Morphing EPW weather variables — morphing

morphing_epw() takes an epw_cmip6_data object generated using extract_data() and calculates future core EPW weather variables using Morphing Method.

morphing_epw(
  data,
  years = NULL,
  labels = NULL,
  methods = NULL,
  warning = FALSE
)

Arguments

data: An epw_cmip6_dataobject generated using extract_data()
years: An integer vector indicating the target years to be considered. If NULL, all years in input data will be considered. Default: NULL.
labels: A character or factor vector used for grouping input years. Usually are the outputs of base::cut(). labels should have the same length as years. If given, climate data of years grouped by labels will be averaged. Default: NULL.
methods: A named character giving the methods of morphing procedures of each variables. Possible variable names are tdb, rh, p, hor_ir, glob_rad, wind. Possible values are: "stretch", "shift" and "combined". For example: c(tdb = "stretch", rh = "shift"). "combined" is only applicable to tdb. The default morphing method for each variable is listed in the Return section. If NULL, the default methods will be used. Default: NULL.
warning: If TRUE, warnings will be issued for cases with input data less than a decade (10 years). This is because using data that only covers a short period of time may not be able to capture the average of future climate. Default: FALSE.

Value

An epw_cmip6_morphed object, which is basically a list of 12 elements:

No.	Element	Type	Morphing Method	Description
1	`epw`	eplusr::Epw	N/A	The original EPW file used for morphing
2	`tdb`	`data.table::data.table()`	Stretch	Data of dry-bulb temperature after morphing
3	`tdew`	`data.table::data.table()`	Derived	Data of dew-point temperature after morphing
4	`rh`	`data.table::data.table()`	Stretch	Data of relative humidity after morphing
5	`p`	`data.table::data.table()`	Stretch	Data of atmospheric pressure after morphing
6	`hor_ir`	`data.table::data.table()`	Stretch	Data of horizontal infrared radiation from the sky after morphing
7	`glob_rad`	`data.table::data.table()`	Stretch	Data of global horizontal radiation after morphing
8	`norm_rad`	`data.table::data.table()`	Derived	Data of direct normal radiation after morphing
9	`diff_rad`	`data.table::data.table()`	Stretch	Data of diffuse horizontal radiation after morphing
10	`wind`	`data.table::data.table()`	Stretch	Data of wind speed after morphing
11	`total_cover`	`data.table::data.table()`	Derived	Data of total sky cover after morphing
12	`opaque_cover`	`data.table::data.table()`	Derived	Data of opaque sky cover after morphing

Each data.table::data.table() listed above contains 19 columns below or an empty data.table::data.table() if the corresponding variables cannot be found in the input epw_cmip6_data object.

No.	Column	Type	Description
1	`activity_drs`	Character	Activity DRS (Data Reference Syntax)
2	`institution_id`	Character	Institution identifier
3	`source_id`	Character	Model identifier
4	`experiment_id`	Character	Root experiment identifier
5	`member_id`	Character	A compound construction from `sub_experiment_id` and `variant_label`
6	`table_id`	Character	Table identifier
7	`lon`	Double	The averaged values of input longitude
8	`lat`	Double	The averaged values of input latitude
9	`dist`	Double	The averaged spherical distances in km between EPW location and grid coordinates
10	`interval`	Factor	The label value used to average raw input data
11	`datetime`	POSIXct	The datetime value with fake year generated by calling the `Epw$data()` method with the input EPW
12	`year`	Integer	The original year of the raw EPW data
13	`month`	Integer	The month value of the morphed data
14	`day`	Integer	The day of the morphed data
15	`hour`	Integer	The hour of the morphed data
16	`minute`	Integer	The minute of the morphed data
17	Variable Name	Double	The morphed data, where `Variable Name` is the corresponding EPW weather variable name
18	`delta`	Double	The shift factor. Will be `NA` for derived values
19	`alpha`	Double	The stretch factor. Will be `NA` for derived values

Details

The EPW weather variables that get morphed are listed in details.

The Morphing procedure

Here Morphing is an algorithm proposed by Belcher etc. 2005 used to morph the present-day observed weather files (here the EPWs) to produce future climate weather files. The EPW data is used as the 'baseline climate'.

The first step before morphing is to calculate the monthly means of climatological variables in the EPW file, denoted by $<x_0>_m$. The subscript '0' is to denote the present day weather record, and 'm' is to denote the month.

The morphing involves three generic operations, i.e. 1) a shift; 2) a linear stretch (scaling factor); and 3) a shift and a stretch:

$$ \textrm{Shift: } x = x_0 + \Delta x_m $$

$$ \textrm{Stretch: }x = \alpha_m x_m $$

$$ \textrm{Shift + Stretch: } x = x_0 + \Delta x_m + \alpha_m (x_0 - <x_0>_m) $$

Shift

If using a shift, for each month, a shift $\Delta x_m$ is applied to $x_0$. $\Delta x_m$ is the absolute change in the monthly mean value of the variable for the month $m$, i.e. $\Delta x_m = <x_0>_m - <x>_m$. Here the monthly variance of the variable is unchanged.

Stretch

If using a stretch, for each month, a stretch $\alpha _m$ is applied to $x_0$, where $\alpha _m$ is the fractional change in the monthly-mean value of a variable, i.e. $\alpha _m = <x>_m / <x_0>_m$. In this case, the variance will be multiplied by to $\alpha^2_m$

Combined Shift and Stretch

When using a combined shift and stretch factor, both the mean and the variance will be switched off altogether.

For more details about morphing, please see (Belcher etc. 2005)

References

Belcher, S., Hacker, J., Powell, D., 2005. Constructing design weather data for future climates. Building Services Engineering Research and Technology 26, 49–61. https://doi.org/10.1191/0143624405bt112oa