This file documents the process for creating hdd.json, which defines a
heat degree day value for a set of polygons.
1. Inputs
- file
nrg_chdd_afrom eurostat's HDD data - yearly HDD data for the member states of the EU. - small scale cultural geography map from naturalearthdata.
2. Method
Eurostat HDD data is defined as follows (source):
If Tₘ ≤ 15°C Then [HDD = ∑ᵢ(18°C - Tⁱₘ)] Else [HDD = 0] where Tⁱₘ is
the mean air temperature of day i.
As HDD values are trending upwards over time, despite Eurostat data going back to 1979 only the average of the last 5 years was used.
However, the THERMOS heat demand model was trained using HDD data for
Copenhagen where the base temperature was 17°C: HDD = ∑ᵢ(17°C - Tⁱₘ).
To remove this discrepancy, the Eurostat HDD values are multiplied by
HDD(Copenhagen, 17°C)/HDD(Copenhagen, Eurostat).
2.0.1. HDD(Copenhagen, 17°C):
2006 2990.5 2007 2819.8 2008 2901.1 2009 3134.4 2010 3642 2011 3007.6 2012 3248 2013 3190.9 2014 2745 2015 2885.8 2016 2910.7 2017 2986.8 avg 3038.55
For HDD(Copenhagen, 17°C), the average of all the years was used as the model was trained on all the years.
2.0.2. HDD(Copenhagen, Eurostat)
2943.88466666667
The average of the latest 5 years was used as in the other data from Eurostat.
2.0.3. Conclusion
All Eurostat HDD values were multiplied by 3038.55 / 2943.88 =
1.03215660396116, and added as properties to the geoJSON from
naturalearthdata.