Cant See the Woods for the Trees

Can’t See The Woods For The Trees

If my internet connection is working then I listen, most days, to Radio 4. I start at 5:45am with Farming Today and then stream through until the shipping forecast. So when I awoke on the 7th December I was surprised to hear the program open with fears that England was suffering from deforestation.

The Deforestation of England

woodland-trustThe Woodland Trust, and Confor, the confederation of forest industries, a UK trade organisation,  argued that England, which has just “10% forest cover compared to the 38% average for Europe,” is suffering from deforestation. The government, which is committed to planting 11 million trees over the next forty five years, planted only 700 ha of the 5000ha needed to meet that target in 2015. The Woodland Trust, who are themselves planning on planting 64 million trees over the next ten years, argue more needs to be done to protect ancient woodland and hedgerows “which are being lost to roads, quarries and housing”. There is currently no national logging of this loss with the trust relying on a network of volunteers to spot negative impacts at the local planning level. The trust argues that these woodlands and hedgerows need to be buffered, extended and connected to other woodland through gaping and hedgerow maintenance if the government is to meet it’s own targets.

Protecting Forests Through Global Supply Chains

traseAS I am in Morocco and it is just a bus and a train ride down from the mountains, so not a big load onto my carbon footprint, I made a brief visit to the COP22 in Marrakesh last month. I went primarily to attend the climate law and governance day but I also attended the launch of TRASE ( Transparency for Sustainable Economies) a new online open-access tool that uses publicly available data to unravel supply chains and reveal the origin of commodities such as soybean, beef, palm oil and timber.



One of the ten most important agricultural crops soybean production reached 336 million tons in 2015 making it the 7th most important crop globally. The USA, the largest single producer, was responsible for 118 million tons (35%) whilst South America (Brazil, 102 million tons, Argentina, 57 million tons and Paraguay, 10 million) was responsible for 169 million tons (50%) of global soybean production. With Canada’s contribution (2%) the America’s are responsible for 87% of the Worlds soybean production; much of it however on deforested land. The TRASE platform addresses this through the “use of trade and customs data to identify the producers, traders and transporters involved in the flow of globally-traded commodities” to bring transparency to the global supply chain so that business can identify commodities originating from deforested land.

Palm Oil

The situation is mirrored in Indonesian where deforestation for palm oil production has put the orangutan on the critically endangered list. The clearing, draining and setting alight to the peat of the Indonesian swamp forests in preparation for palm oil plantations in 2015 further led to 100,000 deaths across Asia from the thick belching smoke. Releasing over a billion tons of CO2 in the process and pushing Indonesia into 4th place behind the USA, China and India as the Worlds leading greenhouse gas emitter. [Costing the Earth]theforesttrust

The Starling project, a collaborative venture between The Forest Trust, Airbus, and SarVision uses “high-resolution optical satellite and radar imagery to monitor forest cover in real time” and provide the tools “to enable companies to provide evidence of how they are implementing their No Deforestation Commitments.” As with TRASE the Starling project seeks to provide transparency in global supply chains; supply chains in which just four products, beef, palm oil, timber and soybean are responsible for two thirds of global deforestation . In addition to habitat and species loss the deforestation undertaken to grow these crops is responsible for over 10% of global greenhouse gas emissions.


Another major deforestation crop is Cocoa, produced by just six million smallholders worldwide it is a global crop controlled by less than a dozen companies. In West Africa, the source of 68% of the Worlds chocolate and home to four million cocoa farmers, cocoa is the principal cause of deforestation.

Logging and Land Tenure

Subsistence farmers are blamed for much of the remaining deforestation and whilst they undeniably contribute, in a recent report from the Congo researchers identified logging and land tenure rather than farming as the principle cause of deforestation. Tropical forests are the most diverse ecosystems on planet Earth, they store and clean water, influence the climate and act as large reservoirs for carbon that has accumulated over the lifetime of the forest. When the forest is removed that diversity is lost, the ground dries up and the carbon stored in the forest and it’s soils is released. Over 65% of tropical deforestation and 7% of global carbon emissions now result from the cultivation of less than a dozen crops.

A Carbon Neutral Future From Forests?

Old Forests, be they ancient English woodland, Indonesian swamp forests, or Amazonian rain forest are all bigger carbon sinks than what can be captured in new plantations. It’s likely that for every hectare of old forest felled two hectares or more of new forest are needed to offset the carbon released. New forests that similarly don’t have the diversity or provide the habitat of the ones lost.

In a recent Inside Science program it was claimed that to reach a negative carbon balance; where carbon captured as biomass is used to fuel an energy plant, and the CO2 produced is then captured and stored; would require an area of land 1-2 times the size of India (3-6 million km2) to grow the biomass. The idea is somewhat over complicated and risks turning captured carbon into the climate equivalent of nuclear waste as we struggle to store billions of tons of CO2. If instead that energy was produced by other carbon neutral sources (i.e. solar, wind and water) and the biomass grown as both the carbon capturing and carbon storing device, we would not require a network of silo’s storing ‘dry ice’ or it’s equivalent, but the World would still need to plant a forest the size of India to capture and store the carbon the 20th century has released.

Restoration and Regeneration

Whilst stopping deforestation completely would be the best course of action, restoring recently deforested land would result in capturing more carbon than planting a new forest on agricultural or marginal land. The Amazon has lost 20 %, one million km2, over the last 40 years and whilst it is unlikely that all that loss can be recovered, with strategic planting perhaps 20% of what has been lost could be recaptured. If the same strategy could be applied to West Africa, the Congo and Indonesia; perhaps as much as ½ million km2 of tropical forest could be restored within 50 years. It is though just 10% of what is needed. If Europe were to similarly increase it’s forest cover by 10% then that rises to 20% of what is needed but again will take 50 years to reach fruition.


The UK has 450,000km of hedgerow but ,as a consequence of the plough up policy of 1948, has lost 121,000km.  Gap filling the existing and reducing field sizes to recreate the hedgerows lost could increase the UK’s wood cover by as much as 5%; more than meeting the UK governments target of 2% over the next forty five years.

Whilst deforestation and afforestation are the key issues in Africa, South America and Asia, hedgerow planting could similarly contribute the equivalent of a billion km2 or more of new forests in agricultural regions of the tropics. In many instances planting could encourage diversification with trees and shrubs for the production of fruit and nuts, oilseed for bio-fuel production, biomass for energy or trees for soil remediation and erosion prevention. Such diversification provides commercial value and resilience as well as contributing to climate mitigation.

Monitoring Restoration

Whilst projects such as TRASE and Starling are providing the tools for businesses to identify commodities originating from deforested land or to verify no deforestation commitments, there needs to be additional tools to further monitor and measure restoration and afforestation strategies. If deforestation was to end tomorrow, the World would sill need to create 5 million km2 of new forest and woodland if it is to meet its commitment to prevent global temperatures rising above the 2 degrees C threshold agreed in Paris. It is not enough just to arrest the damage, we must repair it too.

However there are no supply chains, no customs or logistics data to mine to see if a forest or a hedgerow has been replanted or is being maintained. Satellite data, a resource the Starling project has utilized to reveal changes in forest canopy in palm oil production could similarly be utilized to monitor afforestation efforts. An approach that applies as much to reafforestation strategies in the UK as it does to the tropics. However monitoring is only half the story and whilst carbon sequestration and habitat creation are important global functions of trees they are not their sole function. Trees are a commodity, we need the wood but they also perform other crucial functions within the local environment, be it improving flood defences, arresting soil erosion, removing air and noise pollution or just providing beauty and enjoyment; trees are an integral part of the human landscape.

Human Beings and Climate Mitigation Strategies

The first principle of the Rio Declaration was that “human beings are at the centre of concerns for sustainability”. [Earth Summit I, 1992 ]. Keeping human beings at that centre should similarly be the first principal of all future environmental and climate mitigation strategies. The people whose lifestyles need to change need to be involved in that change; for whilst human beings are at the centre, they are also the cause, our environmental and climate crisis is a problem of our our making and it is within our own humanity that the solutions similarly lie.

The world has similarly changed considerable since 1992, there was no remote sensing, the internet had not yet gone ‘viral’ and Glastonbury would have been regarded as the centre of the Gig economy. However somethings have not changed; we have continued to lose habitats at an alarming rate and have bought more species to the brink of extinction over the last twenty four years than since the demise of the dinosaurs. If we continue on this trend for another twenty four years there may well be nothing left to save.

The Satellite’s Eye: Remote Sensing

The Forestry Commission’s Corporate plan for England identified it’s priorities for English Woodland as to “protect, improve and expand”. Seeking to “bring two thirds of all woodland under management by 2018” and to create a total of 2600km2 of new woodland by 2060. The plan further commits to provide “support for mechanisms and payments for ecosystem services” and calls for “more trees and woodlands in and around towns and cities.

In order to meet those targets the commission needs to know the real time state of Woodland across England. As only 57% of England’s woodland is currently sustainably managed and the commission has prioritised bringing only 2/3rd’s under management by 2018, a significant proportion of Woodland will remain without any mechanism to assess threats to it for the foreseeable future. Unprotected it is, as the Woodland Trust identified, “at threat of being lost to roads, quarries and housing” as well as to the disease and climate threats the Forestry Commission prioritise in their plan. With no mechanism in place to record these losses the trust has resorted to relying on a network of volunteers to spot negative impacts at the local planning level. However with the advent of satellite imagery and the internet the ability to log and record existing woodland remotely and to similarly record any impending changes to it now exists.

All land use management strategies, at both the planning and monitoring stages rely on maps. Maps that can be greatly enhanced with the use of satellite imagery. In some instances, as with the Starling project those images are used to identify deforestation and species changes in tropical forests; it is a specific task that can and is performed by machine learning, but in many others, particularly mapping trees in complex environments such as cities and towns, the process still requires human intervention.

Crowd Sourcing The Map: A Place To Plant Your Tree

AfSIS (African Soil Information Service) and their partner QED  have been using satellite imagery to map land use in Africa for the last year. The process relies on volunteers, ‘citizen scientists’ to annotate images to identify buildings, cultivated land and forestry in a 250m2 grid. The same method could be used for England, if not the whole of the UK, to map the current state of woodland and land use in general [Mapping UK Habitats] Done correctly such a map could not only identify existing woodland by when interpolated with other data such as soil maps, hydrological or species distribution could identify where the benefits of planting new woodland and restoration of existing woodland can be best realized. The same map could also be used to monitor the health and, with weather maps, predict the movements of threats to woodland from pests or diseases.

A single map into which all environmental data can be interpolated so as to give a complete and accurate reflection of the state of the environment at any scale and to any stakeholder who needs it. [DFM]

Such a map would be the means to calculate the quantitative and qualitative benefits and cost of a given action to the environment. To build such a map at a working resolution requires a large network of volunteers, the same network that will later be required to monitor and update the map. It may be possible to develop machine learning but in the interim, and to give the machines something to learn from, the map requires human input. That input similarly performs another crucial function; it engages the very people it needs to change.

Citizen scientists have and continue to contribute to many existing mapping projects but in reality relying on volunteers to create and maintain critical environmental maps in order to meet our climate objectives is perhaps a policy that is as likely to succeed as relying on governments to voluntarily abide to environmental agreements. It is doomed to failure for there are not enough volunteers to provide the level of coverage needed to map and then monitor global land use and climate mitigation strategies to the extent required to achieve COP22 objectives.

The Gig Economy: A Tree Hugger’s Paradise?

Paying stakeholders to both maintain and monitor climate mitigation strategies is likely the only viable way of bringing about the level of change and monitoring that is required to meet our climate objectives. If we are to plant a forest the size of India we will need a lot of spades to do so; In this respect the gig economy may well come to the rescue for we do not have a lot of time in which to raise the number of tree huggers needed to map, plant and monitor the planet’s forests and woodland. In England an army will be needed to both identify and plant the 13000km2 of land required for the 64 million trees the Woodland trust aim to plant. Five times more ambitious than the Forestry Commission, if successful it is a plan that would double the UK’s woodland cover but it is similarly a plan that will need a radical new approach to be successful.

Carbonizing the Blockchain

The block chain and the concept of smart contracts makes paying a large number of citizen scientist to analysis and annotate satellite imagery possible. It further offers the potential to automate payments to farmers and landowners for carbon mitigation efforts. Using smart contracts carbon credits could be earned by farmers and landowners using quantifiable metrics that provided payments over an extended period; a trickle system that ensured trees and plantations are not only planted but maintained to a standard that achieves the environmental and climate objective. A three point verification system between the farmer, the satellite and the gig economy. Such a mechanism is not limited by scale , an individual planting a single tree in a garden or a landowner planting a 1000 ha stand; each would be paid according to the benefits achieved by their efforts.

Developing New Technology 

Whilst much of the technology already exists, the satellite images, the software to annotate those images and the smart contracts to pay the cartographers and monitors: new apps for phones that allow for newly planted trees to be quickly recorded and uploaded along with the gps co-ordinates need to be created. Satellite imagery could then be used to confirm the work and machine learning used to estimate the canopy cover and the amount of carbon sequestered. This could be performed over the lifetime of the tree or stand and used to adjust and refine payments to ensure carbon mitigation strategies are similarly maintained.

The use of new technologies such as blockchain, machine learning and remote sensing offer real opportunity to make climate mitigation strategies a reality that works at the local and global scale. The World though has a tough hill to climb, not only must it plant a forest the size of India but it must similarly cease from decimating the existing forests. It must also find a solution to our addiction to fossil fuels, as we to our consumerism, as it is these habits that are the root cause of our environmental and climate woes. There is though no such thing as a free lunch or a technological fix for greed; each and everyone of us needs to reduce our own personal carbon footprint, our own personal consumption of fossil fuels and forest products: for this is not only the most effective way but similarly essential if we are to mitigate climate change and arrest deforestation.

Citizen Science on the Map

Putting Citizen Science on the Map

Citizen Science

Crowdcrafting, a web-based service that invites volunteers to contribute to scientific projects, defines Citizen science as “the active contribution of people who are not professional scientists to science. It provides volunteers with the opportunity to contribute intellectually to the research of others, to share resources or tools at their disposal, or even to start their own research projects. Volunteers provide real value to ongoing research while they themselves acquire a better understanding of the scientific method.”

As definitions go it is though a little unfair, for it somewhat trivialises the importance of the citizen scientist in achieving Sustainable Development Goals. We are in this together. If any has more importance than another then it’s the Citizen Scientist, it is the amateur, not the professional who is pivotal to our sustainable future. Without the citizen scientists most projects would be impossible or take decades to achieve a fraction of what the crowd can do in a matter of months. It is only because of the collective effort of the crowd that much of this data can be gathered and or analysed. With more and more data being generated then this volume is set to grow exponentially. As it does so the importance in recruiting citizen scientists will similarly grow, for without citizen scientists all efforts to turn this blue planet green will be little more than academic exercises: the half hearted efforts of fools and sycophants holding summits in capital cities.

Desk Top and Field Work

For Citizen Scientists the highs come from taking part and seeing completed projects, projects that can be divided into two distinct types: desktop, where citizen scientists look at images/data on a screen and then answer questions or annotate the image in some way; and field work, where citizen scientists take an active part in providing the data. Examples of desktop studies would be the analysis of satellite or drone images, the identification of insect or plant specimens, or the measure of lichen growth or plant populations. Examples of field work projects would be Open Street Map, mapping seaweed distribution along the UK’s shoreline or insect distribution inland.

There is though an ever growing number of citizen science projects that one can now get involved with. Below are links to some of the larger collections and larger projects. Most are built using open software platforms with those platforms similarly providing a platform upon which to promote the projects. They encourage anyone and everyone to create a project and so there are opportunities to volunteer ones services for everything from the sublime to the ridiculous. The last three sites provide search facilities to find more projects but this is in no way a complete list, for the complete list needs it’s own dedicated site.

The Open Street Map Project

Whilst Google’s Map, Earth and Street View projects are great applications they are similarly subject to copyright restrictions. Google owns the resources and whilst it currently allows ‘anyone’ to use them it retains the option to change its mind at some point in the future. An eventuality made more likely should google ever become the world’s sole cartographer; fortunately there’s the Open Street Map project.

osm2The open source alternative to Google, OpenStreetMap is the largest and most contributed to citizen science project to date. Supported by the OpenStreetMap Foundation the OSM project is “an initiative created to provide free geographic data, such as street maps, to anyone”.  A not-for-profit organization maintained by donations and membership fees (You can join here) the OSM foundation supports rather than controls the development of the OpenStreetMap Project. Over the last ten years, the OSM foundation has mobilised more than 2 million volunteers to create free open street maps across the globe. If you are looking to get involved and want to find a project in your area I have created a twitter list of all the OSM projects and resources. The list current contains over 50 country and city projects as well as resources to support OSM groups; if you know of anymore please advise.



ZooniverseZooniverse declares itself as “the world’s largest and most popular platform for people-powered research. Research made possible by hundreds of thousands of volunteers from around the world who come together to assist professional researchers in work that would not be possible, or practical, otherwise.” With just 46 projects in 11 categories (art, biology, climate, history, language, literature, medicine, nature, physics, social, science, space) that might seem like a hollow boast but Zooniverses software is used by some of the biggest scientific data projects in the World. This professionalism comes through in the projects Zooniverse showcases which are well documented and fun.


crowdcraftingIf Zooniverse is the biggest then Crowdcrafting, a “web-based service that invites volunteers to contribute to scientific projects developed by citizens, professionals or institutions”, with over 400 projects in 6 categories is the broadest. As with the previous two projects crowdcrafting uses and develops open source software to “help solve problems, analyze data or complete challenging tasks that cant be done by machines alone.” However it is non too discerning about the projects it promotes and some, whilst valid are perhaps a little too esoteric for most. If though odd and obscure is what you are looking for then crowdcrafting is a good place to start looking.

NASA Earth Observatory

nasa terraNASA’s Terra satellite, which they describe as being about the size of a school bus and falling about the Earth on a circular sun-synchronous polar orbit carries five imaging and monitoring instruments to aid understanding air and water processes and quality. There’s not too much I can say about this project after the successful planting of a mind gif containing a yellow school bus slaloming it’s way around the planet on an endless loop. If you live in the States and that idea doesn’t make you dizzy then you could help NASA by doing some essential ground measurements and observations in air and water quality to support the work of the school bus falling above you.


geosurveyJust as the data from the Earth Observing Systems of NASA’s Terra Satellite is being used to map air and water quality in the USA, so data from the European Space Agencies Sentinel-2 satellite is being used to map Africa’s soils and habitats. Geosurvey in conjunction with the African Soil’s Information Service are currently running 30 desktop land use projects in over 10 African countries. Whilst geosurvey is built on the open source web framework Django the land classification survey is unique in that it has all been built in house. The net result is perhaps the most user friendly interface that I have so far come across. Geosurvey’s simple clean approach makes it a good place to begin citizen science mapping work.


The Natural History Museum

bigseaweedThe natural history museum have a mission to digitize their collection of 80 million specimens. They are similarly one of the big data collection holders who use Zooniverse software to harness the efforts of citizen scientists to digitize those collections. Current projects range from transcribing microscope slides from the comfort of your home to visiting the coast to map the distribution of seaweed or report whale and dolphin stranding. The natural history museum have eight projects currently in progress and provide a huge amount of support and educational material around those projects. If you want to get really active and turn trips to the seaside into science expiditions then the natural history museum is the place to go.

Centre for Ecology and Hydrology


As with the Natural History Museum the Centre for Ecology and Hydrology uses Zooniverse software to power their citizen science projects. Taking further advantage of smartphone technology the CEH has been mapping the distribution of more than 10,000 insect species in the UK and in other projects giving volunteers a backpack containing a particle monitor and a GPS logger to monitor their personal exposure to air pollution. The Centre for Ecology and Hydrology also offer guides on best practice and designing and executing citizen science projects. If the natural History museum’s citizen science projects are designed for kids then the centre for ecology and hydrology citizen science projects are designed for their grown up counterparts, those who can’t stop fiddling with their smart phones, drones or wearable tech.. now you can do it all in the name of science.


British Geological Society (BGS)

mysoilThe BGS have been using crowdmap for amateur geologists to record observations and share images about temporary geological exposures for a number of years. The BGS also manage the mysoil app which allows citizen scientist to record soil properties and upload to the BGS database to refine the UK’s aging legacy soil maps.

RSPB Birdwatch

The RSPB (Royal Society for the Protection of Birds) have been campaigning for the protection of birds for over 100 years. Conducting surveys amongst their members for over half a century they are perhaps the earliest adopters of citizen science and continue to be a major contributor with national birdwatch surveys. On the other side of the Atlantic ornithologically based citizen science projects are run by the Cornell Lab of Ornithology.




If you haven’t found the perfect citizen science project to get involved in through any of the above links then Scistarter might be able to help. Scistarter has a mission to “bring together the millions of citizen scientists in the world,” a mission objective I share and support. It does however and despite being in receipt of a Mark Shuttleworth foundation fellowship/flash grant, have one of the most retro looking web sites I’ve seen in years; a real blast back to 1995.. It’s a great concept but it needs some serious web designing on top and some lateral thinking underneath.



Scientific America

Scientific American has a similar but somewhat smaller mandate than SciStarter. A little more discerning they want volunteers to “Help make science happen by volunteering for a real research project”. What qualifies as ‘real research’ is perhaps a little subjective however that said they do give citizen scientists the opportunity to engaged with 240 of what they determine to be real projects (i.e. coming from subscribing institutions). That’s not to say these projects are not valid, they are, they are as real as any that  Scientific America feels it has the authority to say are not.


An EU funded initiative Socientize received 710,000 euro to fund a 2 year project with the “aim of creating a common forum for cooperation between e-Infrastructure providers and citizen science infrastructures providers.” Four years after receiving that funding Socientize have five projects listed and a simple list of links to other projects (most of which can be found through the previous links above). Whilst I don’t doubt that this was money well spent the results of the investment are not immediately apparent through the web site.

Paid or Voluntary

Whilst there are a great variety of projects that need the help of citizen scientists to be found in the above links nearly all have one thing in common; non of them pay citizen scientists for their efforts.  Some, such as the the geosurvey website award points which are later converted into chances to win a prize. But I have yet to come across a site that pays even a nominal sum to citizen scientists for their efforts. It’s likely that millions of contributors to thousands of scientific projects to date have not even received an acknowledgement. The crowds contribution is both anonymous and voluntary but the researchers is likely neither. Whilst it’s not a fair deal on the citizen scientist it similarly shortchanges the research project. Paying citizen scientist would result in greater participation, greater accuracy and earlier completion. It’s also extremely cheap as a draft paper I created explains pay to map

A Common Framework

Whilst all the projects above conform to a single voluntary principal with respect to citizen scientists none make any real effort to conform when it comes to categorizing projects. None make it particularly easy to find or organize projects according to factors such as execution (desktop/field), geographical location, study subject or research objectives. In most cases the research objectives are not even shared suggesting that citizen scientists are either disinterested in the objectives of the projects they freely give their time and effort to, or that researchers are disinterested in the very people they want and need to engage. If researchers only looked after the citizen scientists.. then the citizen scientists would complete their project.

The Future of Citizen Science

Citizen Science, crowd mapping and collective distributed actions are unlikely to be passing fads. The technological and data driven path we now find ourselves on is not one we can choose to simply step off. Whilst individually we have the option to not engage and act only in a selfish capacity, as a society we don’t. It’s a whole new paradigm.

To take advantage of this and reap the collective rewards citizen science sites need to be highly functional. A dashboard that allows a visitor to easily see mapping projects according to geographical relevance, desktop or field, study and research objectives, and organized into environmental, social and economic categories. A dashboard that is interactive and visual.

Projects need to provide detailed information on the nature of the study, where is it based, who is running and who is sponsoring it. Once the study is finished if not before citizen scientist should be given access to the project data so that they can see the fruits of their labours.

Consideration should be given to identifying projects suitable for education of school children. To fully engage schools in citizen science projects additional educational material to assist teachers in using and contributing to projects as part of or in sympathy to the curriculum should be provided.

There is no such thing as free lunch that isn’t stale or lacking in flavour. To make and keep people more engaged a nominal payment system that allows citizen scientists to earn convertible value, such as a crypto currency, should be introduced. The payment does not need to be substantial but sufficient to allow an industrious individual to earn up to a $1 an hour. This would make participation attractive to groups who have the time and access to an interface but would not otherwise engage. Agency and zero hours contract workers, the unemployed, homeworkers, disabled, carers and the retired would all be more likely to engaged if they earned a nominal amount for their efforts. Paying contributors I would aver would result in higher recruitment and faster individual work rate shortening study periods considerably. Time and funds previously wasted on promotion can be reclaimed as the prospect of reward encourages viral promotion and longer participation. Overall paying contributors would likely result in lowering overheads and improving efficiency.

It’s a whole new Paradigm is this Citizen Science, Data Driven, Sharing, IoT Society Thingy…