-
Notifications
You must be signed in to change notification settings - Fork 11
/
README.Rmd
245 lines (162 loc) 路 16.9 KB
/
README.Rmd
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
96
97
98
99
100
101
102
103
104
105
106
107
108
109
110
111
112
113
114
115
116
117
118
119
120
121
122
123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
144
145
146
147
148
149
150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
---
output:
github_document:
html_preview: true
---
```{r, echo = FALSE, warning=FALSE}
knitr::opts_chunk$set(
collapse = TRUE,
comment = "#>",
fig.path = "man/figures/",
warning = FALSE,
message = FALSE
)
library(fasstr)
```
# fasstr <img src="man/figures/fasstr.png" align="right" />
<!-- badges: start -->
[](https://github.com/bcgov/repomountie/blob/master/doc/lifecycle-badges.md)
[](https://opensource.org/licenses/Apache-2.0)
[](https://github.com/bcgov/fasstr/actions/workflows/R-CMD-check.yaml)
[](https://cran.r-project.org/package=fasstr) [](https://CRAN.R-project.org/package=fasstr) [](https://cran.r-project.org/web/checks/check_results_fasstr.html)
<!-- badges: end -->
</div>
The Flow Analysis Summary Statistics Tool for R ('fasstr') is a set of [R](http://www.r-project.org) functions to tidy, summarize, analyze, trend, and visualize streamflow data. This package summarizes continuous daily mean streamflow data into various daily, monthly, annual, and long-term statistics, completes annual trends and frequency analyses, in both table and plot formats.
### Reference
[fasstr package 馃摝 home page and reference guide](https://bcgov.github.io/fasstr/)
### Features
This package provides functions for streamflow data analysis, including:
- data tidying (to prepare data for analyses; `add_*` and `fill_*` functions),
- data screening (to identify data range, outliers and missing data; `screen_*` functions),
- calculating summary statistics (long-term, annual, monthly and daily statistics; `calc_*`functions),
- computing analyses (volume frequency analyses and annual trending; `compute_*` functions), and,
- visualizing (data plotting the various statistics; `plot_*` functions).
Useful features of functions include:
- the integration of the `tidyhydat` package to pull streamflow data from a Water Survey of Canada [HYDAT](https://www.canada.ca/en/environment-climate-change/services/water-overview/quantity/monitoring/survey/data-products-services/national-archive-hydat.html) database for analyses;
- arguments for filtering of years and months in analyses and plotting;
- choosing the start month of your water year;
- selecting for rolling day averages (e.g. 7-day rolling average); and,
- choosing how missing dates are handled, amongst others.
This package is maintained by the Water Management Branch of the British Columbia Ministry of Water, Land and Resource Stewardship.
### Installation
You can install `fasstr` directly from [CRAN](https://cran.r-project.org/package=fasstr):
```{r, echo=TRUE, eval=FALSE}
install.packages("fasstr")
```
To install the development version from [GitHub](https://github.com/bcgov/fasstr), use the [`remotes`](https://cran.r-project.org/package=remotes) package then the `fasstr` package:
```{r, echo=TRUE, eval=FALSE}
if(!requireNamespace("remotes")) install.packages("remotes")
remotes::install_github("bcgov/fasstr")
```
To use the `station_number` argument and pull data directly from a [Water Survey of Canada HYDAT database](https://www.canada.ca/en/environment-climate-change/services/water-overview/quantity/monitoring/survey/data-products-services/national-archive-hydat.html) into `fasstr` functions, download a HYDAT file using the following code:
```{r, echo=TRUE, eval=FALSE}
tidyhydat::download_hydat()
```
### Using fasstr
There are several vignettes and a cheatsheet to provide more information on the usage of `fasstr` functions and how to customize various argument options.
- [Getting Started](https://bcgov.github.io/fasstr/articles/fasstr.html)
- [Users Guide](https://bcgov.github.io/fasstr/articles/fasstr_users_guide.html)
- [Computing an Annual Trends Analysis](https://bcgov.github.io/fasstr/articles/fasstr_trending_analysis.html)
- [Computing a Volume frequency Analysis](https://bcgov.github.io/fasstr/articles/fasstr_frequency_analysis.html)
- [Computing a Full fasstr Analysis](https://bcgov.github.io/fasstr/articles/fasstr_full_analysis.html)
- [Internal fasstr Workflows](https://bcgov.github.io/fasstr/articles/fasstr_under_the_hood.html)
#### Cheatsheet
<a href="https://github.com/bcgov/fasstr/raw/main/fasstr_cheatsheet.pdf"><img src="man/figures/fasstr_cheatsheet.png" width="630" height="252"/></a>
#### Data Input
All functions in `fasstr` require a daily mean streamflow data set from one or more hydrometric stations. Long-term and continuous data sets are preferred for most analyses, but seasonal and partial data can be used. Other daily time series data, like temperature, precipitation or water levels, may also be used, but with certain caution as some calculations/conversions are based on units of streamflow (cubic metres per second). Data is provided to each function using the either the `data` argument as a data frame of flow values, or the `station_number` argument as a list of Water Survey of Canada HYDAT station numbers.
When using the `data` option, a data frame of daily data containing columns of dates (YYYY-MM-DD in date format), values (mean daily discharge in cubic metres per second in numeric format), and, optionally, grouping identifiers (character string of station names or numbers) is called. By default the functions will look for columns identified as 'Date', 'Value', and 'STATION_NUMBER', respectively, to be compatible with the 'tidyhydat' defaults, but columns of different names can be identified using the `dates`, `values`, `groups` column arguments (ex. `values = Yield_mm`). The following is an example of an appropriate data frame (STATION_NUMBER not required):
```{r setup, include = FALSE}
data <- tidyhydat::hy_daily_flows("08NM116")
data <- data[,c(1,2,4)]
```
```{r flow_data, echo=FALSE}
head(data.frame(data))
```
Alternatively, you can directly pull a flow data set directly from a HYDAT database (if installed) by providing a list of station numbers in the `station_number` argument (ex. `station_number = "08NM116"` or `station_number = c("08NM116", "08NM242")`) while leaving the data arguments blank. A data frame of daily streamflow data for all stations listed will be extracted using `tidyhydat` and then `fasstr` calculations will produce results of the functions.
This package allows for multiple stations (or other groupings) to be analyzed in many of the functions provided identifiers are provided using the `groups` column argument (defaults to STATION_NUMBER). If grouping column doesn't exist or is improperly named, then all values listed in the `values` column will be summarized.
#### Function Types
##### Tidying
These functions, start with either `add_*` or `fill_*`, add columns and rows, respectively, to streamflow data frames to help set up your data for further analysis. Examples include adding rolling means, adding date variables (WaterYear, Month, DayofYear, etc.), adding basin areas, adding columns of volumetric discharge and water yields, and filling dates with missing flow values with `NA`.
##### Analysis
The analysis functions summarize your discharge values into various statistics. `screen_*` functions summarize annual data for outliers and missing dates. `calc_*` functions calculate daily, monthly, annual, and long-term statistics (e.g. mean, median, maximum, minimum, percentiles, amongst others) of daily, rolling days, and cumulative flow data. `compute_*` functions also analyze data but produce more in-depth analyses, like frequency and trending analysis, and may produce multiple plots and tables as a result. All tables are in tibble data frame formats. Can use `write_flow_data()` or `write_results()` to customize saving tibbles to a local drive.
##### Visualization
The visualization functions, which begin with `plot_*`, plot the various summary statistics and analyses as a way to visualize the data. While most plotting function statistics can be customized, some come pre-set with statistics that cannot be changed. Plots can be further modified by the user using the `ggplot2` package and its functions. All plots functions produce lists of plots (even if just one produced). Can use `write_plots()` to customize saving the lists of plots to a local drive (within folders or PDF documents).
#### Function Options
##### Daily Rolling Means
If certain n-day rolling mean statistics are desired to be analyzed (e.g. 3- or 7-day rolling means) some functions provide the ability to select for that as function arguments (e.g. `rolling_days = 7` and `rolling_align = "right"`). The rolling day align is the placement of the date amongst the n-day means, where "right" averages the day-of and previous n-1 days, "centre" date is in the middle of the averages, and "left" averages the day-of and the following n-1 days. For your own analyses you can add rolling means to your data set using the `add_rolling_means()` function.
##### Year and Month Filtering
To customize your analyses for specific time periods, you can designate the start and end years of your analysis using the `start_year` and `end_year` arguments and remove any unwanted years (for partial data sets for example) by listing them in the `excluded_years` argument (e.g. `excluded_years = c(1990, 1992:1994)`). Alternatively, some functions have an argument called `complete_years` that summarizes data from just those years which have complete flow records. Some functions will also allow you to select the months of a year to analyze, using the `months` argument, as opposed to all months (if you want just summer low-flows, for example). Leaving these arguments blank will result in the summary/analysis of all years and months of the provided data set.
To group analyses by water, or hydrologic, years instead of calendar years, if desired, you can set `water_year_start` within most functions to another month than 1 (for January). A water year can be defined as a 12-month period that comprises a complete hydrologic cycle (wet seasons can typically cross calendar year), typically starting with the month with minimum flows (the start of a new water recharge cycle). If another start month is desired, you can choose it using the `water_year_start` argument (numeric month). The water year identifier is designated by the year it ends in (e.g. a water year from Oct 1, 1999 to Sep 30, 2000 is designated as 2000). Start, end and excluded years will be based on the specified water year.
For your own analyses, you can add date variables to your data set using the `add_date_variables()` or `add_seasons()` functions.
##### Drainage Basin Area
Yield runoff statistics (in millimetres) calculated in the some of the functions require an upstream drainage basin area (in sq. km) using the `basin_area` argument. If no basin areas are supplied, all yield results will be `NA`. To apply a basin area (10 sqkm for example) to all daily observations, set the argument as `basin_area = 10`. If there are multiple stations or groups to apply multiple basin areas (using the `groups` argument), set them individually using this option: `basin_area = c("08NM116" = 795, "08NM242" = 22)`. If a STATION_NUMBER column exists with HYDAT station numbers, the function will automatically use the basin areas provided in HYDAT, if available, so `basin_area` is not required. For your own analyses, you can add basin areas to your data set using the `add_basin_area()` function.
##### Handling Missing Dates
With the use of the `ignore_missing` argument in most functions, you can decide how to handle dates with missing flow values in calculations. When you set `ignore_missing = TRUE` a statistic will be calculated for a given year, all years, or month regardless of if there are missing flow values. When `ignore_missing = FALSE` the returned value for the period will be `NA` if there are missing values. To allow some missing dates and still calculate statistics, some functions also including the `allowed_missing` argument where you provide a percentage (0 to 100) of missing days per time period.
Some functions have an argument called `complete_years` which can be used, when set to `TRUE`, to filter out years that have partial data sets (for seasonal or other reasons) and only years with full data are used to calculate statistics.
### Examples
#### Summary statistics example: long-term statistics
To determine the long-term summary statistics of daily data for each month (mean, median, maximum, minimum, and some percentiles) you can use the `calc_longterm_daily_stats()` function. If the 'Mission Creek near East Kelowna' hydrometric station is of interest you can list the station number in the `station_number` argument to obtain the data (if `tidyhydat` and HYDAT are installed). Statistics over several months can also be calculated, if of interest. See the summer statistics (from July to September) in this example.
```{r example1}
calc_longterm_daily_stats(station_number = "08NM116",
start_year = 1981,
end_year = 2010,
custom_months = 7:9,
custom_months_label = "Summer")
```
#### Plotting example: daily summary statistics
To visualize the daily streamflow patterns on an annual basis, the `plot_daily_stats()` function will plot out various summary statistics for each day of the year. Data can also be filtered for certain years of interest (a 1981-2010 normals period for this example) using the `start_year` and `end_year` arguments. We can also compare individual years against the statistics using `add_year` argument like below.
```{r plot1, fig.height = 4, fig.width = 10}
plot_daily_stats(station_number = "08NM116",
start_year = 1981,
end_year = 2010,
log_discharge = TRUE,
add_year = 1991)
```
#### Plotting example: flow duration curves
Flow duration curves can be produced using the `plot_flow_duration()` function.
```{r plot2, fig.height = 4, fig.width = 7}
plot_flow_duration(station_number = "08NM116",
start_year = 1981,
end_year = 2010)
```
#### Analysis example: low-flow frequency analysis
This package also provides a function, `compute_annual_frequencies()`, to complete a volume frequency analysis by fitting annual minimums or maximums to Log-Pearson Type III or Weibull probability distributions. See the volume frequency analyses documentation for more information. For this example, the 7-day low-flow quantiles are calculated for the Mission Creek hydrometric station using the Log-Pearson Type III distribution and method of moments fitting method (both default). With this, several low-flow indicators can be determined (i.e. 7Q5, 7Q10).
```{r example2}
freq_results <- compute_annual_frequencies(station_number = "08NM116",
start_year = 1981,
end_year = 2010,
roll_days = 7,
fit_distr = "PIII",
fit_distr_method = "MOM")
freq_results$Freq_Fitted_Quantiles
```
The probability of observed extreme events can also be plotted (using selected plotting position) along with the computed quantiles curve for comparison.
```{r plot3, fig.height = 4, fig.width = 7}
freq_results <- compute_annual_frequencies(station_number = "08NM116",
start_year = 1981,
end_year = 2010,
roll_days = c(1,3,7,30))
freq_results$Freq_Plot
```
### Project Status
This package is set for delivery. This package is maintained by the Water Management Branch of the British Columbia Ministry of Water, Land and Resource Stewardship.
### Getting Help or Reporting an Issue
To report bugs/issues/feature requests, please file an [issue](https://github.com/bcgov/fasstr/issues/).
### How to Contribute
If you would like to contribute to the package, please see our
[CONTRIBUTING](CONTRIBUTING.md) guidelines.
Please note that this project is released with a [Contributor Code of Conduct](CODE_OF_CONDUCT.md). By participating in this project you agree to abide by its terms.
### License
```
Copyright 2023 Province of British Columbia
Licensed under the Apache License, Version 2.0 (the "License");
you may not use this file except in compliance with the License.
You may obtain a copy of the License at
http://www.apache.org/licenses/LICENSE-2.0
Unless required by applicable law or agreed to in writing, software
distributed under the License is distributed on an "AS IS" BASIS,
WITHOUT WARRANTIES OR CONDITIONS OF ANY KIND, either express or implied.
See the License for the specific language governing permissions and
limitations under the License.
```
---