Green is the new grey

The benefits of trees are well documented: they improve air quality, absorb carbon dioxide, provide shade to cool streets and homes, mitigate storm water runoff, increase property values and improve our quality of life. Yet, urban forests continue to steadily decrease in significant numbers. A study published last year by the US Department of Agriculture’s Forest Service found that the US is losing four million urban trees per year.

However, over the past decade, politicians around the world who are aware of the economic, environmental and emotional benefits of trees have begun launching ambitious initiatives designed to add more green to the urban grey by planting trees. In the US, Los Angeles, New York City, Denver, Salt Lake City, and 13 counties in Pennsylvania, New Jersey and Delaware all aim to plant one million trees in their cities. London in the UK and Shanghai in China also have similar goals. â©

However, with these well-intentioned missions comes the million-dollar question ‘How can cities achieve this goal?’ and the equally important, second question: ‘Is this goal even achievable with the existing urban space?’â©

Know your canopyâ©

“If you don’t know how much tree canopy you have and you don’t know how much room you have to plant trees, it’s quite difficult to set a tree-canopy goal,” says Jarlath O’Neil-Dunne, director of the University of Vermont’s (UVM) Spatial Analysis Lab. “It’s equally difficult to prioritise tree-planting efforts.”

Since 2006, O’Neil-Dunne and a small team of scientists at UVM have been helping city officials answer those ‘how much’ questions by integrating spatial data such as satellite imagery, LIDAR, aerial photos and GIS datasets into image-analysis software to produce urban tree canopy (UTC) assessments. The UTC information indicates how much tree canopy exists, where the trees are and what space is left in which to plant, enabling planners to set realistic goals for increasing their green infrastructure, and develop targeted campaigns to maintain the canopy they have. It was this kind of work that helped to launch New York City’s MillionTreesNYC initiative in 2007.â©

“When you think of New York City, you think of Manhattan with lots of concrete and skyscrapers, so we didn’t know if there would be adequate planting space for more trees along the city streets or in park land or on private property,” says Jacqueline Lu, director of GIS and analytics at the New York City Department of Parks and Recreation (NYC Parks). “The UTC data showed us that it was possible and gave us the direction we needed to prioritise activities to meet the goal.” â©

Integrative and accurate UTC datasets such as these have been giving communities such as those in Baltimore, New York, Pittsburgh and Washington DC the ability to answer the core question of how much tree canopy they have and the more sophisticated, targeted questions such as: ‘Where should we plant to improve our carbon footprint?’ or ‘Which areas are most cost-effective for cooling the air or improving the water?’ It’s a new perspective that is providing the definitive means to develop prioritisation strategies to meet city goals, be they for improving tree-canopy inequalities among neighbourhoods, urban renewal or storm-water management. â©

Sprouting the UTCâ©

O’Neil-Dunne first began developing UTC analyses in 2006, when he worked with the USDA Forest Service and Maryland Department of Natural Resources to help map the urban forest of Baltimore as part of a large ecosystem study. Originally, he was given existing land cover data, derived using standard pixel-based processing software, which estimated the city’s existing tree cover at 20 per cent. Based on that assessment, city leaders set a goal in 2007 of increasing the city’s tree canopy to 40 per cent by 2037. However, in practice, more detail was needed to create effective tree-planting strategies.

“With standard, pixel-based methods, you identify large patches of forest,” says O’Neil-Dunne. “To make a well-informed, urban forestry plan, decision makers need to readily see the natural resources of the urban landscape down to the property parcel level. Object-based image analysis allows you to clearly distinguish different vegetative types, including individual street trees, enabling you to see how much is planted, where it is planted and how much more could be planted.”â©

In 2008, the UVM team performed a second assessment with the addition of LIDAR imagery and Trimble’s eCognition object-based image analysis (OBIA) software. Baltimore’s tree canopy rose to 27 per cent.â©

“The city gained seven per cent canopy cover just from better data, better software and better data analysis,” says Morgan Grove, a research scientist with the USDA Forest Service. “More importantly, parks and planning departments now had the essential detailed map of the city’s urban forest and the visual means to plot a path to achieving the 40 per cent goal.”â©

Since that initial partnership, word spread of the usefulness of the UTC assessments, leading more municipalities to begin sending their geospatial datasets to O’Neil-Dunne for their own tree-canopy analyses. To date, his team has completed more than 70 UTC assessments for cities across the US and Canada.â©

Tree report cardsâ©

The UVM analysts are typically given a combination of geospatial imagery, such as high-resolution aerial photos, satellite images and LIDAR data, and GIS datasets such as roads and buildings. They incorporate the imagery into eCognition and create customised workflows to automatically classify specific vegetative types and environs, which are then mapped according to user specifications.

“A unique feature of eCognition is that you can teach it to consider an object’s contextual and relational surroundings,” says O’Neil-Dunne. “For example, you can classify trees based on their geographic position – trees only thrive in certain locations.” â©

Each client usually receives a report detailing its tree canopy percentage, a complete land cover map and the customised UTC GIS datasets, enabling users to examine and view the tree cover down to individual neighbourhoods or the entire city.â©

Stemming the storm water in Baltimoreâ©

At last count, Baltimore has 2.8 million trees and an estimated urban tree canopy of 27.4 per cent. However, the city’s impervious surfaces – streets, buildings and car parks – comprise 43 per cent and that imbalance is problematic, particularly for managing storm-water runoff. Without the abundance of trees to absorb rainfall, polluted storm water flows directly into the Chesapeake Bay, the largest estuary in the US.

With the UTC assessment, leaders were given the base layer upon which to build a greener infrastructure to address surface runoff as well as other urban renewal improvements. And as city budgets are tight, the dataset also allows them to focus on maintaining the urban forest they have. â©

In 2012, foresters and volunteers planted 8,200 trees, bringing the total number planted since 2009 to 27,000.â©

“Simply put, you manage what you measure,” says Grove. “The UTC data is a complete census of the landscape that you can integrate with other city data layers. You can start resolving questions like, ‘Where can we plant trees to reduce storm water runoff?’ or ‘Which tree areas are in greatest need of maintenance?’ Then you can begin to prioritise your efforts to make the most effective use of limited funds.”â©

Urban renewal in Pittsburghâ©

The compelling as-is view of Pittsburgh’s UTC map, coupled with its land-use metrics, has also been a boon for Tree Pittsburgh, a local advocacy organisation, and its committed campaign to extend the city’s tree coverage by 20 per cent in 20 years.

With an estimated UTC of 42 per cent, it would seem that Pittsburgh and its 2.6 million trees provide a rather commendable green landscape. Indeed, Tree Pittsburgh reports that the city’s urban forest provided more than $7m in savings in 2011. However, that 42 per cent is misleading, says Danielle Crumrine, Tree Pittsburgh’s director.â©

“Large swaths of our trees are on steep hillsides, which is not where our residents live,” she says. “So they’re not shading our homes or pulling storm water off the streets. That’s an issue of imbalance that the UTC data, as part of our urban forest master plan, is helping us to resolve.”â©

Incorporating the UTC analysis with other tree-related data, Tree Pittsburgh created an urban forest master plan in 2011 to serve as a roadmap to manage and grow the city’s tree canopy. With such a tool, Crumrine and her team have transformed their former plant-by-request model into a pre-planned, targeted approach to address tree inequities, and energise residents and businesses to help blanket the city in green.â©

“The UTC gives us direction and a way to prioritise our urban renewal projects,” says Crumrine. “We’re able to look at the distribution of trees, identify patterns – poor neighbourhoods have fewer trees – and devise strategies to improve them.” â©

Using that proactive model, Tree Pittsburgh launched a pilot project last year to address underserved communities. They used the UTC dataset to identify the 25 lowest canopy neighbourhoods, and with additional funding by the Pittsburgh Shade Tree Commission, teams inventoried 23 parks, pruned 35 trees, removed three trees and planted 80 trees. The project will continue in the fall, and Crumrine plans to use the same approach in other areas. â©

A million trees in NYCâ©

Sprouting a new tree-planting model has also been one of the significant benefits of the UTC analysis for Lu and her colleagues at the New York NYC Parks.

“Before MillionTreesNYC we were funded at a level where we could only fulfil individual tree requests from the public,” says Lu. “With MillionTreesNYC, we receive so much funding for planting that responding to just tree planting requests would never help us reach our goal. The UTC data enabled us to change to a block-planting model, so we can target an area and plant it out completely. It’s much more effective.” â©

Indeed, 820,330 trees have been planted across New York’s five boroughs since the MillionTreesNYC initiative began, nearly guaranteeing the city will reach its one-million-trees goal by 2017. â©

Although the MillionTreesNYC programme is well funded, planners and foresters have needed to be smart about how to prioritise workloads and put the business of planting both street trees (at an average cost of $1,450 each) and reforesting parks into practice. The tree canopy data has formed the basis for a prioritisation framework to identify tree deficits and provide direction for where teams should dig first. Then they implement customised outreach initiatives and tree-planting events to bring in volunteers.â©

“For New York, the resulting land cover dataset, including the tree canopy assessment, was more than 90 billion pixels: that’s seven times more data detail than in the USGS’ entire 30m land cover map of the US,” says O’Neil-Dunne. “That level of land-use detail, down to a six-foot-tall tree, not only enables the city to make tactical tree-planting decisions, it gives it endless possibilities to further study the relational benefits of trees across the board.” â©

The city of treesâ©

Decisions rooted in trees is what the advocates at Casey Trees (CT), a local non-profit organisation, is striving to achieve through its own canopy goal of 40 per cent. Founded in 2001, CT’s focus has been to restore, enhance and protect the tree canopy of the nation’s capital, a mission that has not been easy. Based on the UVM’s 2011 UTC assessment, Washington’s tree coverage is 36 per cent, but its impervious surface coverage is 41 per cent. â©

The extent of the hard surface in the district began to pose enough of a issue in managing storm water that the US Environmental Protection Agency imposed regulations requiring it to adopt measures to reduce rainfall runoff. Spurred by that decree, and by CT’s canopy goal, the mayor launched the Sustainable DC Plan in 2008, which included achieving a 40 per cent tree canopy by 2035.â©

To achieve that, existing trees need to be preserved and 216,300 new trees must be planted. Aided by the UTC analysis data, CT in partnership with government groups have planted more than 30,000 trees since 2008, and devised maintenance strategies to protect its existing green landscape.â©

“Informative tools like the UTC data are enabling us to keep a pulse on our patient,” says Mark Buscaino, CT’s executive director. “Most importantly, they give us an effective way to educate, plan, secure investment and measure our progress.”â©

With detailed, accurate datasets such as UTC assessments, planners and foresters can get to the root of their canopy problem, devise efficient, tactical greening strategies to better the canopy, and possibly nip future issues in the bud. After all, the best time to plant a tree was yesterday. â©

If you don’t know how much tree canopy you have, and you don’t know how much room you have to plant trees, it’s quite difficult to set a tree-canopy goalâ©

Mary Jo Wagner is a writing and editing consultant and contractor ([email protected])â©