Climate Explorer User Guide

Climate is the long-term behavior of the atmosphere generally presented as averages. i.e. average monthly rainfall, average annual temperature.

Weather is what we observe over short periods such as days or weeks.

Climate projections cannot tell us what will occur in any specific day, month, or year in the future.

Climate scientists project future climate using global climate models (GCM). These are complex models run on computers that project future climate based on state of the art science.

Many groups of researchers have developed their own climate models meaning there are multiple projections for the future climate.

There are many sources of uncertainty in climate projections. These include uncertainty in greenhouse gas emission projections.

Different climate models simulate the impacts using different methods leading to diverse projections. There also remain areas where climate scientists continue to improve their understanding of the climate and its responses.

"Uncertainty" is used in the scientific sense of the term. Scientists acknowledge that there is a range of possible outcomes from climate change.

An overwhelming majority of scientists agree that climate change is occurring and is caused by human activity.

Most climate models project climate over fairly large areas, using a grid with each "cell" in it being 100-600 kilometers along each side in the real world.

To make these outputs more useful for understanding local effects, these data are downscaled. The data in Cal-Adapt and used here have been downscaled to a grid of about 7km covering California.

Climate projections tell us how conditions are likely to change on average, not what the weather will be on a specific date.

The averaged values across different model projections are considered more likely than any individual model value. But if you are trying to plan for the future, it is also important to look at individual model values and consider the full range of model outcomes.

Think of each greenhouse gas emissions scenario as a separate possible future instead of averaging them.

Two emissions scenarios were used to define the level of greenhouse gasses in the climate projections.

RCP 8.5, a high-emissions scenario, in which greenhouse gases continue to rise over the 21st century.

RCP 4.5, a low-emissions scenario, with emissions leveling off around the middle of the century and tapering to below 1990 levels by the end of the century.

Ensembles: Many of the data presented are averages of a group or ensemble of climate models. The resultant averages better represent the likely future climate than any individual model.

Individual Models: As the product of a single modeling effort or team, individual models represent one possible set of outcomes for a future climate.

 Cal-Adapt  and the  4th Climate Change Assessment  reference ensembles of 4 and 10 models from a set of 32 Global Climate Models (GCMs).

Selection of the models is described in detail by  Pierce et al (2018) .

The 10 and 4 model ensembles are described in the next slides.

The ensemble includes downscaled projections from 10 GCMs that closely simulate California's climate. Table 1 in  Pierce et al. (2018)  identifies these 10 models.

The models in the 4 model ensemble are selected from within the 10 model ensemble which produces a warm/dry (HadGEM2-ES), average (CanESM2), cooler/wetter (CNRM-CM5), and a model (MIROC5) least like the others to improve coverage of the range of outcomes.  (Pierce et al, 2018) 

The  Localized Constructed Analogs  (LOCA) statistical method was used to downscale all 32 global climate models (GCMs) from 100km resolution to ~7km.

Each of the GCMs represents climate variability. Model outputs from a single year will likely be either higher or lower than the long-term trend. Using results from a single year's model outputs may lead to incorrect and invalid conclusions.

Climate scientists for Cal-Adapt recommended using a 30-year average to better understand the underlying trends.

There are two types of 30-year averages used in the Climate Explorer App.

Named climate periods of modeled outputs for the historic baseline period (1961-1990), mid-century (2035-2064), and end of century (2070-2099)

Rolling averages which provide an average of the 15 years before and after a date. i.e., 2050 is the average for 2036-2065.

When an individual model is selected and a location selected on the map an additional dotted line will be visible.

These include extreme heat days and extreme precipitation.

Extreme heat days are the projected number of days in a year where the daily maximum temperature is greater than the 98th percentile of the April-October maximum temperatures from 1961-1990 for that location.  More Information  (Cal-Adapt.org)

An "estimated return level" is an estimate of the maximum precipitation projected to occur once in a century. These high and low confidence levels are presented for the historic baseline (1961-1990), mid-century (2035-2064), and end of century time (2070-2099) periods.  More Information  (Cal-Adapt.org)