Can air pollution be controlled by drones?

1) Air pollution is referred to any contamination of the indoor or outdoor environment that modifies the characteristics of the atmosphere. Although more research is required to further understand the role poor air quality and multi-pollutant exposure plays in health. According to the World Health Organization (WHO) in 2012, an estimated 6.5 million deaths were associated with indoor and outdoor air pollution together, this r4epresents 11.6% of all global deaths.

Major sources of air pollution vary from country to country and in every city, depending on their infrastructure and industrial activities, but in general, common sources include motor vehicles, household combustion devices and waste burning, coal-fired power plants, and industrial activities.

Pollutants of major public health concern include particulate matter, carbon monoxide, ozone, nitrogen dioxide and sulfur dioxide. (EPA)

The most relevant issue in this problem is that once the pollution has been produced it is very difficult to clean because of dispersion, and while reducing the pollution production is a relatively simple solution ( through filters and catalytic combustion, etc) once the pollution is diffused in the environment, it becomes a very complex problem to solve, mainly because of the dimension of it. In other words, it is impossible to clean all the air in the atmosphere. china-may-use-drones-to-kill-the-smog-problem

2)Nonetheless, China is investing a large quantity of money to fix this difficult problem, and many innovative solutions are being implemented, such as air purifiers the size of buildings, or mist cannons that nebulizer liquids to trap harmful particles, but still, these solutions face the great challenge of diluted pollution and large dimensions.

A novel solution that is being tried is the smog-busting drones, the idea is to use drones to spray chemicals [liquid nitrogen],  to solidify pollutants in the air and fall to the ground.

“When liquid nitrogen is dispersed in air, it readily absorbs heat from the surrounding atmosphere, causing water vapor to instantly condense. The condensate would drag down particulate matter along with it as it falls to earth” Emily Carino-postdoctoral researcher in chemical engineering with a PhD from the University of Texas

The chief executive of the company, Ma Yongsheng, he explained that drones have a reach of 5km and can carry 700 kg of smog-clearing chemicals, and have conducted over 100 hours of testing flight. [South China Morning Post]

Nonetheless, there is a big international opposition to this approach of geoengineering, mainly because of two postures, exposed by Emily Carino, from the University of Texas.

The first one is the effect that this chemical rain can have on citizens. “Used improperly, liquid nitrogen is dangerous: It can cause severe cold burns if it comes in contact with skin, and items cooled by liquid nitrogen can stick fast to human skin when touched.”

And the second is the risk of the unintended consequences of such chemical reaction, not only because the reaction occurs so quickly, but also because nitrogen outperforms all other agents during the reaction.

3) This technology is mainly to be deployed by the government since air pollution or air quality is a public good.

4) The steps to deploy this technology are to run a pilot and have an important and solid baseline to compare the improvements as well as the unintended consequences.

Nonetheless, I am very skeptical about the impact of this technology application due to the unintended consequences, and most importantly it is important to point out the irrational thinking of creating rain from pollutants, instead of deploying filters and air pollution control systems in the industry and combustion vehicles.


Images sources: Image 1, Image 2

Protecting the oceans through information sharing

coral-reef-1024x682-1024x6821) Oceans cover three-quarters of the Earth’s surface, contain 97 % of the Earth’s water, feed over a billion people and over three billion people depend on marine and coastal biodiversity for their livelihoods. (UN) Unfortunately, oceans and the species that depend on them— including humans—face an uncertain future. The combined effects of overfishing, poorly managed resource extraction, serial depletion of marine species and warming waters have contributed to the collapse of 70% of the world’s most productive fisheries, the loss of a third of the world’s coral reefs, 90 percent of the oceans’ top predators(WCS).  Illegal fishing and poorly manage resource extraction represent a significant problem in fisheries. (Mission Blue)

Screen Shot 2017-11-09 at 12.51.58 PM.png

2) Vessel monitoring systems, together with satellite tracking and machine learning allow to create transparency in the seafood supply chain, promoting sustainability.

Many initiatives such as Global Fishing Watch, Project Eyes on the Sea and TransparentSea. org use the data of Vessel Monitoring Systems (VMS) and Automatic Identification Signals (AIS) from ships at sea, and corroborates the information with satellite images. That information is then processed with artificial intelligence tools to distinguish individual vessel tracks and identify when the vessels are fishing based on their movements fleets to understand their activities and impacts. Finally, most of this information is being shared for free. This enables to share information about where fishing activity is happening and has happened. For both, monitoring and conservation purposes, the former being relevant for law enforcement and later for research purposes.

The sharing of this type of information is important for efficient decision making, revealing where, when, and how much fishing occurs can help decision makers develop informed policy as well as strategic management and enforcement programs that create sustainable fisheries and reduce Illegal Unregulated and Unreported fishing (IUU).

“Developing datasets and analyses of global fishing activity can support research on ecology, the environment and human interaction with the ocean.  Global Fish Project

3) This technology could be deployed by policymakers, resource managers and in general all decision makers.

Conservation Organizations and NGOs can use this information to support campaigns to reduce impacts of fishing on the ocean, revealing fishing activity and pressuring officials and industry for changes that will improve sustainability on the oceans.

Industry Members could create and create premium markets and use this tool for fishers and seafood suppliers to verify how and where their products are caught.

Researchers could use this datasets and analyses of global fishing activity to support research on ecology, the environment and human interaction with the ocean.

The beauty of open source data sharing is that the application can be very diverse and varies between different stakeholders.

4) The next steps to deploy this technology is to share best practices in order to create awareness of the possible impacts of enabling data sharing could have, since the information is already public and for free.

Could Virtual Reality help us reduce traveling needs?

1.- Climate change is the result of a mix of carbon emission produced mainly by humans. CO2 emissions, are primarily generated from the combustion of fossil fuels, unfortunately, these have risen dramatically since the start of the industrial revolution.

Transportation plays a major role in the green-hous-gas (GHG) per sector, in fact, according to the UN the global car fleet is expected to triple by 2050. GHG emissions from transport are growing faster than any other sector. (UN)

cait-global-emissions-sectorSource: Center for climate and energy solutions

When we analyze into further details the purposes of transportation, not so much regarding items but specifically humans, there is a significant amount of travels for meetings and business purposes. There is a considerable amount of emissions generated by transporting people to the workspace as well as to meetings in different cities or countries.

2.- Virtual Reality

Virtual Reality headsets with antler-like sensors attached to the goggles enables the user to join a virtual-reality environment in which they see digital avatars of themselves moving around a simulated environment. (Wall Street Journal)

Although conference calls have had a great impact on the way we communicate, there are still some shortfalls, such as limitation of hand gestures, and human contact. The novelty of having an avatar that represents yourself and others in a shared common space enables a set of different interactions.

VR meetings will allow for nuanced nonverbal communication—proper eye contact, subtle cues such as interpersonal distance, and eventually virtual touch and smell (when desired),” Prof. Bailenson– founding director of Stanford University’s Virtual Human Interaction Lab

Virtual-reality systems could replace video conferencing as a common tool for business meetings, that could represent not only time savings but also GHG emission reductions, since traveling for meetings within a city, a country or in the world for meetings won’t be necessary.

3.- The next step to deploy this technology is to improve the internet infrastructure. VR, as well as conference calls, rely on a good internet connection, in many cases is not good enough. Furthermore, enabling VR experiences, through movies and short films can be a compelling start into deploying this technology, by gaining public curiosity and acceptance.

4.- Finally, I see this technology being primarily deployed in the private sector by companies. This could have also a great beneficial impact in terms of cost savings since work travels are expensive, therefore there is a clear incentive for companies to adopt it, and put some effort in the public acceptance. An example of a company working on VR meetings is tvTime.

DNA barcodes for sustainable seafood consumption

dna_barcode1. Fisheries are of primary importance for the livelihood and food security of 200 million people, particularly in the developing world, furthermore, one of five people on this planet depends on fish as the primary source of protein(UN) . The increasing demand for fish and fish products is leading to a rapid growth in market prices. Therefore, fisheries investments have become more attractive, creating even greater barriers for the protection of marine biodiversity due to overfishing. The magnitude of the problem of overfishing is often overlooked. According to a Food and Agriculture Organization (FAO) estimate, over 70% of the world’s fish species are either fully exploited or depleted. The dramatic increase of destructive fishing techniques worldwide destroys marine mammals and entire ecosystems.

2.Based on the DNA sequence of a particular gene it is possible to determine which species the sample belongs to, by comparing the DNA sequence to all of the fish DNA sequences in the universal database GenBank. Less than 1 milligram of fish meat is required to perform this analysis. (FishDNAID)

So far, the publicly-accessible database contains over 1,000 DNA sequences on it until now but planning on adding new sequences on a rolling basis. DNA barcoding provides an additional advantage because DNA among similar species bears a similar resemblance, researchers can usually narrow down a mystery fish. Finally bringing transparency to the seafood supply chain. (Agfunder)

This technology is expected to become highly scalable as it matures,  decreasing the cost, making it more attractive to invest in and easier to use in the field. Soon, fish retailers could rely on a spin-off of the database to verify each import shipment as it crosses their docks, alerting the FDA and refusing the admission into the country, this is expected to reduce the number of illegal fished endangered species.

3.- Although this technology could be developed by a private company, the final adoption of the enforcement authorities it’s crucial. The Fish SCALE (Seafood Compliance and Labeling Enforcement) project at the FDA has been able to accurately classify countless types of edible — and perhaps, more importantly, inedible — types of sea fare based on their DNA, this is a perfect example of how the public sector is developing its own technology innovation programs within the agencies. (Agfunder)

4.-  Although there has been a wave of innovative solution regarding conservation of the marine ecosystem, most of them narrow down to offer two solutions: Monitoring and traceability. For the former, it is important to have strong law enforcement authorities, for the latter it is important to have educated markets. To deploy this technology in the developing world it will be crucial to work hand in hand with the law enforcement authorities as well as to raise awareness among fish retailers.



image source:

Machine learning for vegan food

  1. Meat and animal products are an important source of protein with a relevant nutritional value. Nonetheless, the environmental impact in terms of land use, water consumption and CO2 emissions have become of significant importance, also due to the continuous increasing consumption. By 2050 world meat production is projected to double, most of which is expected in developing countries. [1]

According to the Food and Agriculture Organization, the livestock sector generates more greenhouse gas emissions [Co2e] than transport, by 18%. [2]

It is also a major source of land and water degradation. Livestock’s requires vast tracts of land and a significant demand for feed crops, both contribute to biodiversity loss. Moreover, it’s among the most damaging sectors water resources, not only because of its water consumption, but also contributing to water pollution, eutrophication and the degeneration of coral reefs, due to the manure of livestock.

  1. Machine learning is a subset of AI, that generates algorithms that can learn from data and make predictions on it.  In other words make machines learn from experience, experience coming in form of data and the more the data, the more it learns. Machine learning can be useful for making data-driven predictions or decisions.

To create healthier food, companies like Hamptons Creek are automating the extraction and analysis of plant proteins. This includes examining their molecular features and functional performance such as gelling, foaming, and emulsifying properties. Ultimately, the goal is to feed this research to an AI an through machine-learning algorithms identify the most-promising proteins for use in the creation of vegan food that tastes similar to animal products (mayonnaise, muffins, spreads, and other foods). Finally, we are applying generative design to food production. [3]

According to Lee Chae, Hampton Creek’s head of research and development,  Hampton Creek applies deep machine learning to plant biological data to meet its objective of creating healthier food. [4]

  1. This is not really a new technology rather than a new technology application. To deploy this application, more enterprises in the food industry should be making research in this field, or applying the outcomes. More importantly, I personally don’t think that every enterprise should be conducting the same research rather than to have an open database with the results,  would be a great way to expand the adoption of this knowledge.  Furthermore, this new application raises the question of what other industries could be impacted by machine learning, through generative design? Let’s take for example the polymer industry, imagine to feed the AI data about the plant properties to replace plastics.


  1. Since deep learning and machine learning can be applied to several databases, to make predictions as well as generative design, many industries could benefit from this technology. In my opinion is a private sector driven technology, especially due to its large upfront capital investment.  The first step to deploy it is to have a reliable database, this can be either from a primary or a secondary source, depending on the application.

Although it does require a group of people specialized in artificial intelligence, the results could be applied to the industry with no further disruption, this reduces the barriers to technology adoption.

Internet of things for better decision making

  1. Although most environmental challenges are blocked due to a financial reasons (prices for corrective measures), there is a handful of environmental problems that are due to poor decision making, unnecessary waste production with no secondary markets, energy waste due to a bad control on the thermostat, electricity waste due to a lack of information of consumption of devices, water consumption for irrigation in an unnecessary setting ( high humidity/ after rain), etc. If we analyze the decisionmaking process we tend to see that either there is not enough information at the time the decision had to be taken, or the person in charge to take the decision, did not react to that information.
  2. The Internet of Things allows to have a direct communication between things and the controller, to harvest real-time information, to allow a better decision making. There are 4 major components to the IoT:
  • Hardware: Data collectors or sensors ( temperature, movement, distance,etc.) connects physical items to digital objects.
  • Data: Serves as parameters and metrics for performance. It can be seen as a universal language of things.
  • Software: Once we have the information we need the software to interpret, analyzes the information, and to finally control it, to send out information to react.
  • Connectivity : Enables the communication between the sensors and software, and sends orders.

3. There are multiple applications for the IoT, on a state level, on an industry level, and on a personal level. In my opinion city governments should be the first to deploy these technologies, since it provides a direct financial benefit and a clear reduction in operational cost, furthermore the fact that it’s used at a city level with its main propose is to know either climate conditions ( light, humidity, temperature) or quantity of people ( traffic, public transportation usage, etc.) rather than information of a precise citizen. This is important regarding privacy issues.

4. I think that the early adopters of this technology should be large organizations that can have a large efficiency impacts. Once the technology is mature and has sufficient success stories, the adoption will come naturally.

Policies regarding privacy should be created, establishing in clear terms who owns the collection of the data, and who can use it. This is very important since until now this information has been mainly used to create personal advertising. In my opinion, this should not be the ultimate goal of collecting data.



Internet of Things – Jordan Duffy




Micro Fuel Cell, generating electricity of wastewater

fuel-cell1) According to the UN Water at a global level, 80% of wastewater produced is discharged into the ecosystem untreated, causing widespread water pollution. Wastewater treatment is not only relevant to reduce environmental pollution but also to ensure drinking water supply, around 1.8 billion people use a contaminated source of drinking water. Even though there are numerous processes that can be used to clean up wastewaters depending on the type and extent of contamination, the most common barriers are the affordability of these technologies and their energy consumption.

2) Microbial Fuel Cells

Technology: A microbial fuel cell (MFC), is a bio-electrochemical system that drives an electric current by using bacteria and mimicking bacterial interactions found in nature. “The direct conversion of organic matter to electricity using bacteria” Logan.  These electrochemical cells are constructed using either a bioanode and/or a biocathode. Inside the fuel cell,  anaerobic bacteria releases electrons in an oxygen-free environment. The electrons flow to an anode and then into a circuit to cathodes in a separate chamber on the outside of the membrane.

Potential: MFCs use energy very efficiently than standard, in theory, an MFC is capable of energy efficiency far beyond 50%. Nonetheless, MFCs are only attractive for power generation applications that require only low power, therefore the amount of electricity generated will not compare will a power plant, or even cover the entire processing facility, but it can offset the energy used to clean the water. “The energy we don’t consume is more important than the electricity we might produce” Logan.

Constraints: The fuel cell is ideal for wastewater with a high concentration in organic material, mostly wastewater from agriculture and food processing rather than municipalities. This technology has had interesting advances in the proof of concept but it is still has a  wide range of opportunity to increase processing volumes until it can be implemented at a large scale.

3) Municipal, industrial and agricultural water treatment facilities, will be able to increase the efficiency and reduce the sludge by applying this technology. Furthermore, this technology will enable small-scale decentralized water treatment facilities, owned by either farmers or communities.

4) The next steps to deploy this technology would be to increase processing volumes until it can be implemented at a large scale. Moreover, to evaluate the possibility to automatize the small-scale facilities to reduce operation and maintenance efforts to expand the user market.





Hyperspectral Remote Sensing of Water Quality

  1. Agricultural runoff

Fertilizers and animal manure,  both rich in nitrogen and phosphorus, are the primary sources of nutrient pollution from agricultural sources. Excess nutrients can impact water quality creating toxic algae blooms, that depletes the water of oxygen when the algae die and are eaten by bacteria. Finally, this pollution contributes to creating death zones in the ocean.


2) Design of an agricultural runoff monitoring and reward system for the upper Chesapeake Bay – IEEE- Advancing Technology for Humanity

  • Optical indicators of water quality have the potential to reduce agricultural runoff, by enhancing resource managers to monitor water bodies in a timely and cost-effective manner.
  • “Nitrogen and phosphorus levels can be detected with spectral imagery. As objects on the ground absorb sunlight, they give off a reflectance. The relative intensity of this reflectance is different at certain wavelengths for certain elements, which gives a chemical element’s spectral signature. Multispectral and hyperspectral systems are the two primary systems used today. ” Harry Yoo, George Mason University.
  • Hyperspectral systems can be introduced to monitor nutrient levels on farmland, and work as a data input to create policies and incentivize farmers to adopt best management practices.

3) Environmental Protection Agency, U.S. Geological Survey and the U.S. Department of Agriculture are organizations will need to use the technology.

4) The mapping from some waterbodies will be indispensable to create solutions towards toxic algae bloom caused by agricultural runoff.

  • Capacity building in the different organizations is a first requirement to deploy remote sensing technologies such as hyper-spectral images to collect data from polluted waterbodies. Data collection will require coordination between the different organizations to be more efficient in the impact generated.
  • Once the data is collected it is possible to map the pollution and build a strategy to reduce the runoff through policies.
  • Lastly, to monitor the impact of these policies will be crucial to measure the success of these, and in the given case to re-evaluate the policy strategy.