A Advanced Usage

A.1 Calculating county fractions from raw growth profile data

In the chapter on Growth profiles, things ended with a look at “raw” county-specific growth profile data for BY2011 category #761 “Sanitary Sewers”. And, we could see that the raw data looked a lot like population counts.

This means that we can double-check the BY2011 county fractions, stored in file t1325, against fractions that we compute right here using the raw growth profile data.

Sanitary Sewers example

Let’s try this with our “Sanitary Sewers” example (which was linked to growth profile #657).

#
# Double-check the crosswalk. Category #761 "Sanitary Sewers" should be
# associated with growth profile #657 "Household Population".
# 
BY(2011) %>%
  DB_growth_profile_crosswalk() %>%
  filter_categories(
    "Sanitary Sewers" = 761) %>%
  ensure(
    .$gpf_id == 657)

cat_id	cat_h0	gpf_id	backcast	forecast	category
761	Area Source	657	TRUE	TRUE	Sanitary Sewers

#
# Compute county fractions from the raw growth profile data.
#
# - Extract just the CY2011 values.
# - Create a new column `cnty_frac`.
#
BY2011_raw_growth_profile_data %>%
  filter(
    gpf_id == 657) %>%
  filter_years(
    CY(2011)) %>%
  mutate(
    cnty_frac = gf_qty / total(gf_qty))

gpf_from	gpf_id	year	gf_qty	cnty_abbr	gpf_name	gpf_staff	gpf_date	cnty_frac
t0336	657	CY2011	1535560	ALA	Household Population	Minh H Nguyen	2010-02-25	0.2168986
t0336	657	CY2011	1086820	CC	Household Population	Minh H Nguyen	2010-02-25	0.1535139
t0336	657	CY2011	245700	MAR	Household Population	Minh H Nguyen	2010-02-25	0.0347053
t0336	657	CY2011	134080	NAP	Household Population	Minh H Nguyen	2010-02-25	0.0189389
t0336	657	CY2011	794560	SF	Household Population	Minh H Nguyen	2010-02-25	0.1122320
t0336	657	CY2011	728160	SM	Household Population	Minh H Nguyen	2010-02-25	0.1028530
t0336	657	CY2011	1815700	SNC	Household Population	Minh H Nguyen	2010-02-25	0.2564686
t0336	657	CY2011	309080	SOL	Household Population	Minh H Nguyen	2010-02-25	0.0436577
t0336	657	CY2011	429960	SON	Household Population	Minh H Nguyen	2010-02-25	0.0607321

In the new cnty_frac column, we can see percentages of 21.7%, 15.4%, 3.5%, 1.9%, 11.2%, 10.3%, 25.6%, 4.4%, and 6.1%, respectively.

County fractions pulled from DataBank

Now, let’s compare those to what’s stored in DataBank as “published” (file t1325, accessed via DB_county_fractions()).

#
# Pull the "published" county fractions for BY2011 (stored in `t1325`).
#
BY(2011) %>%
  DB_county_fractions(
    verbose = TRUE) %>%
  filter(
    cat_id == 761)

year	cat_id	cnty_abbr	cnty_frac
BY2011	761	ALA	0.2185224
BY2011	761	CC	0.1331767
BY2011	761	MAR	0.0401685
BY2011	761	NAP	0.0188647
BY2011	761	SF	0.1298252
BY2011	761	SM	0.1096952
BY2011	761	SNC	0.2560921
BY2011	761	SOL	0.0394379
BY2011	761	SON	0.0542175

We can see some mismatches, but there could be good reasons for them. This is just intended to show you how you can do the double-checking.

B Special Cases

B.1 Changes in forecasts for category #284

You might have noticed that the throughput forecast, for category #284, doesn’t follow the same trajectory as the NO_x forecast.

This is not because NO_x emission factors were projected to increase after CY2011. Here’s what happened:

The BY2011 criteria-pollutant calculations were completed.
The growth profile for this category, in DataBank, was updated.
The BY2011 GHG calculations were completed.

Here are some basic exercises (and charts) that may shed some light.

#
# This is the growth in NOx emissions from BY2011 category #284, via the
# published inventory (`BY2011_annual_emission_data`).
# 
BY2011_annual_emission_data %>%
  filter_categories(
    "#284 Water Heating" = 284) %>%
  filter_pollutants(
    "NOx") %>%
  chart_annual_growth(
    base_year = CY(2011),
    flag_years = CY(2020))

#
# This is the growth profile assigned to category #284 via file `t1325`. 
# It's not consistent with the published emission forecast (above).
# 
BY2011_growth_profile_data %>%
  filter_categories(
    "#284 Water Heating" = 284) %>%
  chart_annual_growth(
    base_year = CY(2011),
    flag_years = CY(2020))

#
# This is the growth in emissions, via the reproduced inventory. It's consistent
# with the assigned growth profile, but not with the published forecast.
# 
BY2011_area_source_projection_data %>%
  filter_categories(
    "#284 Water Heating" = 284) %>%
  filter_pollutants(
    "NOx") %>%
  chart_annual_growth(
    qty_var = "ems_qty",
    base_year = CY(2011),
    flag_years = CY(2020))

#
# This is the growth in throughputs, via the reproduced inventory. It's also
# consistent with the assigned growth profile, and with the reproduced
# emissions, but not with the published forecast.
# 
BY2011_area_source_throughput_data %>%
  filter_categories(
    "#284 Water Heating" = 284) %>%
  chart_annual_growth(
    qty_var = "tput_qty",  # changed from "ems_qty" to "tput_qty"
    base_year = CY(2011),
    flag_years = CY(2020))

If it still doesn’t make sense, please feel free to raise a question at one of our meetings!

C Technical Documentation

C.1 Control factors

Documentation for three key functions is printed out below.

DB_control_factors() is a legacy-format API for DataBank (t0326). In its output, Cd, Ed, and Ep are key. See help(t0326) as well.
annualize_DB_control_factors(), which itself depends on calculate_Fd_and_Fc(), implements the annualization algorithm mentioned in the chapter on Control factors. The References section of help(calculate_Fd_and_Fc) contains a reference to a PDF that describes the algorithm in several pages.

help(DB_control_factors)

DB_control_factors

R Documentation

DB_control_factors

Description

Retrieve control factors (keyed by category and date) for a given base year.

You have two choices:

DB_raw_control_factors(), for data organized in legacy format; or
DB_control_factors(), for a tidier version of the same information.

Usage of the latter is recommended. You will generally want to pipe its result through annualize_DB_control_factors() before (a) supplying it to BY_area_source_projections(), (b) charting it, or (c) combining it with other annual data.

See the Details and Examples, below, as well as the documentation for the functions above.

Usage

DB_control_factors(base_year, na.rm = TRUE, verbose = getOption("verbose"))

DB_raw_control_factors(base_year, verbose = getOption("verbose"))

Arguments

`base_year`	see `BY()`
`na.rm`	drop records where `Cd` is zero (“uncontrolled”)
`verbose`	display informative messages

Details

IMPORTANT: cf_qty is the uncontrolled fraction, not the controlled fraction. You should multiply things by cf_qty, not (1 - cf_qty).

In the data returned by DB_control_factors(), Cd is “the maximum percent emission reduction expected from all the changes in equipment, materials and operations due to the regulation.”

In the data returned by DB_raw_control_factors(), the values PM through CO correspond to the same Cd for each controlled pollutant, respectively. (Same data, but in “wide”, rather than “long”, format.) They are also returned as percentages (between 0 and 100), rather than fractions (between 0.0 and 1.0).

Value

Tabular data, with the following columns:

cat_id is the relevant (DataBank) category.
pol_id (and pol_abbr) is the relevant pollutant. There may be more than one, for a given regulation.
Ed is “Regulation Effectiveness. It represents the degree of compliance, in percent, for a given pollutant, or the degree to which the actual control approaches Cd.”
Ep is “Previous Regulation Effectiveness, in percent (equals O if emissions previously uncontrolled).”
cf_date is the effective date of the relevant regulation.
cf_desc is a description or note entered by staff.
reg_abbr is a more human-readable version of reg_id.
reg_name is the full name of the relevant regulation.

Functions

DB_control_factors: provides data organized in “legacy format” (slightly tidier than t0326, with Cd pivoted by pol_abbr).
DB_raw_control_factors: provides data in the same format as t0326.

References

See Tracking Emissions_Regulatory Control Levels_Effectiveness.pdf, located in the References/ subfolder of this package.

Examples

  DB_control_factor_data <- BY(2011) %>% DB_control_factors()
  DB_control_factor_data %>% calculate_Fd_and_Fc()
  DB_control_factor_data %>% annualize_DB_control_factors()

help(annualize_DB_control_factors)

annualize_DB_control_factors

R Documentation

Annualize control factors

Description

Convert “legacy style” control-factor data, such as that supplied by DB_control_factors(), into annualized uncontrolled fractions.

Usage

annualize_DB_control_factors(
  DB_control_factor_data,
  years = CY(1990:2050),
  na.rm = TRUE,
  verbose = getOption("verbose")
)

calculate_F_avg(Fc, F_wt)

calculate_F_wt(yday, n_days = 365)

Arguments

`DB_control_factor_data`	(tabular data) must have columns `Cd`, `Ed`, and `Ep` (see `DB_control_factors()`)
`years`	(CY) years for which output is desired; use `CY(…)`
`na.rm`	(logical) currently unused
`verbose`	(logical) display messages

Functions

calculate_F_avg: calculates F_avg as the mean of Fc, weighted by F_wt.
calculate_F_wt: calculates F_wt as the fraction of the calendar year up until yday.

calculate_Fd_and_Fc

Description

Fd is "the fraction of emissions remaining after the effect of a change in regulation control or its effectiveness.
Fc is the cumulative product of Fd.

Usage

calculate_Fd(Cd, Ed, Ep)

calculate_Fc(Fd)

calculate_Fd_and_Fc(legacy_data, verbose = getOption("verbose"))

Functions

calculate_Fd: calculates Fd from Cd, Ed, and Ep.
calculate_Fc: calculates Fc as the cumulative product of Fd.

Note

DB_control_factors() depends on this function.

References

Mangat T., Nguyen M., and Schultz S. (1999) Tracking Emission Levels Using Regulatory Controls and Effectiveness Factors. (See References/ in this package.)

Examples

legacy_data <- BY(2011) %>% DB_control_factors()
legacy_data %>% calculate_Fd_and_Fc()

Charting Annual Emissions

A Advanced Usage

A.1 Calculating county fractions from raw growth profile data

Sanitary Sewers example

County fractions pulled from DataBank

B Special Cases

B.1 Changes in forecasts for category #284

C Technical Documentation

C.1 Control factors

DB_control_factors

Description

Usage

Arguments

Details

Value

Functions

References

See Also

Examples

Annualize control factors

Description

Usage

Arguments

Functions

See Also

calculate_Fd_and_Fc

Description

Usage

Functions

Note

References

See Also

Examples