Climate regulations and investor demands, rather than voluntary pledges, are now driving the decarbonization of existing buildings.

If you’re like most building portfolio owners, you likely only have access to limited information about your buildings and translating that information into an actionable decarbonization plan remains expensive and challenging. Carbon Signal offers the opportunity to rapidly develop a building-by-building existing conditions characterization and sub-metering level energy use disaggregation across your real estate portfolio. It is powered by a novel combination of two techniques - the proven fidelity of physics-based building energy models, and the speed of machine learning models – which reduces the time and cost associated with building energy audits by two orders of magnitude.

Using only the building size, location, and monthly utility data, Carbon Signal automatically determines the primary energy drivers in each building, develops auto-calibrated building energy models, and utilizes these models to simulate future energy and carbon scenarios - from building envelope and lighting upgrades to operational controls and HVAC system electrification- to estimate the impact these measures in every building across your portfolio.

With both machine learning and physics-based models, Carbon Signal uses an ensemble modeling approach that comprehensively searches and combines all high-fit solutions into a probabilistic model. This generates a distribution of likely conditions that can explain the current building energy use and help forecast future energy scenarios with quantified uncertainties that deliver a clear picture of possible real-world outcomes.

Finally, Carbon Signal allows users to define their own relevant criteria such as project first costs, embodied carbon impact, and equipment replacement timelines to be compared against operational carbon reduction potential. Users can select combinations of building-specific projects grouped by multi-objective optimization recommendations to organize portfolio-wide decarbonization pathways, implementation plans, and portfolio and asset management strategies.

The Carbon Signal engine has gone through rigorous technical reviews. The methodology behind the analysis has been published in some of the most reputed peer-reviewed scientific journals, the results repeatedly validated against detailed building energy audits, and the framework successfully applied toward decarbonization pathways for global portfolios of some of the most demanding real estate owners over the last 10 years.

How does carbon signal help with decarbonization planning?

Carbon Signal helps develop actionable, viable, and technically grounded portfolio decarbonization plans: As building owners commit to ambitious carbon reduction targets and timelines, Carbon Signal identifies the right strategies to achieve these targets and helps measure progress along the way.

Carbon Signal helps strengthen investor confidence as the world transitions to a clean energy economy: As investors realize that arbitrary decarbonization targets over arbitrary timelines are no longer an option, Carbon Signal validates actionable strategies and informs ongoing investment plans.

Carbon Signal helps real estate portfolio managers gain insights from already-monitored asset level data: As energy aggregation and reporting platforms become common, Carbon Signal provides a layer of sophisticated predictive analysis to inform and track portfolio and asset management plans.

Our approach

It takes a fraction of the time to generate solutions compared to traditional methods: Engineering assessments rely on expensive, time-intensive, impossible-to-scale in-person audits. Carbon Signal offers rapid, automated, high-fidelity, scalable building and stock level analysis.

Our approach facilitates auto-calibration against low-res data to produce probabilistic high-res forecasts: Energy models rely on smart-meter interval data to offer demand management recommendations. Carbon Signal forecasts probabilistic hourly load profiles based only on monthly energy use data.

Our approach uses physics-based simulations to evaluate specific strategies accounting for future changes: Most tech platforms rely on machine learning models trained on typology based historical data trends. Carbon Signal considers site specific variations in future climate, usage, and technology changes.

Applications

Carbon Signal has been applied in a variety of settings, from colleges and universities to corporate real estate portfolios, to entire cities.

One of our very first projects, at MIT, identified effi­ciency opportunities in several of their academic research labs, many of which were not designed or upgraded to meet present day energy performance standards. Utilizing only monthly utility data from each building to create auto-calibrated energy models, we were able to help prioritize investments in efficiency upgrades. We rolled this analysis into self-contained mini reports for each building that we shared with building operators who in turn used them to solicit high-quality bids for future audit and performance contracting work.

We’ve also used our approach at a global scale for corporate real estate portfolios, including Amazon’s Global Real Estate Facilities portfolio of office buildings. With nearly 500 buildings spread across the world, Carbon Signal was ideally suited to fast-track the audit process and help Amazon quickly identify the most significant opportunities for decarbonization. Amazon was also able to leverage Carbon Signal to identify gaps in their utility data, thereby creating a more complete accounting of their energy use and carbon emissions.

At the extreme end of the spectrum in terms of scale, we’ve also used Carbon Signal for city-scale analysis, working with the New York State Energy Research and Development Agency (NYSERDA) to evaluate the impact of distributed energy resources in the Brooklyn neighborhood of Sunset Park, and with Urban Green to measure the impact of widespread building electrification in New York City. This analysis helps arm policymakers with the information they need to successfully make the transition to a clean-energy future.

Image © YeyenDesign

‍

‍