Earth Wise

A look at our changing environment.

You are here: Home / Archives for machine learning

machine learning

Seaweed and concrete

By Leave a Comment

Modern civilization is pretty much made of concrete.  People use more concrete than any other substance apart from water.  But concrete is made from cement, and cement is the source of 10% of all carbon dioxide emissions worldwide.

Researchers at the University of Washington and Microsoft have developed a new kind of concrete made by mixing dried, powdered seaweed with cement.  By fortifying cement with seaweed, the global warming potential of the concrete is reduced by 21% without weakening it. 

This novel recipe for concrete was developed using machine learning models, arriving upon it in a fraction of the time it would have taken by traditional experimentation.

Producing cement leads to carbon emissions from the fossil fuels used to heat raw materials and from a chemical reaction called calcination that occurs during the production process.  Seaweed is a carbon sink that pulls carbon dioxide out of the air and stores it while it grows.  By replacing some of cement in concrete, the resultant product has a much smaller carbon footprint.

Machine learning was used to predict the ideal mixture of cement and seaweed to yield concrete with a reduced carbon footprint that still passed mechanical strength tests.  Finding the right mixture would have taken 5 years ordinarily, but the machine learning process took only 28 days.

The researchers plan to generalize their work to different kinds of algae and even to food waste or other natural materials in order to create local, sustainable cement alternatives around the world. 

**********

Web Links

Seaweed-infused cement could cut concrete’s carbon footprint

Photo, posted June 29, 2009, courtesy of Peter Castleton via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

AI and greener cement

By Leave a Comment

Cement pretty much holds the modern world together.  The amount of cement required to create our infrastructure is almost incomprehensible.   By weight, humanity consumes more cement than food, about 3 pounds per person per day.  The cement industry produces around eight percent of global CO2 emissions, which is more than the aviation industry.  So, if the amount of emissions produced making concrete could be reduced by even a few percent, it would make a significant impact.

Cement plants utilize rotary kilns heated to 2,500 degrees Fahrenheit to burn ground limestone down to a substance called clinker.  That energy-intensive combustion process emits large amounts of carbon dioxide.  However, the combustion process accounts for much less than half of the emissions associated with making concrete.  The majority comes from the raw materials needed to produce clinker.

One strategy to reduce concrete emissions is to modify the cement recipe itself, replacing some of the clinker with alternative materials.  Some producers already make use of materials like slag from iron production and fly ash from coal-fired power plants.   

A team of researchers at the Paul Scherrer Institute in Switzerland is making use of machine learning to simulate and optimize cement formulations that would emit significantly less CO2 while maintaining the same high level of mechanical performance.  This AI-based approach eliminates time-consuming experiments and conventional complex simulations.

The Scherrer Institute seeks to discover new materials and the effort has already yielded some promising candidates.  The next steps will be testing some of these recipes in the laboratory. 

**********

Web Links

AI paves the way towards green cement

Photo, posted July 3, 2007, courtesy of Tim Shortt via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

Fighting honey fraud

By Leave a Comment

Honey fraud is a significant issue for the food industry.  What is honey fraud?  Typically, it involves mislabeling where honey was produced or what types of flowers the bees collected nectar from.  Honey made from a single type of flower is often more expensive because of the unique flavor it provides or from potential health benefits.  Sometimes even cheap alternatives like sugar syrups are labelled as honey.  It turns out that honey is one of the most fraud-prone commodities in global trade, with fraud estimated to occur in up to 10% of the honey traded internationally.  Honey from some countries, such as China and India, has had 30% or more of samples found to be fraudulent.

Researchers at McGill University in Montreal have developed an AI-powered method to verify the origin of honey thereby ensuring that what is on the label corresponds to what is in the jar.

The McGill method can determine what kind of flowers the bees visited to produce a particular sample of honey.  Previous honey authentication involved pollen analysis, which is ineffective for honey that was processed or filtered.  The new method uses high-resolution mass spectrometry which captures a unique chemical “fingerprint” from the honey.  Machine learning algorithms read the fingerprint to identify the honey’s origin.

The researchers tested their methodology on a variety of honey samples which they then compared with honey from known botanical sources.  Using previous methods for honey authentication can take days.  The McGill method takes only minutes and works regardless of how the honey was processed.

According to the researchers, people deserve to know that the honey they buy is what it claims to be, and honest honey producers deserve protection from fraudulent competitors.

**********

Web Links

Fighting honey fraud with AI technology

Photo, posted May 6, 2012, courtesy of Emma Jane via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

Lithium in Arkansas

By Leave a Comment

Lithium is the critical raw material in the batteries that power electric cars as well as cell phones, computers, and other gadgets.  The stuff has been nicknamed “white gold” for good reason.  Chile and Australia are the world’s largest producers of the metal, which is mostly extracted from brine in evaporation ponds.  The majority of it is then processed in China.  The energy industry has been increasingly working to produce the raw materials needed to produce lithium-ion batteries in the United States and process those materials domestically.  There are multiple projects at various stages across the country.

Researchers at the US Geological Survey and the Arkansas state government recently announced that they have discovered a vast trove of lithium in an underground brine reservoir in Arkansas.

With a combination of water testing and machine learning, the researchers determined that there could be 5 to as much as 19 million tons of lithium in the geological area called the Smackover Formation.  This is more than enough to meet all the world’s demand for it.

Several companies – including Exxon Mobil, which is covering its bets on the future of oil as an energy source – are developing projects in Arkansas to produce lithium.  If these companies can develop and scale up economical new ways to extract lithium from salty water, the region in Arkansas could become the lithium capital of the world.

Energy and mining companies have produced oil, gas, and other natural resources in the Smackover Formation, which extends from Texas to Florida.  The same brines have long been the source of other valuable substances. 

**********

Web Links

Arkansas May Have Vast Lithium Reserves, Researchers Say

Photo, posted May 22, 2020, courtesy of the European Space Agency via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

Elephants have names

By Leave a Comment

Elephants are considered to be very smart creatures.  They have the largest brain of any land animal and have three times as many neurons as humans.  Of course, a lot of that brain hardware is needed to control those big elephant bodies, but elephants have also demonstrated impressive mental capabilities on many occasions.

A new study by an international team headed by researchers from Colorado State University has found that elephants call each other by name and respond when they hear others call their name.

The researchers analyzed hundreds of elephant calls recorded over more than a year in Kenya.  With machine learning techniques, they were able to identify specific sounds that elephants made when calling each other.  When the researchers played recordings of these sounds, the elephants responded to the sound of their own name by calling back or moving toward the loudspeaker.  Their response to other names was much less enthusiastic.

Very few animals use names at all, and, in most cases, it is more a matter of imitation.  For example, dolphins refer to other dolphins by mimicking their unique sounds.

These findings at least suggest that elephants may be capable of abstract thought and perhaps may have a vocabulary beyond names for each other.  The researchers are investigating whether elephants can identify food, water, or locations using their calls.  The great hope is that it might ultimately be possible to communicate with elephants directly.  It would be wonderful to be able to alert them to poachers or other threats.  The lead author of the study said that he would like to be able to tell elephants things like “Do not come here.  You’re going to be killed if you come here.”

**********

Web Links

Elephants Have Names for Each Other, Study Finds

Photo, posted April 24, 2016, courtesy of Neil Ransom via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

Fireflies are in decline

By 2 Comments

If you are seeing fewer fireflies each year, you’re not alone.  Like many insects, firefly populations are in decline.  A new study by researchers from the University of Kentucky, Bucknell University, Penn State University, and the USDA has shed some light on the precarious situation facing firefly populations across North America. 

The research team used a mix of field surveys from citizen scientists and advanced machine learning techniques to analyze more than 24,000 surveys from the Firefly Watch citizen science initiative.  The study, which was recently published in the journal Science of the Total Environment, identified the factors likely responsible for the declines in firefly populations. 

The research team found that fireflies are sensitive to various environmental factors, from short-term weather conditions to longer climatic trends.  Fireflies thrive in temperate and tropical climates.  As global temperatures rise, these conditions become less predictable and less hospitable.

Light pollution is another threat to fireflies.  Artificial light at night – from things like street lights and billboards – is particularly disruptive to fireflies as it interferes with their bioluminescent communication essential for mating.

Urban growth, including buildings, roads, and sidewalks, poses another significant threat to fireflies by overtaking their natural habitats and decreasing available breeding areas. 

Additionally, certain agricultural practices seem to contribute to the decline of fireflies. 

According to the research team, reducing light pollution, preserving natural habitats, and implementing wildlife-friendly agricultural practices are conservation measures that could help mitigate the decline of fireflies. 

**********

Web Links

Fading lights: Comprehensive study unveils multiple threats to North America’s firefly populations

Photo, posted July 12, 2021, courtesy of Bruce Hallman/USFWS via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

Wind farms and land use

By Leave a Comment

Wind power has become one of the most affordable ways to generate electricity as well as being renewable and not contributing to global warming.  But there has been the perception that wind farms require a lot of land compared with fossil fuel power plants. This issue of land use has made decision-makers reluctant to invest in wind energy in many instances.

A new study by McGill University in Montreal looked at this issue and has found that land use for wind power is far more efficient than previously thought.  The study combined data from geographic information systems with machine learning models to assess land usage associated with nearly 16,000 wind turbines in the western U.S.

The study assessed the land use of 320 wind farms.  They found that wind power infrastructure (mostly the turbines themselves and the roads that lead to them) typically only uses 5% of the land area of a wind farm.  If the wind farm is sited in locations with existing infrastructure, such as on agricultural land, then it can be as much as seven times more land-efficient – meaning the amount energy produced in a given area of land impacted by the infrastructure – than a wind farm that is developed on unused land.

Previous studies of wind farm land usage assumed that all of the land where the wind farm was located was devoted to energy generation.  In reality, most of land in question is often used for other purposes, such as agriculture.

The methods developed by the researchers are potentially useable for future assessments of various energy technologies in terms of their environmental sustainability.

**********

Web Links

Clearing the air: wind farms more land efficient than previously thought

Photo, posted September 29, 2009, courtesy of Tim Green via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

Major species turnover forecasted for North American cities

By Leave a Comment

Climate change affects animal species in many ways.  It induces habitat loss, disrupts migration and breeding patterns, threatens marine life, and facilitates an increased spread of disease.  It may also affect where animals can be found in the future. 

According to a new study led by researchers from the University of Toronto Mississauga and Apex Resource Management Solutions in Canada, climate change may dramatically affect the animal species observed in North American cities by the end of the century.

The researchers used species distribution data combined with machine learning to study the impact of human-caused climate change on more than 2,000 animal species historically found in the 60 most populous North American cities. 

According to the research team, changes in biodiversity are brewing for almost every city it studied by the year 2100.  In fact, cities with a rich history of biodiversity are predicted to have the largest declines and fewest gains in species. Cooler and wetter cities like Quebec, Ottawa, Winnipeg, Kansas City, and Omaha are expected to welcome the most new species.  Warmer cities with higher precipitation – like cities in coastal California – are projected to lose the most species. 

More than 95% of bird and insect species are predicted to experience a change in the number of cities they call home.  Canines, most amphibians, and aquatic birds are expected to experience the greatest losses.  Toads, turtles, mice and pelicans are projected to become more common overall. 

The researchers hope their findings will encourage more efforts to protect biodiversity.

**********

Web Links

North American cities may see a major species turnover by the end of the century

Photo, posted September 29, 2013, courtesy of Jonathan Kriz via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

AI’s Environmental Footprint

By Leave a Comment

Artificial intelligence is everywhere these days.  Some say it is the biggest development since the discovery of fire.  There is a lot of hype regarding AI, and it will be a while before the hype is sorted out from the reality.  But one thing that is certain is that AI is resource-intensive and has a large environmental footprint.

AI use directly produces carbon emissions from its consumption of non-renewable electricity and is also responsible for the consumption of billions of gallons of fresh water.

Various forms of AI run on many types of devices, but the kind of AI we hear about the most – such as ChatGPT – requires specialized computer equipment that runs in large cloud data centers.  There are roughly 10,000 such centers worldwide and more are under construction.  Estimates are that electricity consumption from data centers will double between 2022 and 2026 to a total of 1,000 terawatts, roughly as much electricity as all of Japan uses.

These estimates include all data center activities, not just AI.  Most operators of data centers don’t reveal what percentage of their energy use comes from AI.  One exception is Google, which says machine learning accounts for about 15% of its data center energy use.

Data centers also consume a great deal of water to cool delicate electronics.  In 2022, Google’s data centers consumed about 5 billion gallons of fresh water.

AI has the potential to improve the efficiency of systems, improve climate models, and perhaps help develop new ways to help reduce humanity’s environmental footprint.  But at the moment, it represents an increasing burden on the environment that cannot just be ignored.

**********

Web Links

As Use of A.I. Soars, So Does the Energy and Water It Requires

Photo, posted January 23, 2023, courtesy of Aileen Devlin / Jefferson Lab via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio

The Sounds Of Coral Reefs | Earth Wise          

By Leave a Comment

Coral reefs around the world face multiple threats from climate change, pollution, and other impacts of human activity.  Reef conservation and restoration projects must be able to monitor the health of reefs and that is not such a simple matter.  Surveying reefs generally is labor-intensive and time consuming.  But in a new study, scientists at the University of Exeter in the UK have found a new way to do it.

The fish and other creatures living on coral reefs produce a vast range of sounds.  The meaning of these various sounds is for the most part unknown, but reefs nonetheless have distinctive sonic signatures.

The Exeter researchers decided to make use of machine learning technology.  They trained a computer algorithm using multiple recordings of both healthy and degraded coral reefs.  This essentially taught the computer to learn the difference between them.  A computer can pick up patterns that are undetectable to the human year.  This application of artificial intelligence can tell us faster and more accurately how a reef is doing.

The computer was then used to analyze a set of new recordings, and successfully identified reef health 92% of the time.  The team then was able to use this technique to track the progress of reef restoration efforts.

It is generally much cheaper and easier to deploy an underwater hydrophone on a reef and leave it there instead of having expert divers make repeated visits to a reef to survey its status.  Sound recorders and artificial intelligence could be used around the world to monitor the health of coral reefs and determine whether efforts to protect and restore them are working.

**********

Web Links

AI learns coral reef “song”

Photo, posted January 11, 2015, courtesy of Falco Ermert via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio.

Deep Learning And Dirty Air | Earth Wise

By Leave a Comment

Poor air quality is a major global problem.  According to the World Health Organization, exposure to air pollution is linked to the premature deaths of an estimated seven million people every year.  In fact, 9 out of 10 people breathe air that contains more pollutants than what the WHO considers safe.  Air pollution is the fourth largest threat to human health, trailing only high blood pressure, dietary risks, and smoking.

But predicting pollution levels at a given place and time remains challenging.  According to a new study recently published in the journal Science of the Total Environment, scientists are turning to deep learning to improve air quality estimates. 

According to researchers, satellite observations and ground observations both measure air pollution, but both have major limitations.  For example, satellites may collect data at the same time and at the same location each day, but they miss how emissions may vary throughout the day.  Ground-based observations from weather stations do continuously collect data, but they only do so in a limited number of locations.    

As a result, scientists have turned to deep learning – a type of machine learning – to analyze the relationship between satellite and ground-based observations of nitrogen dioxide around Los Angeles.  Nitrogen dioxide is associated with emissions from traffic and power plants.  The researchers were able to rely on the learned relationship to take daily satellite observations and create hourly estimates of atmospheric nitrogen dioxide levels in approximately three mile grids.     

According to the research team, this study could be repeated for other greenhouse gases, and applied to different cities and regions – or even whole continents. 

**********

Web Links

Scientists turn to deep learning to improve air quality forecasts

Air Pollution

Photo, posted November 4, 2019, courtesy of Ninara via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio.

Climate Change And Harsh Winters

By Leave a Comment

In recent years, there have been some unusually harsh winters in North America and Central Europe. This past January, the Midwestern US experienced extreme cold temperatures. We have all become familiar with the term ‘polar vortex’ and its role in sending cold air to middle latitudes and it is generally agreed that unusual behavior by the jet stream is the primary cause of the extreme winter weather.

For years, climate researchers around the world have been investigating the question as to whether the increasingly common wandering of the jet stream is a product of climate change or is a random phenomenon associated with natural variations in the climate system.

The jet stream is a powerful band of westerly winds over the middle latitudes that push major weather systems from west to east.  These days, the jet stream is increasingly faltering.  Instead of blowing along a straight course parallel to the equator, it sweeps across the Northern Hemisphere in massive waves, producing unusual intrusions of Arctic air into the middle latitudes.

Atmospheric researchers at the Alfred Wegener Institute in Germany have now developed a climate model that can accurately predict the frequently observed winding course of the jet stream.  The breakthrough combines their global climate model with a new machine learning algorithm on ozone chemistry.  Using their new combined model, they can now show that the jet stream’s wavelike course in winter and subsequent extreme weather outbreaks are the direct result of climate change.  The changes in the jet stream are to a great extent caused by the decline in Arctic sea ice, according to the results of the investigations.  The results are not surprising but there is now a detailed model to support the hypothesis.

**********

Web Links

A warming Arctic produces weather extremes in our latitudes

Photo, posted January 11, 2011, courtesy of Carl Wycoff via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio.

Fish And Ships

By Leave a Comment

Solutions to overfishing of certain tuna and shark populations have been hindered by some significant unknowns:  where the fishing is happening, and where the fish are.  But researchers from Stanford University have recently shed some light on this mystery. 

According to a paper recently published in the journal Science Advances, the researchers have developed a map of shark, tuna, and ship movements that could help ocean managers identify regions of the sea where vulnerable species may be at risk. 

The researchers’ work builds on a 2018 study in which four years of satellite vessel movement data was used to develop a machine learning algorithm that mapped the footprint of 70,000 different fishing vessels around the world.  In their paper, the researchers zeroed in on the activities of 900 vessels from 12 countries in the northeast Pacific Ocean to figure out to what degree fishing fleets, sharks, and tunas overlapped. 

The researchers combined the vessel data with the ocean habitat preferences of sharks and tunas obtained from a decade-long tracking program called Tagging of Pacific Predators (or TOPP).  According to the IUCN, most of the 876 individuals tagged in TOPP belong to species that are either threatened or near-threatened.

By synthesizing this data, researchers were able to map where sharks and tunas would have the highest overlap with commercial fishing vessels.  Increasing the transparency of where fish meets fleets will allow ocean managers to determine where international protections may be needed. 

The United Nations is currently developing the world’s first legally binding treaty to protect international waters.  The Stanford University researchers hope their findings can help with this treaty’s formulation. 

**********

Web Links

Tunas, sharks and ships at sea

Photo, posted June 20, 2011, courtesy of Mike Baird via Flickr.

Earth Wise is a production of WAMC Northeast Public Radio.

A Breakthrough In Animal Identification

By Leave a Comment

Researchers from the University of Wyoming have developed a computer model that can identify wild animals in camera-trap photographs with remarkable accuracy and efficiency.

This breakthrough in artificial intelligence (AI), detailed in a paper recently published in the scientific journal Methods in Ecology and Evolution, represents a significant advancement in the study and conservation of wildlife. According to the paper’s authors, “the ability to rapidly identify millions of images from camera traps can fundamentally change the way ecologists design and implement wildlife studies.”

This study builds on previous research from the university in which a computer model analyzed 3.2 million images captured by camera traps in Africa.  The A-I technique called deep learning categorized animal images at a 96.6% accuracy rate.  This was the same accuracy rate as teams of human volunteers achieved, but the computer model worked at a much more rapid pace. 

In the latest study, UW researchers trained a deep neural network on a powerful computer cluster to classify wildlife species using 3.37 million camera-trap images of 27 different animal species.  The model was tested on nearly 375,000 images at a rate of about 2,000 images per minute. It achieved a 97.6% accuracy rate, which is likely the highest accuracy to date in using machine learning for wildlife image classification. 

Artificial intelligence has been used in environmental science in other ways as well. For example, AI has been used to increase agricultural yields in farm fields and to help predict extreme weather. 

Maybe artificial intelligence can prove to be a game changer for the environment.   


**********

Web Links

Researchers Successfully Train Computers to Identify Animals in Photos

Photo, posted January 8, 2012, courtesy of the U.S. Fish and Wildlife Service via Flickr. 

Earth Wise is a production of WAMC Northeast Public Radio.

Big Data + Technology = Improved Global Health

By

Scientists are calling for the creation of a global early warning system for infectious diseases. Such a system would use computer models to tap into environmental, epidemiological, and molecular data – gathering the intelligence needed to forecast where disease risk is high and what actions could prevent outbreaks or contain epidemics.

[Read more…] about Big Data + Technology = Improved Global Health

WAMC Northeast Public Radio

WAMC/Northeast Public Radio is a regional public radio network serving parts of seven northeastern states (more...)