The present map of groundwater temperature distribution is based on measurements from more than 109 temperature measurement points in the State of Berlin carried out in 2015. It is to be noted that for measurement depths below 60 m below the terrain surface the number of available measurement points decreases considerably. In some parts of the city, there are relatively large areas in which no temperature measurements can be obtained from greater depths.
In areas where no current measurements from 2015 are available, temperature measurements from "older measurements" were taken into account where applicable. Also, in areas for which no temperature measurements are available at greater depths, any available near-surface temperature profiles were extrapolated to greater depths, and these extrapolated values were included in the analysis.
Since February 2008, in addition to single measurements, temperature measurements have been carried out at regular intervals at groundwater measurement points especially established for that purpose. They consist of a full pipe with no filter, so that they can be filled with water almost to ground level, which permits monitoring of the annual temperature progression even in subsoil not saturated with groundwater.
As a rule, the measurement of the temperature profiles of the groundwater level up to the measurement depth is carried out with a measurement point interval of 1 m.
The measurement error which arises from calibration and measurement inaccuracy of the device, as well as the measurement inaccuracy of the object of measurement (e.g., due to heat convection at the measurement points) is estimated to be at least ±0.3°C.
The existing temperature measurement points, particularly those which measure the deeper subsoil at depths greater than 50 m, are not equally distributed across the territory of the state of Berlin. There are frequently large gaps, with horizontal intervals between measurement points of more than 3000 m.
Local climatic conditions at the surface can greatly influence the temperature distribution in the near-surface area, and increasingly, too, in the deeper subsoil. To some extent, it is apparent that the local representation of temperature measurement values at certain temperature measurement points can be very different, which has to be taken into account in data evaluation.
The representative range of a temperature measurement point varies according to the site environment. In an urban situation such as that in the state of Berlin, with its varying distribution of buildings and its alternation between uncovered green surface areas and highly impervious urban areas, the representative range is more strongly limited than it is in rural areas with less variable surface conditions which strongly affect the local climate and, as a result, the temperature conditions in the near-surface subsoil.
It is evident that temperature measurement values of a certain measurement point in an urban area often represent the conditions in the radius of less than 10 to 100 m around that measurement point.