All buildings in the Netherlands will have to become "Paris-proof" over the next few years. This is a huge responsibility for property owners such as institutional real estate investors, pension funds, and housing associations. But it's not the only one, because climate change is creating new risks. How vulnerable are buildings to heat stress, flooding, drought, subsidence, and flooding, for example? More and more property owners are trying to get to grips with the measures that are needed. This is where Sweco helps property owners and, to do so, they recently used Spotr, a tool that draws on Artificial Intelligence and Machine Learning to retrieve the necessary data. We sat down with Jeroen van Eekelen and Jelmer van de Ridder, both consultants at Sweco.
Jelmer and Jeroen have been working on climate change risks for years and provide services to organizations in both the public and private sectors. They help municipalities, cities and real estate owners to identify physical climate risks from which they prepare strategies and action plans to mitigate them, thus creating a safe and healthy living environment.
Need for insight
Jelmer and Jeroen have recently seen a growing need among property owners to understand the much needed climate change adaption.
Jeroen: "This is in part due to European legislation, specifically the Sustainable Finance Disclosure Regulation and the associated EU Taxonomy. If you're a real estate investor trying to demonstrate that you're a sustainable fund, you have to show that you're aware of climate risks and addressing them where needed. In other words, you have an information obligation. Real estate owners also see that it's smart to understand all this now because of the measures needed to become Paris-proof. The fact is that combining those measures with climate change efforts is the most efficient approach because it saves you from having to do the same thing twice."
Collecting data on climate risks
A Spotr presentation they attended gave them the idea to get in touch with the company.
Jelmer says: "Dirk Huibers, Spotr's CEO, gave a presentation at Sweco on how Spotr uses AI to visualize the characteristics and surroundings of buildings, and obtain input for making them more sustainable. That stuck with us. So, when a client contacted us to find out whether we could help collect data to screen his portfolio for climate risks, we thought: let's use Spotr to collect it. Our client liked the idea so we decided to set up a pilot program with the tool."
Vulnerability to heat stress and flooding
Jeroen explains: "The goal was to investigate the vulnerability of buildings to heat stress and flooding. To do this, we needed to know the relevant building characteristics. We could have retrieved that data manually, but it would have demanded a lot of time and manpower that our client didn't have readily available. Using Spotr, however, we managed to gather that building data far more quickly and automatically."
"We looked at just over 300 assets," Jelmer adds, "stores, offices and homes. The data we were looking for was, for example, information about the height of the threshold, the presence of a parking garage, basement or vents where water can flow in, and whether there were sunshades, shutters or AC's visible on the outside. We also retrieved information about the color of the wall and roof, the total area of glass and the presence of an overhang that keeps out sunlight."
Jeroen adds: "Based on characteristics like these, we can determine a building's vulnerability. We then work with risk scores and of course we have a method that enables us to give advice on what measures to take. For example: this building has a high risk of inside flooding, and these would be cost-effective measures to control that risk.”
Making data as smart as possible
"We had very smooth dealings with Spotr," Jelmer says. "We were under quite a lot of pressure, so we had to move quickly. We started working together quite quickly. Maybe a little too quick in hindsight, because we also retrieved some data that we weren't able to use immediately. But, despite all the pressure, we were also able to put a lot of that data to good use."
Jeroen chimes in: "That's a good lesson we learned from this pilot program: on the front end, be very precise about what your frameworks are and what Spotr needs to measure. For example, the command "measure the overhang" is too vague because of the uncertainty of what exactly the overhang is. Might it also be a gallery balcony, or a signboard over your windows? That was a learning experience; we should have been clearer about that. Right from the start you have to get together and clearly decide what to examine in a building and what is what. Ensuring that your data request on the front end is as smart as possible improves the quality of your end result."
Combining measures is easier
Jelmer: "There is still a fair amount of uncertainty in the data, because the algorithm still has to learn. But that's a matter of training the AI to produce better and better information. That takes time, but we certainly believe this is the future, particularly in view of the labor shortages in the technical job market and the number of surveyors who will be retiring in the coming decades."
"We also see a lot of value in using Spotr for monitoring purposes," Jeroen says. "One thing would be to have Spotr measure the data every year, thus making it clear to us how a building's vulnerability changes, for example, because of the presence of a roller shutter. That would give us an accurate idea of how the measures are progressing. With that, it actually pays off to invest time at the start of the process, because that would enable us to then keep improving the tool.
"Combining measures will also get easier because of the excellent insights we get from Spotr. So, we don't just look at the climate risks but we also examine the measures needed to become Paris proof and then do the things that have the most impact on both. We don't just look at individual buildings, we examine their surroundings too. How much green space is there in a neighborhood, how paved over is it, and what measures would be needed and are possible to bring about greening? Not just planting more trees and plants, but also looking at the presence of bats, swallows and nesting boxes. This prevents the need for an ecologist to do fieldwork."
Enormous potential in innovative technology
Jelmer: "Looking back on it now, I think carrying out this pilot program was exciting and a new challenge. But we also knew that Spotr had already made its mark with big housing associations and we wanted to give innovative technology a chance. It was very instructive to download certain buildings features based on artificial intelligence. The technology has huge potential and we will definitely start using Spotr more often in the future."
Want to know more?
Spotr uses AI to detect defects and measurements in large property portfolios, so our clients can find investment opportunities or identify risks. Our image sources are satellites, planes, cars and phones. Want to know more? Request a demo here.