GIS News

Mapping global fishing activity with machine learning

Sep 15 2016 [Archived Version] □ Published at Maps under tags  causes & community energy & environment machine learning maps

The world’s oceans and fisheries are at a turning point. Over a billion people depend on wild-caught fish for their primary source of protein. Fisheries are intertwined with global food security, slave labor issues, livelihoods, sovereign wealth and biodiversity but our fisheries are being harvested beyond sustainable levels. Fish populations have already plummeted by 90 percent for some species within the last generation, and the human population is only growing larger. One in five fish entering global markets is harvested illegally, or is unreported or unregulated. But amidst all these sobering trends, we're also better equipped to face these challenges — thanks to the rise of technology, increased availability of information, and a growing international desire to create a sustainable future.

Today, in partnership with Oceana and SkyTruth, we’re launching Global Fishing Watch, a beta technology platform intended to increase awareness of fisheries and influence sustainable policy through transparency. Global Fishing Watch combines cloud computing technology with satellite data to provide the world’s first global view of commercial fishing activities. It gives anyone around the world — citizens, governments, industry, and researchers — a free, simple, online platform to visualize, track, and share information about fishing activity worldwide.

Global_Fishing-Effort.gif
Global Fishing Watch, the first global view of large scale commercial fishing activity over time

At any given time, there are about 200,000 vessels publicly broadcasting their location at sea through the Automatic Identification System (AIS). Their signals are picked up by dozens of satellites and thousands of terrestrial receivers. Global Fishing Watch runs this information — more than 22 million points of information per day — through machine learning classifiers to determine the type of ship (e.g., cargo, tug, sail, fishing), what kind of fishing gear (longline, purse seine, trawl) they’re using and where they’re fishing based on their movement patterns. To do this, our research partners and fishery experts have manually classified thousands of vessel tracks as training data to “teach” our algorithms what fishing looks like. We then apply that learning to the entire dataset — 37 billion points over the last 4.5 years — enabling anyone to see the individual tracks and fishing activity of every vessel along with its name and flag state.

Valmitao.gif
An individual vessel fishing off Madagascar

This data can help inform sustainable policy and identify suspicious behaviors for further investigation. By understanding what areas of the ocean are being heavily fished, agencies and governments can make important decisions about how much fishing should be allowed in any given area. Often, fish populations are so depleted that the only way to ensure they are replenished is to create “no take areas” where fishing is not allowed. Our hope is that this new technology can help governments and other organizations make decisions about which areas need protection and monitor if policies are respected.

PIPA_Closure.gif
Kiribati's Phoenix Island Protected Area transitioning from heavy tuna fishing to a protected area
Partners have already started using Global Fishing Watch and have committed to providing additional data sources for greater impact:

  • Indonesia’s Minister of Fisheries and Marine Affairs, Susi Pudjiastuti, has committed to making the government’s Vessel Monitoring System (VMS) public in Global Fishing Watch in 2017. Ibu Susi has been a progressive leader for transparency in fisheries with other governments now expressing similar interest to collaborate.
  • Food and Agriculture Organization of the United Nations will collaborate on new research methodologies for reporting spatial fishery and vessel statistics, building on Global Fishing Watch and developing transparency tools to support their member states in improving the monitoring, control and surveillance of fishing activities.
  • Trace Register, a seafood digital supply chain company, has committed to using Global Fishing Watch to verify catch documentation for its customers such as Whole Foods.
  • Bali Seafood, the largest exporter of snapper from Indonesia, has teamed up with Pelagic Data Systems, manufacturers of cellular and solar powered tracking devices to bring the same transparency for small scale and artisanal fishing vessels, into Global Fishing Watch as part of a pilot program.
We’ve also developed a Global Fishing Watch Research Program with 10 leading institutions from around the world. By combining Google tools, methodologies, and datasets in a collaborative environment, they’re modeling economic, environmental, policy, and climate change implications on fisheries at a scale not otherwise possible.

Global Fishing Watch was not possible five years ago. From a technology perspective, satellites were just beginning to collect vessel positions over the open ocean, and the "global coverage" was spotty. There has been tremendous growth in machine learning with applications in new fields. Policy and regulatory frameworks have evolved, with the United States, European Union, and other nations and Regional Fishery Management Organizations now requiring that vessels broadcast their positions. Market forces and import laws are beginning to demand transparency and traceability, both as a positive differentiator and for risk management. All of these forces interact and shape each other.

Today, Global Fishing Watch is an early preview of what is possible. We’re committed to continuing to build tools, partnerships, and access to information to help restore our abundant ocean for generations to come.

Go explore your ocean at www.globalfishingwatch.org.


Mapping global fishing activity with machine learning

Sep 15 2016 [Archived Version] □ Published at Maps under tags  causes & community environment machine learning maps

The world’s oceans and fisheries are at a turning point. Over a billion people depend on wild-caught fish for their primary source of protein. Fisheries are intertwined with global food security, slave labor issues, livelihoods, sovereign wealth and biodiversity but our fisheries are being harvested beyond sustainable levels. Fish populations have already plummeted by 90 percent for some species within the last generation, and the human population is only growing larger. One in five fish entering global markets is harvested illegally, or is unreported or unregulated. But amidst all these sobering trends, we're also better equipped to face these challenges — thanks to the rise of technology, increased availability of information, and a growing international desire to create a sustainable future.

Today, in partnership with Oceana and SkyTruth, we’re launching Global Fishing Watch, a beta technology platform intended to increase awareness of fisheries and influence sustainable policy through transparency. Global Fishing Watch combines cloud computing technology with satellite data to provide the world’s first global view of commercial fishing activities. It gives anyone around the world — citizens, governments, industry, and researchers — a free, simple, online platform to visualize, track, and share information about fishing activity worldwide.

Global_Fishing-Effort.gif
Global Fishing Watch, the first global view of large scale commercial fishing activity over time

At any given time, there are about 200,000 vessels publicly broadcasting their location at sea through the Automatic Identification System (AIS). Their signals are picked up by dozens of satellites and thousands of terrestrial receivers. Global Fishing Watch runs this information — more than 22 million points of information per day — through machine learning classifiers to determine the type of ship (e.g., cargo, tug, sail, fishing), what kind of fishing gear (longline, purse seine, trawl) they’re using and where they’re fishing based on their movement patterns. To do this, our research partners and fishery experts have manually classified thousands of vessel tracks as training data to “teach” our algorithms what fishing looks like. We then apply that learning to the entire dataset — 37 billion points over the last 4.5 years — enabling anyone to see the individual tracks and fishing activity of every vessel along with its name and flag state.

Valmitao.gif
An individual vessel fishing off Madagascar

This data can help inform sustainable policy and identify suspicious behaviors for further investigation. By understanding what areas of the ocean are being heavily fished, agencies and governments can make important decisions about how much fishing should be allowed in any given area. Often, fish populations are so depleted that the only way to ensure they are replenished is to create “no take areas” where fishing is not allowed. Our hope is that this new technology can help governments and other organizations make decisions about which areas need protection and monitor if policies are respected.

PIPA_Closure.gif
Kiribati's Phoenix Island Protected Area transitioning from heavy tuna fishing to a protected area
Partners have already started using Global Fishing Watch and have committed to providing additional data sources for greater impact:

  • Indonesia’s Minister of Fisheries and Marine Affairs, Susi Pudjiastuti, has committed to making the government’s Vessel Monitoring System (VMS) public in Global Fishing Watch in 2017. Ibu Susi has been a progressive leader for transparency in fisheries with other governments now expressing similar interest to collaborate.
  • Food and Agriculture Organization of the United Nations will collaborate on new research methodologies for reporting spatial fishery and vessel statistics, building on Global Fishing Watch and developing transparency tools to support their member states in improving the monitoring, control and surveillance of fishing activities.
  • Trace Register, a seafood digital supply chain company, has committed to using Global Fishing Watch to verify catch documentation for its customers such as Whole Foods.
  • Bali Seafood, the largest exporter of snapper from Indonesia, has teamed up with Pelagic Data Systems, manufacturers of cellular and solar powered tracking devices to bring the same transparency for small scale and artisanal fishing vessels, into Global Fishing Watch as part of a pilot program.
We’ve also developed a Global Fishing Watch Research Program with 10 leading institutions from around the world. By combining Google tools, methodologies, and datasets in a collaborative environment, they’re modeling economic, environmental, policy, and climate change implications on fisheries at a scale not otherwise possible.

Global Fishing Watch was not possible five years ago. From a technology perspective, satellites were just beginning to collect vessel positions over the open ocean, and the "global coverage" was spotty. There has been tremendous growth in machine learning with applications in new fields. Policy and regulatory frameworks have evolved, with the United States, European Union, and other nations and Regional Fishery Management Organizations now requiring that vessels broadcast their positions. Market forces and import laws are beginning to demand transparency and traceability, both as a positive differentiator and for risk management. All of these forces interact and shape each other.

Today, Global Fishing Watch is an early preview of what is possible. We’re committed to continuing to build tools, partnerships, and access to information to help restore our abundant ocean for generations to come.

Go explore your ocean at www.globalfishingwatch.org.


Mapping global fishing activity with machine learning

Sep 15 2016 [Archived Version] □ Published at Maps under tags  causes & community environment machine learning maps

The world’s oceans and fisheries are at a turning point. Over a billion people depend on wild-caught fish for their primary source of protein. Fisheries are intertwined with global food security, slave labor issues, livelihoods, sovereign wealth and biodiversity but our fisheries are being harvested beyond sustainable levels. Fish populations have already plummeted by 90 percent for some species within the last generation, and the human population is only growing larger. One in five fish entering global markets is harvested illegally, or is unreported or unregulated. But amidst all these sobering trends, we're also better equipped to face these challenges — thanks to the rise of technology, increased availability of information, and a growing international desire to create a sustainable future.

Today, in partnership with Oceana and SkyTruth, we’re launching Global Fishing Watch, a beta technology platform intended to increase awareness of fisheries and influence sustainable policy through transparency. Global Fishing Watch combines cloud computing technology with satellite data to provide the world’s first global view of commercial fishing activities. It gives anyone around the world — citizens, governments, industry, and researchers — a free, simple, online platform to visualize, track, and share information about fishing activity worldwide.

Global_Fishing-Effort.gif
Global Fishing Watch, the first global view of large scale commercial fishing activity over time

At any given time, there are about 200,000 vessels publicly broadcasting their location at sea through the Automatic Identification System (AIS). Their signals are picked up by dozens of satellites and thousands of terrestrial receivers. Global Fishing Watch runs this information — more than 22 million points of information per day — through machine learning classifiers to determine the type of ship (e.g., cargo, tug, sail, fishing), what kind of fishing gear (longline, purse seine, trawl) they’re using and where they’re fishing based on their movement patterns. To do this, our research partners and fishery experts have manually classified thousands of vessel tracks as training data to “teach” our algorithms what fishing looks like. We then apply that learning to the entire dataset — 37 billion points over the last 4.5 years — enabling anyone to see the individual tracks and fishing activity of every vessel along with its name and flag state.

Valmitao.gif
An individual vessel fishing off Madagascar

This data can help inform sustainable policy and identify suspicious behaviors for further investigation. By understanding what areas of the ocean are being heavily fished, agencies and governments can make important decisions about how much fishing should be allowed in any given area. Often, fish populations are so depleted that the only way to ensure they are replenished is to create “no take areas” where fishing is not allowed. Our hope is that this new technology can help governments and other organizations make decisions about which areas need protection and monitor if policies are respected.

PIPA_Closure.gif
Kiribati's Phoenix Island Protected Area transitioning from heavy tuna fishing to a protected area
Partners have already started using Global Fishing Watch and have committed to providing additional data sources for greater impact:

  • Indonesia’s Minister of Fisheries and Marine Affairs, Susi Pudjiastuti, has committed to making the government’s Vessel Monitoring System (VMS) public in Global Fishing Watch in 2017. Ibu Susi has been a progressive leader for transparency in fisheries with other governments now expressing similar interest to collaborate.
  • Food and Agriculture Organization of the United Nations will collaborate on new research methodologies for reporting spatial fishery and vessel statistics, building on Global Fishing Watch and developing transparency tools to support their member states in improving the monitoring, control and surveillance of fishing activities.
  • Trace Register, a seafood digital supply chain company, has committed to using Global Fishing Watch to verify catch documentation for its customers such as Whole Foods.
  • Bali Seafood, the largest exporter of snapper from Indonesia, has teamed up with Pelagic Data Systems, manufacturers of cellular and solar powered tracking devices to bring the same transparency for small scale and artisanal fishing vessels, into Global Fishing Watch as part of a pilot program.
We’ve also developed a Global Fishing Watch Research Program with 10 leading institutions from around the world. By combining Google tools, methodologies, and datasets in a collaborative environment, they’re modeling economic, environmental, policy, and climate change implications on fisheries at a scale not otherwise possible.

Global Fishing Watch was not possible five years ago. From a technology perspective, satellites were just beginning to collect vessel positions over the open ocean, and the "global coverage" was spotty. There has been tremendous growth in machine learning with applications in new fields. Policy and regulatory frameworks have evolved, with the United States, European Union, and other nations and Regional Fishery Management Organizations now requiring that vessels broadcast their positions. Market forces and import laws are beginning to demand transparency and traceability, both as a positive differentiator and for risk management. All of these forces interact and shape each other.

Today, Global Fishing Watch is an early preview of what is possible. We’re committed to continuing to build tools, partnerships, and access to information to help restore our abundant ocean for generations to come.

Go explore your ocean at www.globalfishingwatch.org.


Patterns of contribution to citizen science biodiversity projects increase understanding of volunteers’ recording behaviour

Sep 13 2016 [Archived Version] □ Published at Po Ve Sham – Muki Haklay's personal blog under tags  biological recording citizen science concepts gis papers

One of the facts about academic funding and outputs (that is, academic publications), is that there isn’t a simple relationship between the amount of funding and the number, size, or quality of outputs. One of the things that I have noticed over the years is that a fairly limited amount (about £4000-£10,000) are disproportionately effective. … Continue reading Patterns of contribution to citizen science biodiversity projects increase understanding of volunteers’ recording behaviour


Published: Why is Participation Inequality Important?

Sep 12 2016 [Archived Version] □ Published at Po Ve Sham – Muki Haklay's personal blog under tags  citizen science concepts cost cost energic cost ic1203

I’ve mentioned the European Handbook for Crowdsourced Geographic Information in the last post, and explained how it came about. My contribution to the book is a chapter titled ‘Why is Participation Inequality Important?‘. The issue of participation inequality, also known as the 90:9:1 rule, or skewed contribution, has captured my interest for a while now. … Continue reading Published: Why is Participation Inequality Important?


New book: European Handbook of Crowdsourced Geographic Information

Sep 12 2016 [Archived Version] □ Published at Po Ve Sham – Muki Haklay's personal blog under tags  citizen science citizen science data cost cost energic cost ic1203

COST ENERGIC is a network of researchers across Europe (and beyond) that are interested in research crowdsourced geographic information, also known as Volunteered Geographic Information (VGI). The acronym stands for ‘Co-Operation in Science & Technology’ (COST) through ‘European Network Researching Geographic Information Crowdsourcing’ (ENREGIC). I have written about this programme before, through events such as twitter … Continue reading New book: European Handbook of Crowdsourced Geographic Information


Geospatial Technology for Building Smarter Cities

Sep 08 2016 [Archived Version] □ Published at GIS Cloud under tags  blog emergency response fire hazards geospatial solutions gis

What is the role of geospatial technologies in building smarter cities? The concept of smart cities became one of the most frequently mentioned buzzwords in the terms of building and designing sustainable urban environments in the 21st century. But, this does not necessarily mean that we all understand what exactly a smart city is, and...


Hailing more ride service options in Google Maps

Sep 08 2016 [Archived Version] □ Published at Maps under tags  maps

Back in March, we introduced a new way for people to find and compare the fastest ways to get around town by adding a new ride services tab when searching for directions in Google Maps. Today, we’re adding two more partners in the U.S., Lyft and Gett. Now Google Maps will display options from 9 ride-sharing partners in over 60 countries, allowing you to compare the fastest, most affordable ride near you, without having to download and open multiple apps.

Say you’re looking to get from the High Line to Times Square in Manhattan. When typing these locations into the Google Maps app, you’ll see a ride services tab appear alongside driving, transit and walking directions. Just tap the icon and you’ll find fare estimates and pick up times from multiple ride service partners, depending on driver availability. We’ll also show various types of services offered by each partner— for instance Lyft may also show options for a Lyft Line ride.

Uber_Lyft.png

Ride options from Lyft will begin appearing across the U.S., while Gett will show availability within New York City. So the next time you find yourself with an appointment across the city, just open the Google Maps app on iOS or Android and take it for a spin.


Hailing more ride service options in Google Maps

Sep 08 2016 [Archived Version] □ Published at Maps under tags  maps

Back in March, we introduced a new way for people to find and compare the fastest ways to get around town by adding a new ride services tab when searching for directions in Google Maps. Today, we’re adding two more partners in the U.S., Lyft and Gett. Now Google Maps will display options from 9 ride-sharing partners in over 60 countries, allowing you to compare the fastest, most affordable ride near you, without having to download and open multiple apps.

Say you’re looking to get from the High Line to Times Square in Manhattan. When typing these locations into the Google Maps app, you’ll see a ride services tab appear alongside driving, transit and walking directions. Just tap the icon and you’ll find fare estimates and pick up times from multiple ride service partners, depending on driver availability. We’ll also show various types of services offered by each partner— for instance Lyft may also show options for a Lyft Line ride.

Uber_Lyft.png

Ride options from Lyft will begin appearing across the U.S., while Gett will show availability within New York City. So the next time you find yourself with an appointment across the city, just open the Google Maps app on iOS or Android and take it for a spin.


Hailing more ride service options in Google Maps

Sep 08 2016 [Archived Version] □ Published at Maps under tags  maps

Back in March, we introduced a new way for people to find and compare the fastest ways to get around town by adding a new ride services tab when searching for directions in Google Maps. Today, we’re adding two more partners in the U.S., Lyft and Gett. Now Google Maps will display options from 9 ride-sharing partners in over 60 countries, allowing you to compare the fastest, most affordable ride near you, without having to download and open multiple apps.

Say you’re looking to get from the High Line to Times Square in Manhattan. When typing these locations into the Google Maps app, you’ll see a ride services tab appear alongside driving, transit and walking directions. Just tap the icon and you’ll find fare estimates and pick up times from multiple ride service partners, depending on driver availability. We’ll also show various types of services offered by each partner— for instance Lyft may also show options for a Lyft Line ride.

Uber_Lyft.png

Ride options from Lyft will begin appearing across the U.S., while Gett will show availability within New York City. So the next time you find yourself with an appointment across the city, just open the Google Maps app on iOS or Android and take it for a spin.