mpdougherty / razviz Goto Github PK

Add the ability to change the output report file name

There are more background parameters that could be added to longitudinal profiles, particularly if the hydraulic parameter is not water surface elevation. It would be nice to be able to add ....

• specific historical floods with a label of the data and return period of the event (this likely could be incorporated into the highwater mark)
• river features that cover multiple river miles (a lake for example)
• spillway connections

Brian asked if we could add confidence intervals to water surface intervals (WSE).

Add the functions required to produce the Goodness of Fit (GOF) tables.

Many users do not like the look of the smoothed levee elevation line on the longitudinal profile plot. Therefore, this should be set to an optional parameter that users can choose to display.

There must be missing data remaining in the long form of the data. Probably need to remove those before plotting.

The last step of the Vignette with the code to make the plots is not in the exported vignette file, but is in the rmd file.

Changed the linetypes to dotted for observed data
Added a legend under both plots
Changed the page headings to include Gage location and removed the run type and number

function (plot_number, hydrograph_df, hg_plot_pages)
{
plot_df <- dplyr::filter(hg_plot_pages, plot == plot_number)
plot <- plot_df$plot
run_type <- as.character(plot_df$Run_type)
run_num <- as.numeric(as.character(plot_df$Run_num))
river_sta <- as.numeric(as.character(plot_df$River_Sta))
hg_df_plot <- dplyr::filter(hydrograph_df, River_Sta == river_sta, Run_type == run_type, Run_num == run_num)

ws <- dplyr::filter(hg_df_plot, Type == "Obs_WS" | Type == "WS_Elev")
q <- dplyr::filter(hg_df_plot, Type == "Obs_Q" | Type == "Model_Q")

WS_cols <- c(WS_Elev = "darkslategray4", Obs_WS = "coral3")
WS_line <- c(WS_Elev = "solid", Obs_WS = "dashed")

Discharge_cols <- c(Model_Q = "darkslategray4", Obs_Q = "coral3")
Discharge_line <- c(Model_Q = "solid", Obs_Q = "dashed")

WS_labels <- c(WS_Elev = "Modeled", Obs_WS = "Observed")
Discharge_labels <- c(Model_Q = "Modeled", Obs_Q = "Observed")

ws_plot <- ggplot(data = ws, aes(x = date, y = value, color = Type), na.rm = TRUE) +
geom_line(aes(linetype=Type) ,size = 1) +
facet_grid(. ~ Event, scales = "free") +
theme_bw() +
scale_color_manual(values = WS_cols, labels = WS_labels) +
scale_linetype_manual(values = WS_line, labels = WS_labels ) +
theme(legend.position = "bottom",
legend.title = element_blank(), axis.title.x = element_blank(), axis.text.x = element_text(angle = 50, hjust = 1)) +
scale_x_datetime(date_labels = "%e %b", date_breaks = "7 days", date_minor_breaks = "1 day") + labs(y = "Elevation (NAVD88 feet)")

q_plot <- ggplot(data = q, aes(x = date, y = value/1000,
color = Type), na.rm = TRUE) +
geom_line(aes(linetype=Type) ,size = 1) +
facet_grid(. ~ Event, scales = "free") +
theme_bw() +
scale_color_manual(values = Discharge_cols, labels = Discharge_labels) +
scale_linetype_manual(values = Discharge_line, labels = Discharge_labels) +
theme(legend.position = "bottom",
legend.title = element_blank(),
axis.title.x = element_blank(), axis.text.x = element_text(angle = 50,hjust = 1)) +
scale_x_datetime(date_labels = "%e %b", date_breaks = "7 days", date_minor_breaks = "1 day") + labs(y = "Discharge (1000 cubic feet per second)")

#removed from title - ws$Run_type,"#", ws$Run_num,

title <-textGrob(label = paste(trimws(ws$River), "River,",
trimws(ws$Reach), "Reach, River Mile ",
ws$River_Sta, "\n",
"Gage", ws$Gage, sep = " "),
x = unit(0, "lines"), y = unit(0,"lines"), hjust = 0, vjust = 0, gp = gpar(fontsize = 12))

hg_plot <- grid.arrange(ws_plot, q_plot, nrow = 2, ncol = 1,
widths = unit(7, "in"), heights = unit(c(5, 5), "in"),
top = title, clip = FALSE)
return(hg_plot)
}

Jon Hendrickson asked if there is a was to display other parameters.

• velocity

This will keep the legends in the same order throughout the report.

function (hydrograph_df)
{
long_hydrograph <- tidyr::pivot_longer(hydrograph_df, cols = c(WS_Elev,
Obs_WS, Model_Q, Obs_Q), names_to = "Type", values_to = "value")
long_hydrograph$Type <- factor(long_hydrograph$Type,levels = c("WS_Elev","Obs_WS", "Model_Q", "Obs_Q"), labels = c("WS_Elev","Obs_WS", "Model_Q", "Obs_Q"))
return(long_hydrograph)
}

If bridge elevations are not available/not wanted, the longitudinal profile will not plot

Error: Error: Discrete value supplied to continuous scale"

Can the longitudinal profile plot function be change to include an if statement that similar to the smoothing function for the levee profiles? This way the bridge elevations are an option input.

Develop code to automatically decide what the min and max stage for gage box should be rather than having to iterate.

Create function to show which files are being pulled for observed data - right now this functionality is within the import_observed_data.R, but the function does not return a specific list of the pathnames.

#loop through each of the Formatted Unsteady Flow Files pathnames and pull the files out of dss
for(p in 1:length(plan_events)){
Formatted_UnsteadyFlowFile <- Formatted_UnsteadyFlowFileList[[p]]
event_year <- plan_events[p]

#Recreate the paths for those DSS files
Observed_Pathnames_Formatted <-c()

for(j in 1:nrow(Formatted_UnsteadyFlowFile)){ #for each row in the data frame
  Formmated_Pathname <- paste0("/",Formatted_UnsteadyFlowFile[j,4],"/",
    Formatted_UnsteadyFlowFile[j,5],"/",
    Formatted_UnsteadyFlowFile[j,6],"/",
    "/",
    Formatted_UnsteadyFlowFile[j,8],"/",
    Formatted_UnsteadyFlowFile[j,9],"/")

  Observed_Pathnames_Formatted  <- c(Observed_Pathnames_Formatted , Formmated_Pathname)
}

assign(paste("Observed_Pathnames_Formatted",Plan_Event, sep="_") ,Observed_Pathnames_Formatted)

Rather than pull all the data - could add parameters for a start and end river mile - likely would need to filter the list coming in from unsteady flow file

Would it be possible to automatically extract inputs from RAS so you don't have to build all of the input .csv files?