loadCalculations: Calculate annual loads and summarize trends
In seawaveQ: SEAWAVE-Q Model

Description Usage Arguments Format Details Value Note Author(s) References Examples

Function to calculate pesticide loads in kilograms per year and summarize trends.

loadCalculations(
  dailyDat,
  pestPredict,
  modRes,
  concTrends,
  yrtype = 1,
  alpha = 0.1
)

`dailyDat`	is the daily streamflow data in the form of a data frame with three columns representing a station ID, date, and streamflow.
`pestPredict`	is the continuous (daily) estimation of pesticide concentrations for one or more pesticides at a single site. This should be in the form of the fourth element of the list returned by `fitswavecav`.
`modRes`	is the first element of the list returned by `fitswavecav` and includes the scale parameter for one or more pesticide trend models at a single site. The scale parameter is used in the bias correction.
`concTrends`	the SEAWAVE-Q trend in flow-normalized annual load. Cannot be different (computationally) from the trend in flow-normalized annual concentration when there is no trend in flow (Oelsner and others, 2017).
`yrtype`	allows one to calculate annual loads based on a calendar year or a water year, where a water year is the 12-month period October 1 through September 30 designated by the calendar year in which it ends. A yrtype of 1 represents a calendar year and is the default because that is the way the original model was developed. A yrtype of 2 represents a water year.
`alpha`	is the significance level or alpha value for statistical significance and confidence intervals.

The first data frame returned has one row for each pesticide-year at a particular site and four columns. The general format is as follows:

pstaid	character	the station identification number
pcode	character	the parameter code for which a load was calculated
year or wyear	numeric	the year or water year for which a load was calculated
load	numeric	the load in kilograms per year

The second data frame returned has one row for each pesticide at a particular site and 11 columns. The general format is as follows:

pcode	character	the parameter code for which a load trend was calculated
mclass	numeric	a value of 1 or 2
mclass	numeric	a value of 1 or 2
alpha	numeric	a significance level
ltndPpor	numeric	the load trend in percent over the period of record
luciPpor	numeric	the load upper confidence interval for the trend in
		percent over the period of record
llciPpor	numeric	the load lower confidence interval for the trend in
		percent over the period of record
baseLoad	numeric	the base load, the load for the first year of trend period
ltndOrigPORPercentBase	numeric	the load trend in original units over
		the period of record
		(calculation based on percent per year and base load)
luciOrigPORPercentBase	numeric	the load trend upper confidence interval
		for the trend in original units over the period of record
		(calculation based on percent per year and base load)
llciOrigPORPercentBase	numeric	the load trend lower confidence interval
		for the trend in original units over the period of record
		(calculation based on percent per year and base load)
ltndlklhd	numeric	is the load trend likelihood

Parameter load (mass) is the product of water-quality concentration (a mass per volume) and an associated streamflow rate (volume per time). This function generates an annual time series of pesticide loads on either a calendar year basis or a water year basis and summarizes load trends.

Two data frames, the first contains the annual loads, the second contains the trend summary.

In this load calculation function, daily pesticide concentration estimates provided by the fitswavecav function are corrected for retransformation bias (the concentration model is built on the base-10 logarithm of concentration; therefore, a bias correction is required when transforming back to the original units) and then used to calculate daily loads. The bias correction is based on the quasi-maximum likelihood estimator (Cohn and others, 1989) that was developed for natural logarithms, with an adjustment for the base-10 logarithm of the concentration. To calculate loads, the bias-corrected concentration estimates (assumed to be in micrograms per liter) are multiplied by daily streamflow and a constant, 0.892998605, which converts the load units (micrograms per liter * cubic feet per second) to kilograms per year. Daily loads are summed to annual values. See page 70 and equation 26 of Oelsner and others (2017) for further details regarding the load calculation and bias correction. Users may modify this function to convert to units other than kilograms per year.

Karen R. Ryberg

Cohn, T.A., DeLong, L.L., Gilroy, E.J., Hirsch, R.M., and Wells, D.K., 1989, Estimating constituent loads: Water Resources Research, v. 25, no. 5, p. 937–942.

Oelsner, G.P., Sprague, L.A., Murphy, J.C., Zuellig, R.E., Johnson, H.M., Ryberg, K.R., Falcone, J.A., Stets, E.G., Vecchia, A.V., Riskin, M.L., De Cicco, L.A., Mills, T.J., and Farmer, W.H., 2017, Water-quality trends in the Nation's rivers and streams, 1972–2012—Data preparation, statistical methods, and trend results (ver. 2.0, October 2017): U.S. Geological Survey Scientific Investigations Report 2017–5006, 136 p., https://doi.org/10.3133/sir20175006.

data(swData)
modMoRivOmaha <- combineData(qwdat = qwMoRivOmaha, cqwdat = cqwMoRivOmaha)
myfit1 <- fitswavecav(cdat = modMoRivOmaha, cavdat = cqwMoRivOmaha, 
tanm = "myfit1", pnames = c("04035", "04037", "04041"), yrstart = 1995, 
yrend = 2003, tndbeg = 1995, tndend = 2003, iwcav = c("flowa30", "flowa1"), 
dcol = "dates", qwcols = c("R", "P"))
MoRivOmahaLoadsYr <- loadCalculations(cqwMoRivOmaha[, 1:3], myfit1[[4]], 
myfit1[[1]], myfit1[[6]])
MoRivOmahaLoadsYr