The Climate Change Dashboard

Long story short: jump to the selected dashboard right away:

Climate Change

Climate change has become one of the most prominent topics in the media and in our everyday lives. Most of us have already noticed that winters tend to bring less snow, while summers are getting increasingly hotter. Extreme weather events, once considered rare, are now happening more often and in many different parts of the world. These changes are no longer abstract predictions; they are visible and affecting people everywhere. Many countries consider the consequences seriously.

Extreme Indicators

There are many indicators that help scientists understand climate change and its impacts. Some are straightforward, such as the average annual temperature. Others are more complex and require additional calculations. In our work, we have compiled 180 such indicators. The list and formulas for the core 26 can be found here, the dashboard contain short descriptions for other. A portion of them can only be calculated when specific measurements are available, for example, the dew point. However, roughly half of the indicators can be derived from basic daily observations like minimum, maximum, and average temperature, along with daily precipitation.

Some of these indicators can be new for you, but they have a sense for specialists. For example, one of the key indicators for construction engineering is the count of zero temperature crossing. This freeze - thaw cycle literally destroys basements, highway coat and railways. Another example is Vegetation Period and the associated metrics - important indicator for the agriculture industry.

We are pleased to share the results of these calculations through our interactive dashboards.

Sources of data

At the moment, we have uploaded two demo databases.

The first contains indicators calculated from 3-hour interval measurements collected at 517 Russian meteorological stations between 1961 and 2024. This dataset is extensive and includes a wide range of climate indicators. The values are aggregated at several levels: annually, monthly, by warm and cold seasons, and by the traditional four seasons.

By the way, there are more than 84.000 official meteo-stations in the World with long-term collections of observations.

The second database focuses only on temperature- and precipitation-based indicators. These are modelled data for the Netherlands covering the period from 2000 to 2070. The data were interpolated onto a uniform grid using outputs from 19 (out of 26 available) CMIP6 climate models. There are many scientific approaches for simulating and projecting future climate conditions, and modern climate models can estimate daily values all the way to 31 December 2100. This dataset represents an ensemble climate projection, meaning that results from all participating models are combined, and the median value is calculated for each grid cell. The years 2000-2015 form the historical period, which allows us to compare observed climate conditions with model predictions. So far, only yearly aggregations are uploaded to the database.

Dashboard

To start, select two indicators in the top-left corner of the dashboard and click Load. These two indicators will then be shown together on the chart on the right. If you select the same indicator in both fields, you can compare it across different periods or different locations. The map in the bottom-right corner always displays the first (red) indicator.

Once the data are loaded, you can choose the node or time-aggregation level for comparing the two indicators in the bar chart. The chart is scaled automatically to make the values easier to compare. A simple linear trend line across the full time period helps show the general direction of change.

The map visualizes the levels of the first indicator across the country using contour-style shading. By default, it shows the average over all years, but you can filter the map to display only selected years.

Please remember, modelled data are computed in the nodes of the regular grid. Some of these nodes got city names for convenience, these are not exact location.

Missing values

There are two main reasons why some values may be missing in the indicators.

1. Missing data in the original measurements
Weather station records are not always continuous: some days may be missing, and occasionally stations stop working for longer periods. When this happens, we sometimes have to exclude entire periods from certain calculations, which is why yearly and monthly results may not always match perfectly.

In future-climate model data, some grid points may also fall over the sea, which means no land-based value can be produced. In model-based data, a few models may produce missing values for certain locations, but the median shown in the dashboard is still based on the remaining available models. This may occasionally lead to unexpected spatial fluctuations, but it reflects the best information available for that grid point.

2. Missing results after the indicators are calculated
Some indicators simply cannot be computed in certain situations. For example, the “average temperature on cold days” cannot be calculated if a particular year has no days below 0°C. While it looks like missed year at the barchart for meteo-observations, in the situation of ensemble estimation, some models can be excluded, but we still observe some temperature computed based on other models.

Please, do not use these charts as a source of exact scientific information. Notes above, perhaps, gave you some flavour of the complexity of the result interpretation. There are thousands of really scientific papers available, please refer to them, or to the University scientists.

Links