In markwh/rcmodel: Calibrate and Validate Rating Curves for Riverine Load Estimation

knitr::opts_chunk$set(echo = TRUE)

I've decided I need to start keeping a notebook for this project, even though the directory structure is somewhat limited as it's a package.

Picking up where I've left off in troubleshooting the data retrieval workflow. See vignettes/retrieval

unique(qData$MonitoringLocationIdentifier)

qData %>% 
  filter(grepl("USGS", MonitoringLocationIdentifier)) %>% 
  summary()

Looks like I need to retrieve USGS flows using pcodes.

names <- getWQPCodes("Characteristicname")
glimpse(names)

filter(names, grepl("NWIS", providers),
       grepl("flow", value)) %>% 
  print(n = 100)

pcodes <- dataRetrieval::pCodeToName

dplyr::filter(pcodes,
       grepl("Discharge", description, ignore.case = TRUE)) %>% 
  tbl_df() %>% 
  print(n = 100)

dplyr::filter(pcodes,
       grepl("flow", description, ignore.case = TRUE)) %>% 
  tbl_df() %>% 
  print(n = 100)

Can I just append the parameter codes to the flow strings in my getFlowData function?

constructWQPURL

Doesn't look like it. Plus they're switching between pCode and characteristicName in a potentially dangerous way.

Done. Try again.

flowsites <- unique(no3Data$MonitoringLocationIdentifier)
flowData <- getFlowData(flowsites, n = 100)

flowData %>% 
  filter(grepl("USGS", MonitoringLocationIdentifier)) %>% 
  summary()

That appears to have worked!

Now check out the multiple observations per day datasets.

qData %>% 
  filter(MonitoringLocationIdentifier == "USGS-01639000",
         # ResultMeasure.MeasureUnitCode == "mg/l",
         # ResultSampleFractionText == "Dissolved",
         ActivityStartDate == "1993-12-05") %>%
  transmute(Date = ActivityStartDate, datetime = ActivityStartDateTime,
            value = ResultMeasureValue, units = ResultMeasure.MeasureUnitCode) %>% 
  arrange(datetime) %>% 
  ggplot(aes(x = datetime, y = value)) +
  geom_point()

Nice, that looks good too.

foo <- qData %>% 
  transmute(detlimTxt = ResultDetectionConditionText, 
            station = MonitoringLocationIdentifier,
            char = CharacteristicName,
            Date = ActivityStartDate,
            detlim = DetectionQuantitationLimitMeasure.MeasureValue,
            detlimUnits = DetectionQuantitationLimitMeasure.MeasureUnitCode,
            value = ResultMeasureValue,
            units = ResultMeasure.MeasureUnitCode,
            comment = ResultCommentText)

foo %>% 
  filter(station == "USGS-01639000")

Looks like wqp_checkBDL is returning detLim as a list instead of a numeric.

wqpData <- flowData
convertTo <- "CFS"
convertedDLs <- convertUnits(
      x = wqpData$DetectionQuantitationLimitMeasure.MeasureValue,
      from = wqpData$DetectionQuantitationLimitMeasure.MeasureUnitCode,
      to = convertTo)

str(convertedDLs)

Fixed.

markwh/rcmodel documentation built on May 21, 2019, 12:26 p.m.