ClimateVision - Documentation

Contents

What is ClimateVision?

Climate Models

Climate Variables

Some Interesting Insights

Data Sources and References

What is ClimateVision?

ClimateVision is a simplified regional climate data platform for viewing 20 different climate variables – including 6 basic variables and 14 derived variables – for any location within the contiguous US starting in the year 2020 and ending in 2095. This data platform is a free community service from Model Paths.

The inputs are simply a US address and one selected climate variable. The address can be a full street address, or simply a city, state, zip code, or any combination of those. The result is a chart displaying the selected variable at 5-year intervals through 2095 for the specific US location. The result includes data for two GHG emission pathways shown side-by-side. Users are allowed up to 10 requests per minute.

The underlying data comes from the NA-CORDEX dataset, which consists of regional climate model outputs for North America based on boundary conditions from the global climate model simulations in the CMIP5 archive.

Climate data platforms like this are essential for developing climate adaptations and designing resilient systems in many different economic sectors including energy production, buildings, infrastructure, agriculture, transportation, insurance, etc. ClimateVision provides a simplified view of the climate data that would be needed in these sectors, intended to show what is possible with the detailed climate models that already exist today.

Contact us for data specific to your industry or geography.

Climate Models

The starting point for the data comes from the global climate model (GCM) simulations. All of the GCM-driven simulations are full transient runs from 1950 to 2100. The historical period spans 1950–2005 and the future period 2006–2100. GCMs provide boundary conditions on an hourly basis for the North American regional climate models (RCMs) to produce dynamically downscaled regional results that have higher spatial resolution and better comprehension of the land surface detail. RCM outputs are then bias corrected against the gridMET gridded observational datasets. For more background on GCMs, RCMs and bias correction, see this paper.

The GCMs and RCMs used to generate the NA-CORDEX dataset are listed in the simulation matrix here. This data platform uses up to nine GCMs and seven RCMs, along with two RCPs. The spatial resolution used in this platform is 50 km (~0.5°), so the simulation grid cells are 50kmx50km.

Two emission (representative concentration) pathways are included: RCP 4.5 and RCP 8.5. RCP 4.5 is IPCC’s moderate scenario in which emissions peak around 2040 and then decline. RCP 8.5 is IPCC’s highest baseline emissions scenario in which emissions continue to rise throughout this century. Note that the GCM/RCM combinations for the two RCPs are not identical – simulation output for each RCP is based on its own set of GCMs and RCMs, so data for the two RCPs will not align in the starting year (2020) used in the charts.

It is important to interpret correctly the simulation results from climate models and the data presented here based on those simulation runs. Climate should be understood as the statistical summary of the weather events occurring in a given season , and long-term climate is a long-term statistical summary of likely weather events. In addition, these simulations are based on possible emission scenarios such as RCP 4.5 and RCP 8.5. The simulation results (including the derived variables) we present here are not forecasts, but projections based on a set of underlying assumptions and scenarios. These projections are actionable -- for example, to adapt to possible future conditions, or to even prevent those conditions from occurring by limiting emissions -- but they are not necessarily predictive of what may actually happen in the future.

Climate Variables

The following basic climate variables are included in this data platform:

  • Near-Surface Air Temperature, °C
  • Precipitation, mm/day
  • Near-Surface Relative Humidity, %
  • Near-Surface Specific Humidity, kg kg-1
  • Surface Downwelling Shortwave Radiation, W m-2
  • Near-Surface Wind Speed, m s-1

Each variable is first extracted on an annual frequency from the NA-CORDEX data archive, then averaged over 5-year intervals (for example, the result for year 2030 is an average of the data from 2028 through 2032), and finally averaged over multiple GCM/RCM combinations for each RCP. The time-averaging smooths out some of the year-to-year variations. Averaging over an ensemble of multi-model simulations provides a more robust picture of future conditions than any one model by smoothing out some of the model uncertainties. Some extreme longitudes (especially on the US East Coast) do not have data in the NA-CORDEX archive – these are filled in when possible using valid data from the nearest longitudes within one degree in either direction.

Note that the near-surface air temperature values are actual temperatures in degrees Celsius and not just the temperature anomalies. Additionally, the results include data for both a moderate (RCP 4.5) emission pathway and a worst-case (RCP 8.5) pathway. It is likely that the final trajectory of the climate variables will be bounded by these pathways through the end of this century, with the spread between the two scenarios becoming significant after mid-century. Trend lines superimposed on the chart in lighter color help compare the climate response to the RCPs.

The platform also includes the following derived climate variables:

  • Dry Spells (longest annual stretch of consecutive days with daily total precipitation below 1 mm), days/year
  • Wet Spells (longest annual stretch of consecutive days with daily total precipitation above 2.5 mm), days/year
  • Dry Days (annual count of days with daily total precipitation below 1 mm), days/year
  • Wet Days (annual count of days with daily total precipitation above 2.5 mm), days/year
  • Heat Waves (annual count of 6 or more consecutive days with max temperature above 100 °F / 37.78 °C), days/year
  • Cold Spells (annual count of 6 or more consecutive days with min temperature below 32 °F / 0 °C), days/year
  • Hot Days (annual count of days with max temperature above 100 °F / 37.78 °C), days/year
  • Cold Days (annual count of days with min temperature below 32 °F / 0 °C), days/year
  • Heating Degree Days (annual count of degree-days with daily average temperature below 65 °F / 18.33 °C), °F-days/year
  • Cooling Degree Days (annual count of degree-days with daily average temperature above 65 °F / 18.33 °C), °F-days/year
  • Adequate Solar Intensity Days (annual count of days with average shortwave radiation above 167 W/m2), days/year
  • Adequate Wind Intensity Days (annual count of days with average wind speed above 5.8 m/s), days/year
  • Solar Energy Potential (total annual solar energy per unit land area), kWh/m2/year
  • Wind Energy Potential (total annual wind energy per unit swept area), kWh/m2/year
  • Fire Weather Index (FWI), upper and lower bounds

The derived variables are extracted on a daily frequency from the NA-CORDEX data archive, and then annual values are calculated at 5-year intervals using the daily values.

Note on FWI: This fire danger rating system requires temperature, relative humidity and wind speed measured at 12 noon local time, and precipitation totaled over the previous 24 hours. Since the necessary climate simulation outputs were only available on a daily frequency, we generated an upper bound for FWI using the maximum daily temperature in conjunction with average daily values for relative humidity and wind speed. For precipitation, we used the daily total for the current day instead of for the last 24 hours. For the FWI lower bound, we used the average daily temperature in place of the maximum. We then validated these bounds using historical FWI data.

Some Interesting Insights

This simplified data platform is built on a robust technical foundation and can produce many useful insights about future climate this century within the contiguous US. Here are a few examples:

  • There is not a significant divergence between the two RCPs before 2050 for any of the variables. This emphasizes the frequently stated point that climate change is baked in for the next two decades or so, and the decisions and changes we make today will really have an impact in the second half of the century. This means that serious adaptation is needed while mitigation efforts continue over the next few decades.
  • Specific humidity increases over time in a pattern that looks very similar to the temperature rise. As temperature increases, the atmosphere can hold more moisture – and the climate simulations are simply illustrating this.
  •  Relative humidity has a completely different pattern because it measures humidity relative to the maximum possible at any given temperature. In New York city, relative humidity increases over time under RCP 4.5 but decreases under RCP 8.5, reflecting both the moisture in the atmosphere and the temperature in future decades.
  • Total direct and diffuse solar radiation hitting the surface (“Surface Downwelling Shortwave Radiation”) generally decreases in many locations as specific humidity increases over time, illustrating an inverse relationship between the two variables. This may have implications for the performance of solar panels among other things.
  • Precipitation generally increases over time (consistent with other indicators like temperature and humidity), but this pattern is not as clearcut as temperature and specific humidity.
  • Cold spells (all derived variables defined above) generally decrease at most locations through the end of this century.
  • The number of cold days generally decreases at all locations.
  • Heat waves and the number of hot days generally increase at most locations. Northern locations begin to see heat waves multiple times a year in the second half of the century under the RCP 8.5 emissions trajectory.
  • The number of wet days per year increases and number of dry days decreases at many northern locations.
  • Heating degree days decrease at all locations while cooling degree days increase, indicating significantly increased energy use in summers for cooling while winter energy use for heating declines.
  • Solar energy potential at most locations decreases by up to 5% between now and the end of the century.
  • Wind energy potential generally declines slightly under RCP 4.5 and increases slightly under RCP 8.5 for some of the existing large wind farm locations in the US.
  • The fire weather index (FWI) -- the most widely used fire danger rating system in the world -- generally increases in the range of 5-20% between now and end of the century under RCP 8.5, suggesting significantly increased fire risk in a high-emissions scenario for many parts of the US. FWI is flat or increases/decreases slightly under RCP 4.5, showing that any increased fire risk in the future (especially in the second half of this century) is still entirely in our control and is a function of the future emissions trajectory.

Data Sources and References