Can tech save the world?
Written by Peter Walker
Whether it be from David Attenborough’s documentaries, Greta Thunberg’s teenage activism or the Extinction Rebellion protests, it’s hard to miss the message about the earth’s impending climate catastrophe.
But as governments miss environmental action targets, corporations greenwash their way to shareholder satisfaction and individuals struggle to make any impact, perhaps salvation comes from the variety of tech startups trying to make a difference.
In September, the UK government announced that scientists and innovators will be able to access up to £1 billion of aid funding to develop and test new technology targeted at tackling climate change in developing countries.
Addressing the UN General Assembly, prime minister Boris Johnson detailed the fund’s aims, which include providing affordable access to electricity through solar technology; enhancing large-scale battery technology to replace polluting diesel generators; working with factories to reduce their carbon output; and designing low-emission and electric vehicles.
“I have always been deeply optimistic about the potential of technology to make the world a better place – if we get this right, future generations will look back on climate change as a problem that we solved by determined global action and the prowess of technology,” he commented.
Speaking ahead of the same event, Microsoft co-founder Bill Gates recently opined that climate activists were wasting their time lobbying investors to ditch fossil fuel stocks, and would do better to put their money and energy behind the disruptive technologies that slow carbon emissions and help people adapt to a warming world.
“Divestment, to date, probably has reduced about zero tonnes of emissions, it’s not like you’ve capital-starved the people making steel and gasoline,” he told the Financial Times. “When I’m taking billions of dollars and creating breakthrough energy ventures and funding only companies who, if they’re successful, reduce greenhouse gases by 0.5 per cent, then I actually do see a cause and effect type thing.”
With that in mind, National Technology News took a broad look at some of the work being done by ingenious innovators to try and curb the climate crisis before it’s too late.
A multi-scholar report in June looked at the ways artificial intelligence (AI) - and machine learning in particular - was being used to make systems more efficient, enable remote sensing and automatic monitoring, and provide fast approximations to time-intensive simulations.
The paper noted that the nature of the data involved poses challenges, as it is often proprietary or includes sensitive personal information. It also warned that machine learning is not a silver bullet, explaining that technologies that could reduce climate change have been available for years, but have largely not been adopted at scale by society.
Priya L. Donti, a PhD student in computer science and public policy at Carnegie Mellon University, suggested that machine learning can accelerate the development of clean energy technologies by improving forecasts of demand, system management and monitoring.
“Since variable generation and electricity demand both fluctuate, they must be forecast ahead of time to inform real-time electricity scheduling and longer-term system planning,” she stated. “Better short-term forecasts can improve electricity scheduling, enabling operators to both reduce their reliance on polluting standby plants and proactively manage increasing amounts of variable sources.”
She did warn that at the same time, applications in this space hold many potential pitfalls; for instance, innovations that seek to reduce greenhouse gas emissions in the oil and gas industries could actually increase emissions by making them cheaper to emit.
Lynn H. Kaack, a postdoctoral researcher in the energy politics group at ETH Zurich, proposed that machine learning could improve vehicle engineering and enable intelligent infrastructure.
“A colossal amount of transport occurs each day across the world, but much of this mileage is used inefficiently, resulting in needless emissions,” she wrote. “With the help of machine learning, the number of vehicle-miles travelled can be reduced by making long trips less necessary, increasing loading and optimising vehicle routing.”
Nikola Milojevic-Dupont, a researcher at the Mercator Research Institute on Global Commons and Climate Change, focused on how machine learning can reduce emissions from buildings and cities.
“In cities, integrating smart systems beyond gadget-level solutions and activating the potential of data-driven urban planning can play key roles for mitigation, while improving the well-being of citizens,” she noted.
Finally, Anna Waldman-Brown, a PhD student in political economy at the Massachusetts Institute of Technology (MIT), pointed out that industrial production, logistics and building materials are leading causes of difficult-to-eliminate emissions.
“Luckily for researchers, the global industrial sector spends billions of dollars annually gathering data on factories and supply chains – aided by improvements in the cost and accessibility of sensors and other data-gathering mechanisms such as QR codes and image recognition,” she stated.
Those in the field can potentially reduce global emissions by helping to streamline supply chains, improve production quality, predict machine breakdowns, optimise heating and cooling systems, and prioritise the use of clean electricity over fossil fuels.
Internet of Things
Nicholas Hill, chief executive of design and development consultancy Plextek, argued that the global warming challenge is beyond fixing by governments.
“Fundamentally, we need to change the behaviour of the six billion people who inhabit the Earth; or at least influence that proportion of the global population whose consumption is generating most of the climate change,” he said.
“With a combination of back-end data on the environmental cost of basic products and services, Internet of Things (IoT) devices and a suitable app, I could have immediate feedback on the effects of my behaviour on the planet, answering the question ‘what impact have I had today?’,” he explained.
“I could monitor my behaviour and compare it to recommendations or targets from governments or environmental groups, compete with friends or family for low impact scores, or experiment with ‘what if’ scenarios before deciding how to get from A to B,” Hill continued, suggesting that if a good proportion of the two billion smartphone users had access to such an app, people wouldn’t have to wait for governments to act.
The greenhouse gas emissions of buildings are also significant, with the combined emissions from lighting, power, heating and cooling whether at home or in the office, contributing almost 20 per cent of global emissions.
Part of the answer is to build smarter cities, powered by IoT technology. Sidewalk Labs, a part of Google parent company Alphabet, has been harnessing digital technologies to solve the most pressing urban problems. One of its current projects involves looking at how traffic flows through a city and how hotspots of congestion might be solved, which could dramatically reduce air pollution in cities.
Transport represents just under a quarter of global energy-related CO2 emissions, but demand is only set to increase. To power the move to electric vehicles, much more efficient batteries and battery-charging technology is required.
Researchers at the University of Surrey recently announced that they have discovered new materials which offer an alternative to battery power and have proven to be between 1,000 and 10,000 times more powerful than the existing supercapacitor battery alternatives.
A statement from the university read: “The new technology is believed to have the potential for electric cars to travel to similar distances as petrol cars without the need to stop for lengthy recharging breaks of between six and eight hours, and instead recharge fully in the time it takes to fill a regular car with petrol.”
Trevor Jackson’s engineering experience in The Navy, for Rolls Royce and then British Aerospace, led him to eventually form his own fuel cell research and development firm. As chairman of Mal R&D, he is responsible for the 'semi-fuel cell', a mix between fuel cells and batteries that produces power by 'digesting' aluminium, using oxygen from the air and a special electrolyte.
This forms aluminium hydroxide - the same as anti-acid tablets - which can go into the aluminium production industry to be converted to aluminium again - using hydro power - releasing the oxygen consumed in the reaction to make power.
“All smelters are reducing their C02 output, so this is a near zero complete power cycle which is independent of fossil fuels,” he stated, adding that it has about nine times the energy of a comparable weight lithium Ion battery and is fully recyclable.
The system comes in the form of a cassette which is exchanged in 90 seconds, meaning there is no need for a recharging network, nor for any waiting time before getting back on the road. Jackson claims the driver’s cost per mile is about half the current cost of driving with petrol or diesel and the cost to the automaker is about one seventh the cost of an equivalent lithium pack.
However, he’s found it hard to scale up the technology, as government policy favours investment in lithium ion batteries to power the next generation of electric vehicles.
“The progress has depended on a trickle of funds over many years, despite the technical side being relatively straightforward,” Jackson commented. “The continual blocking from InnovateUK has been extremely frustrating, especially as I believe Britain should have been able to prosper from commercialising this technology, with the added benefit of being able to export a climate fix.
“From what I can see, short-range lithium rechargeable batteries as a replacement for fossil fuels appears to be a global policy - but who sets that policy is a mystery - and the hidden issue is that the power for recharging these batteries mainly comes from burning oil and gas in power stations,” he added.
Of course, it’s not just the inventors and entrepreneurs pushing climate change technology forward.
The Anthesis Group was formed to address a market gap between small, local sustainability boutiques and large multi-disciplinaries which only see climate crisis mitigation as an add-on service.
Director Brad Blundell explained that the company has developed software tools for Tesco’s Supplier Network, the Cool Farm Tool and its own SCATTER model.
“We’re currently developing a new version of SCATTER - which assists cities to plan their decarbonisation pathways - as well as delivering ongoing developments to the Cool Farm Tool, which helps farmers with sustainability,” he stated.
Simone Aplin, technical director at Anthesis, commented that if companies and authorities really want to make a difference they need to start by collecting and consolidating data, as well as getting that data right. “Particularly in the resource efficiency sector, the role of technology and the data that feeds into it requires infrastructure in the right places and people with the suitable skills for a solution to be fully functioning end-to-end.”
Blundell said a recent uptick in public awareness and activism has all put pressure on the government, which has responded with a legal commitment to deliver carbon neutrality by 2050, while over half of UK local authorities have declared a climate emergency.
“Similarly, in the corporate world, the ever-growing number of organisations that are setting science-based targets is extremely encouraging,” he continued. “However, society as a whole needs to start reacting to the ‘emergency’ status, as when there is an emergency declared around a building fire we tend not to convene international panels of experts and write lots of technical papers to confirm that the building is indeed on fire – the world needs to act now and technology has a critical role to play in making change happen at pace.”
At the start of 2018, China instituted a ban on importing waste to be recycled, leading to countries across Europe, North America and Australasia having to dramatically rethink their refuse policies, as the country previously took in around half of the world’s plastic and paper products.
While previously the Chinese would pay for containers full of recycling waste, now centres are having to pay other countries to take them, so those authorities which can afford it are turning to technology to help improve on-site processing power.
The gap between what’s collected and what’s recycled has led to investments in automated solutions that use AI to separate mixed recyclables at waste treatment plants.
Norwegian corporation TOMRA, Finnish firm ZenRobotics and Canadian sorting specialist Machinex are all providing bespoke sensor-based solutions to the recycling industry, utilising things like robotic arms that use computer vision systems, spectroscopic cameras and data supplied from sensors to separate items on a conveyor belt based on their size, shape, colour and conductivity level.
This automation means that, for example, Machinex’s optical sorter can work through recyclables at up to 3,000 objects per minute, compared to the average of a human worker’s rate of 800 a minute.
Meanwhile, MIT’s Computer Science and AI Laboratory recently unveiled its RoCycle system, which uses soft Teflon grippers with ‘fingertip’ sensors that imitate the sense of touch, allowing the robotic arm to separate objects.
Currently, RoCycle is slower and less accurate than a human worker, but its developers are aiming to improve its dexterity by combining the tactile input with visual data from cameras.
Forecasting the future
One of the best ways to put rapid advances in computing power to use is to improve the modelling algorithms used to forecast everything from farmer’s crops, to hedge fund bets and emergency agency budgets.
Swedish energy company Tekniska verken found that inaccurate weather forecasts were especially damaging, as it has to predict exactly how much electricity needs to be produced from its wind turbines and put into the energy grid 24 hours in advance. When grids aren’t properly balanced, this creates major costs for energy companies – often around seven per cent of total profits.
“If the prediction is either too low or too high, we will suffer the economic consequences,” commented Erik Olsson, business developer at Tekniska verken.
The company has started using Peltarion’s cloud-based AI platform to develop a deep learning model to improve weather forecasting. The Deep Weather system identifies patterns in both live and historical weather data, making predictions in 100 milliseconds on a computer that costs €10,000, compared to the previous three hours on €30 million worth of computer hardware.
Its operators send weather data from their windmills to a meteorological institute, receiving a customised forecast for their exact location, specifying exactly how much energy needs to be produced.
Olsson noted that the system has the potential to create all sorts of new solutions, suggesting that offices could be provided with updated forecasts and automatically adjust their climate systems.
FinTech for good
In October, the Financial Conduct Authority (FCA) published a statement setting out its proposals to improve climate change disclosures by issuers, integration of climate change risks into decision-making at firms and information to consumers on green financial products and services.
The regulator will consult on new rules to improve climate-related disclosures by certain firms and clarify existing obligations, while also finalising rule changes requiring Independent Governance Committees to oversee and report on firms’ environmental, social and governance (ESG) and stewardship policies.
Last October, the FCA launched the Green FinTech Challenge to support innovation. The challenge was open to startups, incumbents and technology providers assisting in the transition to a greener economy, with 22 applications and nine firms accepted by the end of the application window this January.
Among those firms entered into a regulatory sandbox to test their wares on consumers, are Enian, a machine learning tool that uses live satellite data and draws on a proprietary data library of 100,000 commercial solar photovoltaics and wind power assets to predict investment returns for greenfield projects; and Sustainability.Exchange, a platform for corporate ESG performance ratings and a blockchain-based prediction market for ESG-focused investors.
The regulator is also giving direct support to a Barclays impact investing ecosystem platform aiming to connect accelerators, fund managers, institutional investors, investor networks and wealth managers; along with the CoEnergy.app, a platform that automates the setup and management of share registry services for community energy projects; and Cogo, an app that connects consumers with businesses that align with their social and environmental values.
The investment world is slowly waking up to the movement too, with a recent example coming in the form of Clean Energy Ventures (CEV), which recently closed its first Clean Energy Venture Fund with $110 million to spend on seed and early-stage investments in promising environmental energy tech.
“After more than a decade of investing in the advanced energy sector, it's been gratifying that this fund focused on investments that address climate risks, was significantly oversubscribed,” said Daniel Goldman, managing director at CEV. “Investors are clearly seeing that increasing commercial adoption of advanced energy innovations is creating opportunities to earn attractive risk-adjusted returns.”
The data centres enabling our ‘always-on’ lifestyle consume three per cent of the global electricity supply and account for two per cent of greenhouse gas emissions, putting cloud computing on par with the airline industry.
But where airlines face rising fuel costs and falling margins, cloud providers are accelerating attempts to improve energy efficiency. The economic model of cloud computing lies in spreading the data centre costs by co-locating users, which is more efficient compared to private data centres.
Server capacity in the cloud scales up and down to fit fluctuating requirements, so customers use only the energy they need. The biggest providers Adobe, Microsoft Azure and Amazon Web Services have already committed to achieving 100 per cent renewable energy usage.
Hendrik-Jan Boer, head of sustainable and impact equity investing at NN Investment Partners, commented: “Our sustainable equity funds are well positioned to take advantage of the growth in cloud computing - for example, with holdings in companies such as Adobe, Intuit and Microsoft - which has a positive impact on both our carbon footprint and alpha generation.
“As the digital transformation continues and cloud computing becomes an ever more inescapable part of our daily lives, we project growing opportunities for investors seeking to reduce carbon emissions and still benefit from alpha generation.”
With the UK’s drive for zero emissions by 2050, electricity providers have had to invest in regaining control of their networks in order to adequately manage changing energy demand.
Paul Clark, sector director for energy and utilities at SSE Enterprise, said he was starting to see suppliers moving to deploying powerline monitoring technology across their infrastructure. “However, there needs to be widespread uptake if the UK power networks are to become more reliable and resilient – monitoring technology can provide enhanced protection, automation and control to reduce both outages and engineering costs.”
For example, if power companies were to start collecting data from the sensors, they would be able to achieve predictive maintenance of critical assets that are often remote or inaccessible. “Early identification of issues such as ice build-up on power lines, or impending failures in underground cable systems, are also possible through the technology,” said Clark.
“In the coming years, connectivity providers are going to help shape a world where both businesses and consumers are able to take greater control over their energy consumption and even sell energy back into the grid – this will have a profound impact on the reduction of carbon emissions.”
Christian Chudoba, chief executive of Lumenaza, started his business in 2013 to connect and intelligently control producers and consumers of renewable energy. Today, the modular software platform works as a ‘utility-in-a-box’, shortening the time-to-market for companies, from small rooftop photovoltaic systems up to large wind farms.
After expanding in mainland Europe, a project is about to launch in the UK, but Chudoba accepted that they can only move as fast as national rules allow.
“We advocate fast actions; climate change is real and does not wait for us to make decisions,” he stated. “On one hand, there is politics, which must give CO2 a price. On the other hand, we believe in the innovative strength of companies – they can make a valuable contribution to implementing climate protection cost-effectively, efficiently and quickly.”
The rebound effect
A 2017 United Nations report found energy-efficient technology could - on an annual basis - cut 25 billion tons of greenhouse gas emissions, 17 million tons of particular matter and three billion tons of human-toxic waste. Sensors and software could also double the energy efficiency of buildings, reduce metal consumption by up to three quarters and save natural resources like water and land by 200 billion cubic meters and 150,000 square kilometres, respectively.
However, the UN also warned that using green technology may be less beneficial than expected, due to what’s known as the rebound effect – where technologically-driven advances in energy efficiency increase, rather than decrease, consumption leading to net-zero emissions. A paper by Barker, Dagoumas and Rubin suggested this may erode up to half of the eco-benefits promised by green technology by 2030.
Ashley Nunes, an academic at MIT and Harvard University, said one example is that as electric cars cost less to run, consumers may drive them further and more often.
He also pointed out that consumer behaviour affects energy efficiency in other ways, citing British research which found buildings designed to save energy don’t always perform as expected, “partially because occupants behave in more complex ways than designers account for – they open windows, leave doors open, generate body heat, keep tropical fish tanks and install plasma TV screens”.
Mike Childs, head of science, policy and research at Friends of the Earth, commented that while technology must form a significant part of the climate crisis solution - with innovation speeding up the journey towards sustainability and reducing the costs involved - it is not a panacea.
“Because governments have for 30 years failed to listen to scientific warnings, we need to reduce greenhouse gases, but also remove carbon from the air,” he stated, noting emerging technologies which directly capture carbon and store it underground. “It’s currently horrendously expensive, but will definitely be necessary, so that needs development, but we also need more cost reduction in solar and wind power – particularly offshore, floating wind turbines.”
One example is Carbon Engineering, a Canadian startup working on taking carbon dioxide directly from the atmosphere and using it to produce fuel. According to the company, “direct air capture can remove far more CO2 per acre of land footprint than trees and plants”, and it is already running a demonstration plant in British Columbia that is removing one ton of CO2 from the air every day.
Childs said home conversion also has a part to play in moving away from natural gas for heating and power towards using heat storage and distribution systems. “The human/tech interface is important here, helping people control their energy use using smart meters to turn heating on and off within individual rooms, while new tariff systems from the likes of Octopus Energy will also be key.”
He pointed out problems with the wider energy market, which in the UK has only recently been opened up to smaller challengers by regulatory pressure.
“Companies will and do respond to public pressure too, and appear to be evolving in the right direction – some are buying up innovative startups and bringing the technology in-house, getting things to market quicker.”