We sit down with Bruce MacArthur, President and CEO of remote sensing and forestry industry leader, Tesera.
Hello Bruce. We're excited to announce the partnership between UP42 and Tesera! Can you let us know a little more about Tesera as a company? When and why was it founded?
Tesera Systems Inc. is an employee-owned company that is based in Calgary, Alberta, Canada. We operate remotely from home offices across Canada and internationally. Since 1997, Tesera has been delivering innovative IT and geomatics solutions by designing and developing advanced GIS and modeling and web-based applications to serve client needs.
We are an agile company supported by an innovative and multidisciplinary team of information technology and GIS specialists, data scientists, systems analysts, engineers, and environmental resource specialists. Our diverse team encompasses wide-ranging backgrounds and domain expertise.
We believe inspiration and innovation begin with people + data + technology. Using unconventional and collaborative thinking, we build cost-effective, agile, easy-to-use software and solutions to improve business outcomes and sustainable development framework.
We're curious. Where does the name Tesera come from?
Tesera's name is derived from the word "Tessera," which is the tiles that comprise a mosaic. This reflects our interdisciplinary team of specialists and corporate culture of shared values. Our name is also based on "Tesseract," a four-dimensional representation of space (x,y,z) and time (t) which reflects the spatial and temporal nature of the solutions that Tesera offers in the marketplace.
The company primarily focuses on forestry solutions. What's the importance of forestry solutions in today's world?
With the amount of exploitation pressure forests have been under worldwide in the last few decades and their increasing importance in easing the climate and biodiversity crisis, forest management (passive and active) requires comprehensive, reliable, and scalable inventory data. While there are many remote sensing solutions for estimating forest cover and detecting changes or disturbances to existing forest cover, there is a very real need for solutions that provide detailed information on tree types, ages, volumes/biomass, and other more nuanced metrics important for sustainable harvest and regeneration planning and management.
Over the last decade, we have pioneered the development of commercially available LiDAR and airborne imagery-based semi-automated forest inventories in Western Canada, working with some of the world-leading academics and industry players in this big spatial data analytics and cloud computing space to improve access to highly reliable, comprehensive and scalable forest inventory attribute information. Access to highly accurate estimates of forest stand structure and composition is fundamental to ensure forest managers can plan and manage forest landscapes sustainably over time and space.
Climate change represents a major social, economic and environmental challenge and existential threat to the state of forests across the globe. Generally speaking, conventional forest inventory solutions are unreliable, unscalable, and are low resolution, which undermines efforts by forest managers to design and implement sustainable forest management to help realize multi-objectives across forest landscapes.
Healthy forest ecosystems are critical to the well-being of humanity and the ecological integrity of our planet. There is a need for higher resolution forest inventory data and easier to use forestry solutions which is why we created our suite of High Resolution Inventory Solutions (HRIS) to help move the forestry industry forward.
We're really excited to be able to offer the Land Lines Segmentation block to UP42 users. What value do you and Tesera see in partnering with UP42?
UP42 provides access to satellite Earth observation data and easy-to-use tools that help to bridge this forest inventory data and information gap, particularly with advancements in cloud-processing and ML technologies. In the last few years, we are starting to deliver these solutions more rapidly and cost-effectively over larger and larger areas, which is really exciting for the future of sustainable forestry and the well-being of our forest ecosystems and the planet.
We have found that satellite data and other remote sensing and geospatial data are really spread out across a number of providers and platforms, all with different degrees of difficulty to obtain and stream into analytics processes. What we really like about what UP42 has done is putting these big datasets into one place with common building blocks to interface with the data for our use cases.
I think the way UP42 has set up their platform to allow geospatial developers to bring their analytics to the data, instead of the other way around, is a real game-changer to lowering the entry barriers for end-users just looking to see what is possible with this kind of global Earth observation datasets.
And what is the role of the Land Lines Segmentation in that solution? What are the main use cases of that solution on UP42?
So the Land Lines app came as a sort of natural evolution of the image segmentation algorithms we initially developed to automatically delineate forest stand boundaries using LiDAR and high-resolution aerial imagery to create even-aged stand polygons for use in traditional forest timber inventory management.
This automated process we developed replaced the very time-consuming task of trained foresters creating stand boundary polygons using manual image photo-interpretation to visually group stands of trees based on height, age, volume, density, and tree types.
When we began getting requests from forest managers to update their high-resolution inventories for natural or man-made depletions, we needed to find a more cost-effective solution than LiDAR for very large inventories, which are typically required on an annual basis.
The process we developed for automated forest stand segmentation was then adapted to analyze satellite imagery to delineate vegetation changes and disturbances, like fires or cutblocks, at much larger scales both faster and at a lower cost than airborne remote sensing. This linework, while being lower precision than LiDAR, actually ended up being very useful for a number of forestry and landscape vegetation analysis, including land cover classification, vegetation boundary delineation, and large-scale forest stand mapping and analysis.
Having GIS-ready polygons that are suitable for object-based or feature level, rather than pixel-level, analysis and attribution coming right out of the Land Lines process is the very first step we now use for forest inventory and vegetation mapping projects as a broad-level look at the land cover and vegetation boundaries. We use the Land Lines app on UP42 ourselves because it's got a very handy user interface that gives you access to the right sentinel-2 imagery for your project site and makes it super easy to get an end-to-end solution almost out-of-the-box.
What are the benefits users could see from using this block?
The first major benefit is the ability to vectorize imagery to outline land cover features of interest, turning heavy imagery into lightweight linework. The Land Lines app does a really good job of creating polygons that follow water bodies, including rivers, and delineating other major land cover types. When creating maps and models, it's useful to simplify the information from raster to vector to abstract features of interest from a bunch of pixels.
The Land Lines app can analyze hundreds or even thousands of square kilometers in just a few minutes to create these land surface feature polygons at multiple scales for different use cases.
The second key benefit is access to an image segmentation algorithm tailored to free satellite imagery for landscape-level feature analysis. We are a company that builds geospatial solutions using open-source frameworks, and we looked high and low for something off the shelf that does this kind of analysis; and there are not any affordable solutions out there for ready-to-go image segmentation that lets you do segmentation of natural features like this.
The key benefit to doing a classification analysis on image objects rather than individual pixels is you improve the accuracy and relevance of the results in natural settings like forests and other vegetation management use cases. These Land Line polygons lend themselves to many different object-based image analysis workflows for satellite image machine learning applications in natural landscapes, which is not something we have seen made available anywhere else.
Tesera has brought together a really impressive team. Could you speak a little more about that? How do their backgrounds allow Tesera to develop such unique solutions?
Our multidisciplinary team reflects a unique combination of skill sets from forestry, geomatics, remote sensing, software engineering, data analytics, machine learning, systems integration, Amazon Web Services cloud, and web development. Tesera's multidisciplinary team enables out-of-box thinking and creative solutions to embrace a philosophy of adopt, adapt and develop in that order of priority. We strive to act as OneTeam that believes, thinks, behaves, learns, works as one great integrated team.
Our OneTeam members strive to:
- Embrace cultural diversity (e.g., diversity is a strength)
- Achieve seamless integration (e.g., communications, standards, handoffs)
- Build capacity (e.g., across people, expertise, tools, training)
- Scale for success (e.g., efficient, effective, evolving frameworks, procedures, and processes)
- Pursue greatness (e.g., be awesome, always iteratively better)
- Take care and pride in our work in a manner that helps the team and our clients achieve success (e.g., focusing on they do best and stepping up to help other team members when needed)
- Display the "we" attitude (e.g., consideration for others on the team)
Sustainability is a topic that is of utmost importance to both Tesera and the world! Could you speak more about that and where you see the vision and mission of Tesera going in the coming months and years?
Tesera's High Resolution Inventory Solution is currently focused on sustainable forest management and the forest sector; it can also be applied to support many other secondary markets. The comprehensive and reliable data produced by HRIS provides the fundamental data that these markets require to function efficiently across the globe - with the objective of achieving better and more sustainable outcomes.
HRIS offers significant environmental advantages in terms of forest landscape conservation. Our full suite of landscape spatial data attributes makes it easier for decision-makers to plan and manage activities to achieve desired objectives, whether related to economic, ecological, cultural, and environmental values or markets.
Tesera's HRIS Vision: Sustainable forest management is realized through the use of HRIS, the world's most comprehensive, consistent, and reliable forest inventory solution.
Tesera's HRIS Mission: Transform global forest inventories with the world's best forest inventory solution - unlocking better returns on investment and enabling the sustainable management of forestry assets.
How is Tesera and, in a broader context, the geospatial community able to move the needle in solving these complex challenges?
So how do we move the needle? Because HRIS is significantly better than conventional forest inventories in every way. HRIS provides a 25 to 50 times return on a client's investment (ROI). It enables more comprehensive, consistent, and reliable updates to forest inventories. HRIS will provide forest managers the means to achieve the sustainable management of the world's forests.
Solving these challenges requires global effort from mission-driven companies and NGOs. How do you balance the social mission of Tesera with its commercial success?
Our team is excited by the opportunity to conserve forests through sustainable management. A scientific, data-driven, and integrated forest and landscape inventory approach is key to conserving biodiversity and the planet's well-being. HRIS also enables companies, governments, NGOs, investors, and financial institutions to gain improved information about ecosystem assets, risks, and opportunities and to finance more sustainable forestry and natural resource projects.
For many in the geospatial community, solving these challenges is what drives their passion for geospatial. What brings you the greatest satisfaction in your work?
Despite the critical role that our global forests play in mitigating climate and environmental impacts, conventional forest inventory technologies are unable to reliably identify…
"What types of trees are in the forest and where are they located, how fast are they growing, and what is the state of conservation values across the forest landscape?"
By focusing on the inherent weaknesses causing poor forest inventories (and addressing the industry's pain), we have laid the groundwork to transform global forest inventories and unlock higher sustainability, productivity and return on investment across the forestry supply chain.
What a fantastic mission and approach to solving such a complex and vital challenge. Thank you very much for joining us, Bruce!
You can find more information about Tesera's Land Lines processing solution and even try it yourself on the UP42 marketplace.
