CAnVAS, or the Climate Analysis & Visualization for the Assessment of Species Status, is a web-based tool designed to display future climate projections for a number of weather parameters over the southeastern United States.

This tool was the end result of a project led by the State Climate Office of North Carolina and supported by the Southeast Climate Adaptation Science Center, which ran from 2018 through 2023. Over the course of this project, the CAnVAS interface was evaluated and refined through usability testing including eye tracking, along with one-on-one engagement with project stakeholders and end users.

The CAnVAS tool was primarily developed to meet the needs of scientists with the US Fish & Wildlife Service who are responsible for writing Species Status Assessments. These detailed reports cover present and future threats to endangered and threatened species, including climate-based risk factors that may impact habitats, food sources, breeding activity, and other parts of a species' life cycle.

However, the data in CAnVAS may also be useful for other users, including non-scientists, since its visualizations have been designed to be easy to view and understand, and the parameters included may be of broad interest – for instance, the expected number of days per year with extreme hot or cold temperatures.

CAnVAS uses a dataset called MACA, or Multivariate Adaptive Constructed Analogs, which was developed by John Abatzaglou at UC Merced. MACA began with a set of 20 global climate models, which were then statistically downscaled over the continental United States and bias corrected in order to best match historical climate conditions. The result is a set of high-resolution (4-kilometer grid size) historical baseline data and future climate projections through the end of the 21st century.

MACA is unique in the number of climate variables it offers. Its output includes daily maximum and minimum temperature, maximum and minimum relative humidity, precipitation accumulation, downward surface shortwave radiation, wind velocity, and specific humidity. A number of these parameters, and calculations based on them, are accessible in CAnVAS.

Because we don't know exactly how human activities related to greenhouse gas emissions will change in the future, climate models often use multiple emissions scenarios in order to simulate a few of the more likely pathways that our policies and emissions might take.

CAnVAS includes two such scenarios, labeled as RCP8.5 and RCP4.5. "RCP" stands for Representative Concentration Pathway, while the numbers represent the maximum increase in radiative forcing – in Watts per square meter – compared to historical levels. Either scenario can be toggled on or off using the blue toolbar near to the top of the interface.

In recent years, the radiative forcing due to the global emission of fossil fuels has fallen somewhere in between the pathways shown by RCP4.5 and RCP8.5. This makes these two scenarios good upper and lower bounds for short-term projections.

From mid-century and beyond, RCP4.5 assumes a decline and eventual stabilization of radiative forcing to near historical levels, while RCP8.5 assumes a continued increase through 2100. Some scientists treat RCP8.5 as a worst case scenario, while achieving the forcing depicted by RCP4.5 will require an eventual curbing of emissions, which has not yet occurred.

While all species can be affected by the whims of our day-to-day or year-to-year weather conditions, scientists writing Species Status Assessments are more interested in knowing how the overall climate may change, since this is more likely to affect species habitats and life cycles.

In order to represent what the future climate may look like, and how it may compare with what has happened historically, CAnVAS breaks the historical and future time periods into 30-year time slices. This effectively averages out the year-to-year variability and keeps only the long-term climate signal.

The MACA dataset used in CAnVAS is based on 20 downscaled global climate models, and each model may show a slightly different outcome based on their own unique physics and formulations. The visualizations in CAnVAS are therefore designed to show the spread of these 20 model outcomes, which is calculated as 2 standard deviations above and below the multi-model mean.

Note: this does not represent the variability you might expect in any given year, but rather, the spread among model projections for the average annual conditions over each 30-year time slice.

Having this full spread, instead of only the multi-model mean value, can be useful for assessing future changes. For instance, if the entire spread of outcomes for a time slice is greater than or less than the historical average, that should give higher confidence that conditions during that time period will be different than what was observed historically.

As a quick way to see how the spread compares with the historical average, click the “View summary table” button beneath the plot for a given parameter. Any time slices for which the entire spread of outcomes is greater than or less than the historical average are highlighted in bright pink or orange in this table.

Funding for the CAnVAS project expired in 2023, so there is not currently support to update the interface or calculate new parameters. While the MACA dataset used in CAnVAS is not as current as other model projections such as CMIP6, the goal of this project was to develop and evaluate a working interface that could be used to inform Species Status Assessments based on a set of well-documented climate projections with which we had experience using and visualizing. Thus, barring future funding and support for further development, we do not plan to update the data in CAnVAS.

An example citation in APA format is as follows:
CAnVAS Tool. (2023, April). Retrieved December 19, 2024, from https://products.climate.ncsu.edu/canvas

If you have any questions or comments about the CAnVAS interface, contact the State Climate Office of North Carolina at climate-office@ncsu.edu.