, 2021

A Spectroscopic View on Cosmic Polycyclic Aromatic Hydrocarbon (PAH) Emissions

The paper I'll discuss in this post is this one: A Spectroscopic View on Cosmic PAH Emission (Els Peeters, Cameron Mackie, Alessandra Candian, and Alexander G. G. M. Tielens, Accounts of Chemical Research 2021 54 (8), 1921-1933)

Accounts of Chemical Research is a journal to which I used to subscribe, in the old days of paper journals, when I was a kid. Over the years I sort of drifted away from it, although it's a magnificent journal, as I see whenever I do get around to looking at it, ably edited by one of the best professors from whom I ever took a class, where I learned all about cycloadditions. She must be nearing retirement now.

PAHs on Earth are responsible, mechanistically, for a large portion of the deaths from combustion wastes, both from the combustion of dangerous fossil fuels and the combustion of biomass, including, but hardly limited to, tobacco. This is because they are planar molecules, often with oxidizable oxygen functional groups. Because they are planar, they can fit in the grooves of DNA, sometimes bonding to the nitrogen bases causing adducts that can and do lead to cancer, and other diseases.

This account is not about air pollution or even terrestrial life, but is about the formation of these molecules in deep space and their chemistry there. The abstract contains this wonderful statement:

Throughout this Account, we emphasize that huge progress is on the horizon on the astronomical front. Indeed, the world is eagerly awaiting the launch of the James Webb Space Telescope (JWST). With its incredible improvement in spatial resolution, combined with its complete spectral coverage of the PAH infrared emission bands at medium spectral resolution and superb sensitivity, the JWST will revolutionize PAH research. Previous observations could only present spectra averaged over regions with vastly different properties, thus greatly confusing their interpretation. The amazing spatial resolution of JWST will disentangle these different regions. This will allow us to quantify precisely how PAHs are modified by the physical conditions of their host environment and thus trace how PAHs evolve across space. However, this will only be achieved when the necessary (and still missing) fundamental properties of PAHs, outlined in this Account, are known. We strongly encourage you to join this effort.


From the introduction to the paper:

The infrared (IR) spectra of objects associated with dust and gas—including the interstellar medium (ISM), star-forming regions, surface layers of proto-planetary disks (the sites of planet formation), and galaxies out to redshifts of z ? 4—are dominated by strong, broad emission bands (Figure 1). Known collectively as the aromatic infrared bands (AIBs), the main bands are found at 3.3, 6.2, 7.7, 8.6, and 11.2 ?m (3030, 1612, 1299, 1163, and 893 cm–1) along with a plethora of weaker bands which are perched on top of broad emission plateaus. Multiple carriers have been proposed for the AIBs: hydrogenated amorphous carbon (HAC),(6) quenched carbon composites (QCCs),(7) polycyclic aromatic hydrocarbons (PAHs),(8,9) nanodiamonds,(10) locally aromatic polycyclic hydrocarbons,(11) and mixed aromatic–aliphatic organic nanoparticles.(12) However, this emission is generally attributed to a family of PAH molecules with a typical size of 50–100 C atoms (NC).(8,9) Upon absorption of a UV photon, PAHs relax by IR fluorescence via their vibrational modes which are detected as the AIBs (Figure 1). The C–H stretching mode gives rise to the 3.3 ?m? (3030 cm–1) emission, the C–C stretching modes to the 6.2 and 7.7 ?m? (1612 and 1299 cm–1) emission, the C–H in-plane bending mode to the 8.6 ?m? (1163 cm–1) emission, and the C–H out-of-plane bending mode to the 10–15 ?m? (1000–666 cm–1) emission. Within the astrophysical context, the PAH family includes PAH-related species such as, for example, PAHs with side groups, heterosubstituted PAHs, fully or partially (de)hydrogenated PAHs, and PAH clusters (for a definition see ref? 13)). While aromatic species such as benzene, benzonitrile, and fullerenes (C60 and C70) have been detected in space,(14?17) identification of specific members of this PAH family (or classes within) remains a mystery.


They are very involved in the chemistry of space:

To date, PAHs are among the largest and most complex molecules known in space. They emit up to 15% of the total energy output of galaxies and contain ?10% of the cosmic carbon. PAHs pervade the Universe. PAHs play a crucial role in several astrophysical and astrochemical processes and thus influence star and planet formation. They control the temperature of the gas in the neutral ISM through the photoelectric effect and provide large surface areas for interstellar chemistry. They influence the charge balance and thus molecular abundances in interstellar clouds. In turn, the composition of the PAH population (e.g., size distribution, charge balance, molecular structure) is influenced by the local physical conditions via photoprocessing, shocks, and formation and destruction processes...


Pictures from the text, beginning with the advertising graphic that has a nice picture of PAH planar structures, as well as the obviously non-planar fullerene which has also been observed in space:





The caption:

Figure 1. Top: Spitzer-IRAC image of the Andromeda galaxy with the 3.6 and 4.5 ?m (2778 and 2222 cm–1) emission (respectively blue and green) representing the stellar population and the 8.0 ?m (1250 cm–1) emission (red) showcasing the ubiquitous PAH emission. Credit: NASA/JPL-Caltech/P. Barmby (Harvard-Smithsonian CfA). Bottom: The IR spectrum of the star-forming region, the Orion Bar, observed by the Infrared Space Observatory (ISO). The AIBs are shown in red, superimposed on top of broad plateaus. The corresponding vibrational modes are labeled at the top. Adapted with permission from ref (25). Copyright 2004 Astronomical Society of the Pacific.





The caption:

Figure 2. Spectroscopic variation in the 6–9 ?m? (1666–1111 cm–1) PAH emission. Profiles are normalized to the peak intensity (Jy) in the 8 ?m? (1250 cm–1) region. Strong aliphatic emission at 6.9 and 7.25 ?m? (1449 and 1379 cm–1, purple) is seen in class . Class is represented by IRAS 23133 + 6050, class by HD 44179, class by IRAS 13416-6243, and class by IRAS f05110-6616. Adapted with permission from ref (26). Copyright 2002 EDP Sciences.




The caption:

Figure 3. Left: Variation of PAH charge, as traced by the 6.2/11.2 PAH ratio (1612/893, top),(3) and dominating PAH size, as traced by the 3.3/11.2 PAH ratio (3030/893, bottom),(57,58) as a function of distance from the illuminating star (i.e., UV radiation field strength) in NGC 2023. Adapted with permission from refs (3) and (58). Copyright 2017 AAS and 2020 Collin Knight. Right: Illustrating the observed range in the relative strength of the CH out-of-plane bending modes (top panels). The bottom panel shows the edge assignment in terms of CH groups based on harmonic DFT calculations. Adapted with permission from ref (37). Copyright 2009 AAS.





The caption:

Figure 4. Left: IR absorption spectrum of chrysene in the 3.3 ?m? (3333 cm–1) region: harmonic spectrum (top), anharmonic spectrum (middle), and experimental gas-phase spectrum (at 20 K; bottom).(91) Right: IR absorption spectrum of tetracene in the 5–17 ?m? (2000–588 cm–1) region: experimental gas-phase spectrum (at 20 K; top), anharmonic spectrum (middle), and harmonic spectrum (bottom).(93) Reproduced with permission from refs (91) and (93). Copyright 2016 AAS and 2019 EDP Sciences.






The caption:

Figure 5. Measured absorption spectrum of naphthalene (at 300 K; bottom(104)) compared to the calculated spectrum in the anharmonic approximation (blue sticks) convolved with a 0.02 cm–1 Gaussian (at 300 K; top(4)). Anharmonic interactions of the fundamental (?46) with specific modes result in “satellite” bands shifted to lower or higher frequency. Reproduced with permission from refs (4) and (104). Copyright 2018 and 2009 AIP Publishing






The caption:

Figure 6. Calculated emission spectrum of the CH out-of-plane bending mode of anthracene (C14H10) as a function of the excitation energy (colored traces). The black trace is the spectrum calculated following the full energy cascade.


Here is the structure of one of the molecules described below, circumcoronene and a related molecule:




Here is the structure of another of the molecules described below, Circumcircumcoronene:



I have never worked on the synthetic chemistry of PAHs, but I'd guess, even with their high symmetry, these are non-trivial molecules to make. When I was a kid, there were all kinds of very smart people around trying to synthesize fullerene, mostly unsuccessfully. The Nobel Prize winning work of its discovery found that it is a component of lampblack.

(Today there's all kinds of fullerene chemistry going on.)

The last graphic from the paper:




The caption:

Figure 7. Characteristics of circumcoronene (C54H18; left) and circumcircumcoronene C96H24; right) as a function of the strength of the FUV field, G0, and the atomic H density, nH. For high FUV fields, PAHs will be quickly completely dehydrogenated (labeled clusters), while, for high atomic H densities, PAHs will be superhydrogenated (labeled hydro). The two ovals in each panel indicate typical conditions in PDRs (top oval) and in diffuse interstellar clouds (bottom oval). Dashed black lines indicate constant G0/nH ratios.(1) Reproduced with permission from ref (1). Copyright 2016 EDP Sciences.


Recently, during the Q&A section of a talk at the virtual ACS meeting on the subject of the origins of life, which focused on showing that it originated on this planet, I made a case for the panspermic theory that the molecules of life, if not life itself, originated in space. The speaker dismissed my question with a remark that amounted to, "I don't study that, so your question is not worth addressing."

Oh well then.

One thing is certain, astrochemistry is a rich field with much to learn.

I'm not a "let's all take a ride to Mars with Elon Musk and inhabit it and build cities and make oxygen, because well, we've wrecked this planet" kind of guy.

I think in most cases robots in space do a better job than humans, more safely, at less cost. Although I'm convinced we need to focus our attention on the air above us, the water we drink, and the land beneath our feet, I also believe that the highest purpose of humanity is to extend our vision.

I was glad to learn about the James Webb telescope through this paper.

Have a nice day.


May 2, 2021

The anti-nuke squad tries to steal Jim Hansen's thunder and manages to look stupid again.

In 2013, the famous climate scientist, Jim Hansen, working with lead author Pushker A. Kharecha published a now fairly famous and widely cited paper showing that nuclear power saves lives: Prevented Mortality and Greenhouse Gas Emissions from Historical and Projected Nuclear Power( Pushker A. Kharecha and James E. Hansen
Environmental Science & Technology 2013 47 (9), 4889-4895).

Almost immediately, the "renewables will save us" anti-nukes Benjamin Sovacool, and filer of lawsuits against PNAS, Mark Z. Jacobson, piped in to say that "nuclear energy is too expensive...too dangerous...blah...blah...blah..." and that wind and solar and finding whatever rivers are left to destroy and destroying them, would save the world better than nuclear could. (Jacobson's lawsuit against PNAS involved PNAS publishing a criticism authored by a plethora of other scientists, including some at his own institution, of his "renewable energy will save the world," paper published in um, PNAS, suggesting that his "save the world" ideas wouldn't work.) Apparently neither 1.8 million lives (as of 2013) already saved by nuclear power were "too expensive" and "too dangerous." This took place in the comments section of a subsequent issue of EST.

Kharecha and Hansen's dismissal of their whining was a joy to behold.

Of course, I've been hearing my whole adult life that "renewable energy will save the world," and I'm certainly not young. It's been going on loudly, pronounced by anti-nukes like Sovacool but it hasn't done shit to save lives historically. Since 2013, eight years ago, somewhere around 50 million people lost their lives to air pollution, this while trillion dollar quantities of money were thrown at solar and wind fantasies for no result.

Global, regional, and national comparative risk assessment of 79 behavioural, environmental and occupational, and metabolic risks or clusters of risks, 1990–2015: a systematic analysis for the Global Burden of Disease Study 2015 (Lancet 2016; 388: 1659–724) One can easily locate in this open sourced document compiled by an international consortium of medical and scientific professionals how many people die from causes related to air pollution, particulates, ozone, etc.

Now Sovacool, is pretending to have a brain by parroting Hansen's methodology and applying it to putative solar and wind fantasies that haven't come, aren't here, and won't be here at any time on any scale that is meaningful despite more than half a century of loud cheering: Positive Externalities of Decarbonization: Quantifying the Full Potential of Avoided Deaths and Displaced Carbon Emissions from Renewable Energy and Nuclear Power (Benjamin K. Sovacool and Chukwuka G. Monyei, Environmental Science & Technology 2021 55 (8), 5258-5271.

Let me reprint the title of their paper: Positive Externalities of Decarbonization: Quantifying the Full Potential of Avoided Deaths and Displaced Carbon Emissions from Renewable Energy and Nuclear Power.

The italics and bold in the repeated title are mine.

It is worth noting that Kharecha's and Hansen's paper included the word "historical," which means that it's based on real data, and not modeling or projection.

I have called up the full text of this appalling paper of Sovacool's. Nowhere in it do the words "steel" or "copper" or "lanthanides" (or the oft used term for "lanthanides," "rare earths" ) appear. No LCA (life cycle analysis) is mentioned except to cite a single 2008 paper (among thousands of papers) about the LCA of nuclear energy, with no data from it reported in the text. Um, firing up a steel retort filled with coke made by heating coal with coal does in fact, kill people, in normal operations.

I know from experience with reading Benny-boy's tripe that he himself knows essentially zero about nuclear engineering, or for that matter, nuclear fuel. (I've read his Ph.D. thesis which involved among other things, interviewing anonymous power company executives.)

It took Sovacool 8 years, again measured by 50 million air pollution deaths, to come up with this list of offering, in lieu of real data about real energy produced, a list of scenarios, about so called "renewable energy," "could" do if only this or only that... All of these scenarios have been falling by the wayside decade after decade while people die and the atmosphere collapses, wildernesses are destroyed to chase after bourgeois fantasies, poor people are consigned to dig cobalt for electric cars and otherwise ignored, etc., etc., etc.

This statement from the paper is as fallacious as a statement that chloroquinine cures Covid:

Third, despite a declining contribution of major fossil fuel sources such as oil and coal to primary energy generation, projections show natural gas maintaining a significant role as a primary energy source. For instance, while gas is estimated to contribute to 13.2% of the historical emissions, computed projections show it increasing its share of emissions to 18.3%. This is despite a considerable decrease of about 62% between historical and projected values. A reason for this may be hinged on the ability of gas to help balance the intermittency of some renewable energy sources, notably wind and solar, and to also assist with compressed air energy storage. However, caution is advocated to ensure gas serves as a transitory energy provider rather than acting as a permanent replacement.


The use of every fossil fuel in this century has been rising, not falling, and gas is hardly "transitional" - this is the really, really, really big lie - because the increase in the use of gas has been vastly outstripping the use (measured in units of energy, exajoules) much faster than all the solar and wind energy generated on the planet.

Here is the data posted this morning of the Mauna Loa Carbon Dioxide Observatory over the last 5 days as of this writing:

April 30: 420.54 ppm
April 29: 419.85 ppm
April 28: 419.75 ppm
April 27: 419.49 ppm
April 26: 420.01 ppm
Last Updated: May 1, 2021

Recent Daily Average Mauna Loa CO2 (Accessed 5/2/21, 12:01 pm)

The first time we ever saw a weekly average reading higher than 400 ppm (400.03 ppm), was on May 26, 2013, the year Hansen's paper was published. Less than 8 years later, we're scraping 420 ppm.

In the year 2000, around the time that Benny Boy types and their cohorts got their way in a major industrial nation when Germany signed on to shut its nuclear plants and go reactionary with so called renewable energy, the highest reading of that year was recorded for the week beginning April 26, 2000: 372.42 ppm.

Twenty-one years, almost 50 ppm higher! And you know what, the rate of increases in carbon dioxide is accelerating, not decelerating.

Sorry Benny Boy, but the time for so called "renewable energy" to shit or get off the pot was decades ago. It didn't. In fact, there is a reason it was abandoned in the 19th century. You ought to look it up. It's way too fucking late for more "scenarios." All the money in the world is not going to make so called "renewable energy" work. Maybe, just maybe, it's beginning, slightly, to seep in, I did see the Benny Boy paper about how we could "just" mine the ocean floor for metals because so called "renewable energy" might use up all those on land.

To wit:

Although mining in terrestrial areas is likely to continue to meet the demands of low-carbon technologies in the nearer term, we need to carefully consider mineral sources beneath the oceans in the longer term. The International Seabed Authority, set up under the United Nations (UN) Convention on the Law of the Sea, is in the process of issuing regulations related to oceanic mineral extraction. This process is a rare opportunity to be proactive in setting forth science-based environmental safeguards for mineral extraction. For metals such as cobalt and nickel, ocean minerals hold important prospects on the continental shelf within states' exclusive economic zones as well as the outer continental shelf regions. Within international waters, metallic nodules found in the vast Clarion-Clipperton Zone of the Pacific as well as in cobalt and tellurium crusts, which are found in seamounts worldwide, provide some of the richest deposits of metals for green technologies.


Sovacool et al., Science 03 Jan 2020: Vol. 367, Issue 6473, pp. 30-33 (Sustainable minerals and metals for a low-carbon future.)

One would have to be completely out one's fucking mind to write something quite so delusional. You're going to mine the ocean floor powered by the sun and the wind only?

Out of your fucking mind...

This would be a good time to start thinking Benny boy, not that I expect it, since I've been suffering your insufferable ignorance for a long time.

Ignorance kills people; since 2013, again, about 50 million of them, all dead because we didn't commit to going nuclear, long ago, as Hansen suggested we should have done long ago, by appeal to historical data.

If I sound angry, it's because I am.

Have a pleasant Sunday.

May 1, 2021

Magnetic Separation Method for Isolating Rare-Earth Elements and Zirconium from Molten Salts

The paper to which I'll briefly refer is this one, published by Russian scientists in an English edition: Alekseeva, L.S., Savinykh, D.O., Orlova, A.I. et al. Magnetic Separation Method for Isolating Rare-Earth Elements and Zirconium from Molten Salts. Inorg Mater 56, 583–590 (2020).

Strontium ferrites are magnetic particles that find utility in many electronic storage devices. This paper uses magnetic particles that are also strontium based, particles which apparently have a Curie Point higher than the temperature of molten salts.

From the introduction:

Pyroelectrochemical technology utilizing molten salts, primarily molten alkali chlorides, as solvents is commonly thought to have considerable potential. Pyrochemical processes are characterized by small volumes of waste, its high specific activity, and the near absence of liquid technological high-level waste. This technology has a number of advantages: melts are good solvents, they are stable at high temperatures and under irradiation, and the use of melts allows one to run the process without water, necessary for all living systems. Work with chloride melts, with application in pyroelectrochemical technology, has been and is being carried out in many countries all over the world: in Russia [1–7], the United States, Japan [8], Korea [9], and the United Kingdom [3, 10] (zeolite sorption). The behavior of U, Np, Pu, and Am in LiCl–KCl melts has been the subject of extensive studies [1, 11–25]. The diversity of studies, in particular, in recent years [20–23], indicates that this is a topical issue.

Extraction of actinides (Am and Cm), rare-earth elements (REEs), and zirconium from melts is one of the key issues in this technology, in particular in Russia [5, 6, 24, 25].

Extraction, preconcentration, and consolidation methods under development for subsequent isolation of actinide fission products from the biosphere include precipitation, ion exchange, zone crystallization, and sorption.

In this work, we propose using a magnetic field for extracting components dissolved in a LiCl–KCl melt.

Actinide- and rare-earth-containing strontium hexaferrite with the magnetoplumbite structure and zirconium-containing spinel ferrites are of interest as magnetic carriers capable of absorbing melt components to be extracted.


The authors synthesized spinel magnetic ferrite particles isomorphous with a magnetic mineral containing lead, magnetoplumbite. They noted that many of the compounds in this class of isomorphs can incorporate lanthanide elements (aka "rare earths.&quot

However their spinel structure was a compound, again isomorphous, of strontium, copper, iron and oxygen.

They heated a molten salt composed of potassium and lithium chloride at 450C and were able to extract neodymium and zirconium.

It was hardly quantitative according to a table in the paper:



There are lots of people who are molten salt reactor kind of guys and gals; I used to be one myself, but I changed my mind. However, in recent years I have begun to rethink molten salts in all kinds of chemical separations processes, and redox processes, including but not limited to separation and recovery of valuable materials in used nuclear fuels. (It should be noted that there are an infinite number of molten salts, given the existence and wide study of "ionic liquids" some of which are partially or wholly organic molecules.

I'm not sure this work has much practical import, although one can imagine these things in certain kinds of molecular filters, but any magnetic material with a curie point high enough to be stable in a molten alkali metal salt is interesting. For the record, I do not favor the use of potassium salts in nuclear fuel settings since the separation from cesium is problematic. Rubidium, which is present mostly in a form (Rb-87) that occurs naturally all over the planet as a fission product is a better choice.

I trust you're having a nice Saturday.
May 1, 2021

Thermoelectric Power Generation and Water Stress in India: A Spatial and Temporal Analysis

The paper I'll discuss in this brief post is this one: Thermoelectric Power Generation and Water Stress in India: A Spatial and Temporal Analysis (Chao Zhang, Joonseok Yang, Johannes Urpelainen, Puneet Chitkara, Jiayi Zhang, and Jiao Wang Environmental Science & Technology 2021 55 (8), 4314-4323).

The connection between water and energy, particularly in an electrified world is inexorable.

It is worth noting (or confessing?) that the form of electrical energy (and other) energy that I regard as sustainable and environmentally acceptable, nuclear energy, is the worst offender for the use of water for power generation, at least as presently practiced. I am convinced that this problem can be engineered away, but it has not, as of now, been engineered away.

The paper under discussion is an overview of the water situation in India, a developing nation now undergoing a terrible Covid tragedy for which all decent human beings must extend sympathy and where possible aid. This said, the long term, environmental tragedy of India may make Covid pale in the end. For this reason, this paper, a policy paper, is worth some consideration.

From the introduction:

Water and energy are inextricably linked since most types of power generating processes, and thermal power in particular, require a significant amount of water, especially for steam generation, ash-handling, and cooling. Similarly, energy is required for all phases of water processing, including extraction, treatment, transportation, and consumption.(1?3) However, climate change and the sharp increase in global water demand pose serious threats to the sustainable supply of these critical resources. This results in a high degree of interdependency in a self-reinforcing cycle: water stress can be intensified by an increase in electricity production,(4?6) and at the same time, water scarcity can hamper the secure and sustainable supply of energy.(7,8)
Among different sources of energy, thermoelectric power is of central importance in this regard. Thermoelectric power generation constitutes 75% of the world’s net electricity generation(9) and has been the largest source of water withdrawal (86%) in the energy sector.(10) In recent years, a growing number of studies have investigated the relationship between thermal power generation and water scarcity in different countries and regions, including the United States,(11,12) the United Kingdom,(13) Middle East and North Africa,(14) South Africa,(15) and developing countries in Asia.(16)

Particular attention has been paid to China, where the thermal power industry has grown rapidly and water scarcity is a significant environmental challenge.(17?19) Existing studies have investigated various aspects of the electricity-water nexus in China, such as estimating aggregate water use by thermal power generation based on top-down data and methodologies (e.g., refs (20and21)); compiling spatial explicit water use inventories and water stress maps through bottom-up data at the power plant or generating unit-level (e.g., refs (22?25)); accounting for virtual water transfers via inter-regional or interprovincial electricity transmission (e.g., refs (26?29)); and projecting future trajectories of water use by using scenario analysis tools or energy system modeling (e.g., refs (30?32)). These analyses have revealed numerous spatial and temporal characteristics of the electricity-water nexus in China. Of note are the impacts in the water sector due to technological changes and policy interventions in the power sector. For example, historical analysis that is based on a detailed unit-level water use inventory conducted by ref (23) shows that thermal power generation in catchments under high water stress (with a water withdrawal-to-availability ratio greater than 0.4) has grown by a factor of 4.5 from 2000 to 2015 (from 643 TWh in 2000 to 2,885 TWh in 2015). At the same time, significant policy-induced technological changes, especially those in cooling technology structures, have successfully decoupled thermoelectric freshwater withdrawal from growing power generation.

Thermal power generation, especially coal-fired power, also plays a major role in India’s electricity mix. Driven by a growing population and economic development, India’s generation capacity is being rapidly expanded. Projections estimate that energy demand in India will increase by 1005 Mtoe from 2016 to 2040, more than any other country during the same period.(33) India is also projected to have the largest population increase worldwide, and will surpass China as the world’s most populous country in 2027...(34)


The paper notes that water scarcity is a huge problem in India as well as in China, and in fact, pretty much everywhere in the world. The failure to make even a weak stab at addressing climate change - all of our fantasies about electric cars, vast wind industrial parks in former wildernesses, and electric generating glass coated with millions of tons of mined, carbon reduced, and often toxic elements of the periodic table have done nothing at all except generate more fantasies - will make this worse.

The story can be told in graphics and tables from the paper:



The caption:

Figure 1. Spatial Distribution of Thermal Power Plants and Baseline Water Stress at Catchment Level. State boundaries are illustrated by light blue lines. Abbreviations of state name: AP, Andhra Pradesh; AR, Arunachal Pradesh; AS, Assam; BR, Bihar; CT, Chhattisgarh; GA, Goa; GJ, Gujarat; HR, Haryana; HP, Himachal Pradesh; JK, Jammu and Kashmir; JH, Jharkhand; KA, Karnataka; KL, Kerala; MP, Madhya Pradesh; MH, Maharashtra; MN, Manipur; ML, Meghalaya; MZ, Mizoram; NL, Nagaland; OR, Odisha; PB, Punjab; RJ, Rajasthan; TN, Tamil Nadu; TR, Tripura; UP, Uttar Pradesh; UT, Uttaranchal; WB, West Bengal. The base map is obtained from WRI’s Aqueduct Water Risk Atlas which is drawn according to defacto status and shows actual control situation. Disputed territories are supplemented and illustrated with shade.




The caption:

Figure 2. Electricity generation and freshwater withdrawal classified by baseline water stress categories.




The caption:

Figure 3. Relative Changes of Power Generation and Water Use. The base year level in 2009 is assigned as 1.0.




The caption:

Figure 4. Thermoelectric Freshwater Withdrawal (a) and Consumption (b) by the top 10 States in 2009, 2018, and 2027




The caption:

Figure 5. Spatial distribution of freshwater withdrawal, freshwater consumption, and thermoelectric water stress in 2018, and changes during the period of 2009–2018 and 2018–2027. The base map is obtained from WRI’s Aqueduct Water Risk Atlas which is drawn according to defacto status and shows actual control situation. Disputed territories are supplemented and illustrated with shade.


Some tables from the text:



Some additional text, which is involved with something called "reality:"

India’s thermal power sector has rapidly expanded in the past decade (2009–2018) with total generation capacity, including coal-fired, natural gas-fired, and nuclear power, increasing by 130% (from 101 GW in 2009 to 232 GW in 2018). This expansion was mainly driven by the commissioning of new coal-fired power plants, whose share in the total power capacity increased from 79% (80 GW) in 2009 to 86% (201 GW) in 2018. Natural gas-fired and nuclear power capacity accounted for the other 11% and 3% in 2018, respectively.

Net electricity generation increased by around 80% from 615 TWh in 2009 to 1106 TWh in 2018. The role of coal-fired power generation grew during the past decade, and it contributed to 92% (1017 TWh) of total thermal power generation in 2018, compared to 83% (615 TWh) in 2009. The lack of natural gas resources in India has impeded the development of natural gas-fired power generation. Although the installed capacity of natural gas-fired power plants rose during the study period, electricity generation decreased 43% from 88 TWh to 50 TWh. Nuclear power contributed to around 3.5% of overall generation.


It is often represented in our modern fantasies that "coal is dead" coupled with the stupid remark that the reason "coal is dead" is because of so called "renewable energy." It is telling however, and a dose of reality in the face a delusion every bit as serious - in some ways worse - that Covid is a "hoax," that coal is not dead where there is no ready access to dangerous natural gas, the use of which will scar the planet's atmosphere, water and land for eternity, all for short term comfort. This fact is highlighted in text I have bolded in this text just above. So called "renewable energy" did not save the world, is not saving the world, and will not save the world. It is a trivial form of energy, completely dependent on access and use of dangerous natural gas, and all the denial in the world will not change this fact.

Facts matter.

Nevertheless, there is discussion in the text of solutions with an eye of chasing the chimera that so called "renewable energy" is sustainable, even though one of the features of escaping poverty - poverty being an issue in India - is the abandonment of trying to survive on so called "renewable energy." This de facto discussion is here:

SWITCH is a long-term capacity expansion planning model which minimizes the cost of power system expansion and operations subject to reliability, operational, and resource availability constraints. It also helps model large, real-world power systems with high temporal and spatial resolutions relatively quickly. This is useful because including renewable energy assets in power systems requires an accurate representation of grid operations, through high temporal and special resolutions.


Whatever. The italics are mine.

India has huge thorium resources, and it has now supplanted Canada in the use of CANDU type heavy water nuclear reactors. It has also piloted a plant for the extraction of uranium from seawater.

Ironically enough - I've discussed this topic many times elsewhere - the solution to water demands for power plant cooling is ironically enough, to raise the internal temperatures so as to drive heat networks that extract additional work (exergy) from that heat, i.e. raise the efficiency of the plants to unprecedented methods. One possible inclusion in a heat network are systems that desalinate sea water. Desalination is not a panacea, and many environmental issues need to be addressed to employ it, but I believe they can be addressed by an educated, dedicated and respectful version of humanity, one not here immediately, but one that is certainly possible.

We are at a cross roads.

The functional "renewables will save us," anti-nuke guru who resides in a super energy efficient aerie on a mountain in the wealthy enclave of Snowmass, Colorado - which is represented in congress by Lorena Bobbitt, delusional gun nut Trump humper - is Amory Lovins. Amory Lovins will tell you that the goal of energy efficiency is to reduce the use of energy. Of course, Amory Lovins couldn't care less about poor people.

Maybe delusion is contagious in Snowmass.

I am disinterested in delusional bourgeois brats. Unlike the guru in Snowmass I am familiar with Jevon's paradox, well known and often observed since the 19th century, which states that increases in energy efficiency increase the use of energy. I'm fine with that because I believe that sustainable energy is possible. For me increases in efficiency are all about making energy available to the billions of people who lack it, a significant portion of whom live in India.

Have a nice weekend.

April 30, 2021

Global groundwater wells at risk of running dry

The paper I'll discuss in this post is this one: Global groundwater wells at risk of running dry (BY Scott Jasechko and Debra Perrone Science 372, 418–421 (2021) 23 April 2021)

A news item referring to the paper in the same issue of Science which is likely to be open sourced is here: The hidden crisis beneath our feet (James S. Famiglietti1,2, Grant Ferguson, Science 23 Apr 2021: Vol. 372, Issue 6540, pp. 344-345)

If memory serves me well, I have referred a number of times to a lecture I attended by Dr. Robert Kopp of Rutgers University, this one: Science on Saturday: Managing Coastal Risk in an Age of Sea-level Rise, wherein, in answer to a question from the audience (mine), Dr. Kopp explained that about 10% of sea level rise can be attributed to water pumped out of the ground and allowed to evaporate. Most of this water is used for agriculture, but it is also used in cities and in homes. I am guilty here. The water in my home is well water, which as I happen to know, contains some interesting and unpleasant substances.

This suggests - I've been musing about this a lot lately - that one path to addressing climate change would be restoring shorelines would include refilling the aquifers we've drained in the last 50 years or so in yet another expression of our accelerating contempt for all future generations, along with dried up lakes like Owens Lake in California, and dying bodies of water like Lake Baikal, the Aral Sea, the Caspian Sea, etc...

My most recent episode of such musings out loud is here: Ion-capture electrodialysis using multifunctional adsorptive membranes, which is preceded by a long dull digression on reactor physics and the mercury content of the oceans.

To anyone who argues that refilling ground water by desalination of ocean water, refilling dead lakes, dead seas etc. is geoengineering, let me say this.

There is a television show on the Science channel called Engineering Catastrophes which I watch from time to time to give me ammunition to make fun of my son the engineer. It's all about failures of bridges, buildings, roadways, etc. from things like hydrogen embrittlement, improper analysis of stresses, poor materials etc. However everything on this show is small potatoes compared to the biggest engineering catastrophe, the existing haphazard, uncontrolled and unplanned geoengineering of the entire planet, most of which has been a catastrophe of the last one hundred years.

This is not really a water problem; it is an energy problem. No, wind turbines in lobster fisheries off the coast of Maine, nor any other wind schemes designed to industrialize wilderness, will not produce even a fraction of the energy required owing to thermodynamic, mass intensity, reliability and lifetime considerations. The sooner we lose our affection for this nonsense, the more hope we will have.

The goal of geoengineering should be to restore, and nothing more.

Anyway future generations are in a world of hurt according to this paper. From the paper's introduction:

Groundwater provides nearly half of the water used for irrigated agriculture and supplies drinking water for billions of people (1, 2). Its perennial and distributed nature makes groundwater suitable for providing reliable access to potable water and supporting efforts to adapt to climate change (3). Nevertheless, access to groundwater to meet water demands may be hindered when groundwater levels decline. Groundwater levels are declining in many aquifers around the globe [e.g., (4–6)], causing groundwater wells (7) to run dry (8–10) and leaving farmers unable to irrigate and households without access to reliable and convenient drinking water.

When groundwater wells run dry, one common adaptation strategy is to construct deeper wells (11). Deeper wells tend to be less vulnerable than shallower wells to climate variability (12) and groundwater level declines (13), but even relatively deep wells are not immune to long-term reductions in groundwater storage. Despite the role of wells as a basic infrastructure used to access groundwater, information about the locations and depths of wells has never before been compiled and analyzed at the global scale. Analyzing groundwater resources at the global scale is becoming increasingly important because of groundwater’s role in virtual water trade, international policy, and sustainable development (14, 15). Nevertheless, we emphasize that groundwater is a local resource influenced by hydrogeologic conditions, water policies, sociocultural preferences, and economic drivers.

Here, we compiled ~39 million records of groundwater well locations, depths, purposes, and construction dates (supplementary materials section S1), which provide local information at the global scale. These observations promote a better understanding of spatiotemporal patterns of well locations and depths (16, 17). Groundwater data are notoriously difficult to collect and collate (18, 19). The ~39 million wells are situated in 40 countries or territories that represent ~40% of global ice-free lands (average data density of 0.7 wells/km2; Fig. 1). Half of all global groundwater pumping takes place within our study countries or territories [fig. S2, reference (2), and supplementary materials section S5.1], which are home to >3 billion people (table S1).


Fig. 1 Groundwater well depths in countries around the globe.



The caption:


The dataset contains ~39 million wells in 40 countries or territories (see inset map). Blue points mark shallower wells and red points mark deeper wells in Canada, the United States (white areas mark states where data are unavailable), and Mexico (A); Argentina (partial coverage), Bolivia (partial coverage), Brazil, Chile, and Uruguay (B); Iceland, Portugal, Spain, France, Germany (partial coverage), Denmark, Sweden, Norway, Italy, Slovakia, Slovenia, Belgium, Poland, Latvia, Estonia, Czech Republic, Hungary, United Kingdom, and Ireland (C); Namibia and South Africa (D); Thailand and Cambodia (E); and Australia (F). National-level analyses are available in figs. S37 to S435.


Fig. 2 Fraction of areas with deeper well drilling versus shallower well drilling for five time periods among study countries.



The caption:


Well-deepening trends are more common than well-shallowing trends for the majority of time periods for the majority of countries (national-level analyses in figs. S37 to S435). (A) The fraction of 100 km by 100 km study areas where wells are being drilled deeper over time (Spearman ? > 0; regression of well completion depth versus well construction date). Each diamond represents well construction depth variations over time for a given country (e.g., blue diamonds are regressions for years 2000–2015; see legend above figure); some points overlap. (B to L) Spearman ? determined by regression of well depth versus construction date for the time interval 2000–2015 in 11 countries [i.e., the fraction of all 100 km by 100 km areas in these maps with Spearman ? > 0 correspond to the blue diamonds in (A)]. Blue shades mark 100 km by 100 km areas where wells are being constructed shallower over time (i.e., Spearman ? < ?0.1); orange and red shades mark 100 km by 100 km areas where wells are being constructed deeper over time (i.e., Spearman ? > 0.1.)


Fig. 3 Well construction depth trends at locations close to ( 0) significantly (Spearman P < 0.05) over time.



The caption:

Results derive disproportionately from Australia, Canada, France, Mexico, New Zealand, Sweden, and the United States (tables S64 to S69). The histograms are distributions of Spearman rank coefficients as determined through regression analyses of well completion depths versus well completion dates at sites close to a monitoring well where water levels have declined significantly over time (Spearman ? > 0 and Spearman P < 0.05 determined through regression analyses of measured depth to water in monitoring well versus water level measurement date). Each plot in (A) to (E) represents one study time interval [e.g., (A) presents results for 1950–1975]. Although all distributions skew right (i.e., well deepening where groundwater levels are declining), a substantial number of sites indicate that newer wells are not being constructed substantially deeper than older wells where groundwater levels are declining. Curly brackets above each plot in (A) to (E) provide the percentage of well construction depth trends falling within the following Spearman ? ranges: ?1.0 to ?0.2, ?0.2 to 0.2, and 0.2 to 1.0.


An excerpt of the conclusion of this short paper:

From India to the United States, wells are already running dry because of groundwater level declines (22, 26, 34, 35). When a well runs dry, there are a number of ways to adapt: (i) construct a new and deeper well or deepen the existing well that ran dry (10, 13); (ii) access alternative water sources (10, 29); (iii) conserve water through behavioral changes or technological adoptions [e.g., switching to crops that are less water intensive; adopting irrigation approaches that reduce water losses (25)]; (iv) sell the property (10); or (v) abandon activities that require high water inputs, such as irrigation (35). In Karnataka, India, some households that have had their well run dry have offset lost agricultural profits by seeking off-farm income (35). The transition away from agriculture can have longer-term impacts on the household, such as debt accrual while seeking alternative employment and lost educational opportunities for children who join the workforce early to supplement household income (35). In California’s Central Valley (United States) and several other agricultural hubs around the globe, typical agricultural wells are deeper than domestic wells (supplementary materials sections S5.5 to S5.6); as a result, rural domestic wells are running dry disproportionately relative to agricultural wells...

...Groundwater wells supply water to billions of people around the globe (2, 13, 14, 36). Groundwater depletion is projected to continue in some areas where it is already occurring, and even expand to new areas not yet experiencing depletion (37)...

...Our work highlights the vulnerability of existing wells to groundwater depletion because (i) many wells are not much deeper than the local water table, making them likely to run dry with even modest groundwater level declines (supplementary materials section S6), and (ii) deeper well construction is common but not ubiquitous where groundwater levels are declining...


History will not forgive us, nor should it.

Have a nice weekend.
April 28, 2021

A day in the life.

April 28, 2021

The rate of denovo genetic mutations in the children of parents exposed to Chernobyl radiation.

The paper I'll discuss in this post is this one: Lack of transgenerational effects of ionizing radiation exposure from the Chernobyl accident M. Yeager et al., Science 10.1126/science.abg2365 (2021).

This week marks the 34th anniversary of the steam explosion at the Chernobyl Nuclear Power Plant. The graphite moderator in plant burned for weeks and much of the inventory of volatile fission products was released into the environment, and detected in easily measurable amounts all across Europe, as far away as Scotland.

This event garnered huge attention at the time and still garners significant attention.

It is reported that about six to seven million people die each year from air pollution, meaning that about 200-250 million people have died from air pollution since the explosion of the nuclear reactor, proving in the minds of many that nuclear power is "too dangerous," while apparently the combustive use of fossil fuels and biomass, responsible for these 200-250 million deaths, is not "too dangerous." Neither is climate change "too dangerous." These facts, deaths from air pollution, and the acceleration of climate change despite endless talk about and expenditures on so called "renewable energy," unlike Chernobyl do not garner huge attention.

Go figure.

From the introduction to the paper:

Nearly all inherited genetic variation is present in the germline DNA of at least one parent; however, a small number of transmitted variants are unique, having arisen due to random mutations in gametes (sperm and oocytes), and are known as de novo mutations (DNMs). DNMs are critical building blocks of evolution and the only class of genomic variation that has not undergone extensive evolutionary purifying selection (purging of highly deleterious but non-lethal variants), making DNMs a unique form of inherited variation different from the genetic variation investigated in mapping complex traits and diseases (1). DNMs have been of intense interest because of their role in human disease, particularly neurodevelopmental disorders (2, 3).

Only recently has it been feasible to comprehensively investigate DNMs genome-wide at the population level in humans by whole-genome sequencing (WGS) of mother/father/child trios. Recent reports of human DNMs characterized by WGS of trios estimate between 50 and 100 new mutations arise per individual per generation (2, 4–8), consistent with the population genetic estimate that the human mutation rate for single-nucleotide variants (SNVs) is approximately 1×10?8 per site per generation (9, 10). The strongest predictor of DNMs per individual is paternal age at conception (2–6, 8) with an increase of 0.64-1.51 per one-year increase in paternal age (6, 8, 11) while a maternal effect of approximately 0.35 per one-year increase in age was observed (6, 8, 12). Transgenerational studies of radiation exposure have primarily focused on disease (cancer, reproductive, and developmental) outcomes and have reported inconclusive results (13, 14).

Exposure to ionizing radiation is known to increase DNA mutagenesis above background rates (15, 16). Animal and cellular studies suggest high doses of ionizing radiation can lead to DNMs in offspring, particularly through double-stranded breaks (13, 17). Human studies have sought a biomarker of prior radiation injury (13, 18, 19), but have examined a small number of minisatellites and microsatellites, yielding inconclusive results (20–23). A WGS study of three trios from survivors of the atomic bomb in Nagasaki did not reveal a high load of DNMs (20), while a single-nucleotide polymorphism (SNP) array study of 12 families exposed to low doses of Caesium-137 from the Goiania accident in Brazil reported an increase in large de novo copy-number variants (24). No large comprehensive effort has explored DNMs genome-wide in children born from parents exposed to moderately high amounts of ionizing radiation yet possible genetic effects have remained a concern for radiation-exposed populations, such as the Fukushima evacuees (25).

Herein, we examine whether rates of germline DNMs were elevated in children born to parents exposed to ionizing radiation from the 1986 Chernobyl (Chornobyl in Ukrainian) disaster, where levels of exposure have been rigorously reconstructed and well-documented (26). Our study focused on children born of enlisted cleanup workers (“liquidators”) and evacuees from the town of Pripyat or other settlements within the 70-km zone around Chernobyl Nuclear Power Plant in Ukraine (27) after the meltdown, some of whom had extremely high levels of radiation exposure and several of whom experienced acute radiation syndrome. We performed Illumina paired-end WGS (average coverage 80X), SNP microarray analysis, and relative telomere length assessment on available samples from 130 children from 105 mother-father pairs. The parents had varying combinations of elevated gonadal ionizing radiation exposure from the accident (tables S1 to S3), and included a combination of exposed fathers, exposed mothers, both parents exposed and neither parent exposed (27)...


Some figures from the paper:



The caption:

Fig. 1 Detected DNMs per genome based on distributions of parental age at conception.
Analyses are presented by increasing paternal and maternal age at conception, paternal and maternal dose, birth year of child, and paternal and maternal smoking at conception. All plots are univariate and do not account for other potentially correlated variables (for example, maternal age does not account for high correlation with paternal age).




The caption:

Fig. 2 Distribution of de novo SNVs by type of nucleotide change across six studies.
n = number of children sequenced (adapted from (39)). Liftover was used to convert coordinates to hg38 for all studies and the reference for CpG sites were defined with respect to that reference sequence. Only autosomes were included. Error bars show binomial 95% confidence intervals. Studies included Kong (2); Wong (8); Francioli (4); Michaelson (3); Jónsson (6); and Chernobyl (present study).


Some additional text:

Although it is reassuring that no transgenerational effects of ionizing radiation were observed in adult children of Chernobyl cleanup workers and evacuees in the current study, additional investigation is needed to address the effects of acute high-dose parental gonadal exposure closer to conception. The upper 95% confidence bound suggests the largest effect consistent with our data is less than 1 DNM per 100 mGy from paternal or maternal exposure (Table 3 and tables S8 and S9). Previously, Dubrova et al. (22, 29) reported a two-fold increase in mini-satellite mutations in children born to parents living in a highly exposed region of Belarus. Weinberg et al. (34) reported an increase in the mutation rate at microsatellite loci among children born to cleanup workers. Subsequent small studies have not reported an increased mini-satellite or microsatellite mutation rate in children of cleanup workers, including those with low doses (0.09-0.23 Gy) (21, 30, 35) or in children of the atomic-bomb survivors of Hiroshima or Nagasaki (31).

Our study evaluated peripheral blood from adult children conceived months or years after the Chernobyl accident, which limited the ability to assess exposure closer to conception; however, there was no evidence of notable differences in DNMs in children born the following year (1987). Since these families were recruited several decades after the accident, we acknowledge potential survivor bias among sampled children, although this is unlikely since there is no consistent demonstration in humans of sustained clinical effects of preconception ionizing radiation exposure (36)...

....Although it is reassuring that no transgenerational effects of ionizing radiation were observed in adult children of Chernobyl cleanup workers and evacuees in the current study, additional investigation is needed to address the effects of acute high-dose parental gonadal exposure closer to conception. The upper 95% confidence bound suggests the largest effect consistent with our data is 1 Gy), but lower maternal dose was not associated with elevated DNMs, consistent with animal studies (13). Furthermore, our analysis of 130 adult children from 105 couples using 80X coverage of short-read technology suggests that if such effects on human germline DNA occur, they are uncommon or of small magnitude. This is one of the first studies to systematically evaluate alterations in human mutation rates in response to a man-made disaster, such as accidental radiation exposure. Investigation of trios drawn from survivors of the Hiroshima atomic bomb could shed further light on this public health question


For the record, I am an older father, and therefore there is a higher probability that my sons have DNM's than there may be to the children of men who were younger than I was before coming a father. If so these were happy mutations, since my sons are better men than I was at their age. I love my mutant boys.

Have a nice day tomorrow.

Profile Information

Gender: Male
Current location: New Jersey
Member since: 2002
Number of posts: 33,516
Latest Discussions»NNadir's Journal