This function creates a seasonally detrended flow data set for selected USGS gages. The created data set is used to support application of GAMs that include a hydrologic term as one of the independent variables. The output from this function should be stored as an .rda file for repeated use with baytrends.
detrended.flow(
usgsGageID,
siteName,
yearStart,
yearEnd,
dvAvgWinSel = c(1, 5, 10, 15, 20, 30, 40, 50, 60, 90, 120, 150, 180, 210),
dvAvgWgtSel = "uniform",
dvAvgSidesSel = 1,
lowess.f = 0.2,
span = 10,
max.fill = 10
)USGS GageIDs (e.g., "01491000")
USGS SiteName (only used for plots)
start year (recommended as at least one year before corresponding water quality data set)
end year
Averaging window (days) for smoothing the residuals of the seasonally adjusted daily flow values [default = c(1, 5, 10, 15, 20, 30, 40, 50, 60, 90, 120, 150, 180, 210)]
Averaging method ("uniform", "weighted", or "centered") for creating weights. If using "weighted" then use dvAvgSidesSel=1. If using "centered" then use dvAvgSidesSel=2. [default = "uniform"]
If dvAvgSidesSel=1 only past values are used, if dvAvgSidesSel=2 then values are centered around lag 0. [default = 1]
lowess smoother span applied to computed standard deviation (see Details). This gives the proportion of points which influence the smooth at each value. Larger values give more smoothness. [default = 0.2]
maximum number of observations on each side of range of missing values to use in filling in data [default = 10]
maximum gap to fill in [default = 10]
Returns a list of seasonally detrended flow data. You should save the resulting list as flow.detrended for use with baytrends. This function also creates diagnostic plots that can be saved to a report when this function is called from an .Rmd script.
This function returns a list of seasonally detrended flow and companion statistics; and relies on USGS' dataRetrieval package to retrieve daily flow data.
It is the user responsibility to save the resulting list as flow.detrended for integration with baytrends.
For the purposes of baytrends, it is expected that the user would identify all USGS gages that are expected to be evaluated so that a single data file is created. To best match up with water quality data, we recommend retrieving flow data for one year prior to the first year of water quality data. This allows for creating a time-averaged flow data set and not loose the first few months of water quality data due to lack of matching flow data. Data retrievals should also be made in light of the time needed by the USGS to review and approve their flow records.
After retrieval, the following computation steps are performed to create a data frame for each USGS gage (the data frame naming convention is qNNNNNNNN where NNNNNNNN is the USGS gage ID):
1) The daily flow data are converted to cubic meters per second [cms] and stored as the variable q.
2) The day of year (doy) is added to the data set. We use a 366 day calendar regardless of leap year.
3) The Log (ln) flow is computed and stored as LogQ.
4) A seasonal GAM, i.e., gamoutput <- gam(LogQ ~ s(doy, bs='cc')) is evaluated and the predicted values stored as qNNNNNNNN.gam.
5) The GAM residuals, i.e., "residuals(gamoutput)" are extracted and stored as the variable, d1.
6) Based on the specifications for dvAvgWinSel, dvAvgWgtSel, and dvAvgSidesSel, the values of d1 are time averaged and additional variables dxxx are added to the data frame where xxx corresponds to list of averaging windows specified in dvAvgWinSel. These values of dxxx are used in GAMs that include a hydrologic independent variable.
After the above data frame is created, the following four (4) additional data frames are created for each USGS gage and combined into a list named qNNNNNNNN.sum:
mean – For each doy (i.e., 366 days of year), the mean across all years for each value of d in the above data frame, qNNNNNNNN.
sd – For each doy (i.e., 366 days of year), the standard deviation across all years for each value of d in the above data frame, qNNNNNNNN.
nobs – For each doy (i.e., 366 days of year), the number of observations across all years for each value of d in the above data frame , qNNNNNNNN.
lowess.sd – Lowess smoothed standard deviations. (These values are used for computing confidence intervals in the flow averaged GAM.)
The process of creating the above data frame, qNNNNNNNN, and list, qNNNNNNNN.sum, is repeated for each USGS gage and combined together in a single list. The beginning of the list includes meta data documenting the retrieval parameters.
This function can be used in conjunction with an RMD file to knit (create) a report (DOCX or HTML).
if (FALSE) { # \dontrun{
# Define Function Inputs
usgsGageID <- c("01491000", "01578310")
siteName <- c("Choptank River near Greensboro, MD",
"Susquehanna River at Conowingo, MD")
yearStart <- 1983
yearEnd <- 2016
dvAvgWinSel <- c(1, 5, 10, 15, 20, 30, 40, 50, 60, 90, 120, 150, 180, 210)
dvAvgWgtSel <- "uniform"
dvAvgSidesSel <- 1
lowess.f <- 0.2
# Run Function
flow.detrended <- detrended.flow(usgsGageID, siteName, yearStart, yearEnd
, dvAvgWinSel, dvAvgWgtSel, dvAvgSidesSel
, lowess.f)
} # }