The goal of wtkapi is to get data from the Gridded Atmospheric Wind Integration National Dataset Toolkit in tidy format to leverage technical analysis of atmospheric data.
It is required to sign up for an API key on the NREL Developer Network. Please also consider the API call rate limits.
You can install the development version of wtkapi from GitHub with:
# install.packages("devtools")
devtools::install_github("floresfdev/wtkapi")At the moment this package is heavy on dependencies (see Imports on DESCRIPTION), so in order to have more control over your environment, avoid install/upgrade dependencies with:
devtools::install_github("floresfdev/wtkapi", dependencies = FALSE, upgrade_dependencies = FALSE)This is a basic example which shows you how to get data:
library(dplyr)
library(lubridate)
library(ggplot2)
library(glue)
library(wtkapi)
# API parameters
endpoint <- "https://developer.nrel.gov/api/hsds/"
host <- "/nrel/wtk-us.h5"
api_key <- readr::read_lines(here::here("api_key.txt"))
# Get the info of the available datasets
datasets <- get_datasets(endpoint, host, api_key)
# Info to select the year 2007, with 1-hour resolution data
datetime_info <-
compute_datetime_info(datetime_from = "2007-01-01 00:00:00",
datetime_to = "2007-12-31 23:00:00",
datetime_step = 1)
# Info to select the location: New York
latitude <- 40.7128
longitude <- -74.0059
# Info to select the datasets as columnar data
datasets$title## [1] "DIF" "DNI"
## [3] "GHI" "coordinates"
## [5] "datetime" "inversemoninobukhovlength_2m"
## [7] "precipitationrate_0m" "pressure_0m"
## [9] "pressure_100m" "pressure_200m"
## [11] "relativehumidity_2m" "status"
## [13] "temperature_100m" "temperature_10m"
## [15] "temperature_120m" "temperature_140m"
## [17] "temperature_160m" "temperature_200m"
## [19] "temperature_2m" "temperature_40m"
## [21] "temperature_60m" "temperature_80m"
## [23] "winddirection_100m" "winddirection_10m"
## [25] "winddirection_120m" "winddirection_140m"
## [27] "winddirection_160m" "winddirection_200m"
## [29] "winddirection_40m" "winddirection_60m"
## [31] "winddirection_80m" "windspeed_100m"
## [33] "windspeed_10m" "windspeed_120m"
## [35] "windspeed_140m" "windspeed_160m"
## [37] "windspeed_200m" "windspeed_40m"
## [39] "windspeed_60m" "windspeed_80m"
dataset_titles <- c("temperature_10m", "windspeed_10m", "winddirection_10m",
"temperature_40m", "windspeed_40m", "winddirection_40m",
"temperature_80m", "windspeed_80m", "winddirection_80m")
# Get data from the API
wind_temp_2007 <-
get_dataset(datasets,
api_key,
dataset_titles,
datetime_info,
latitude,
longitude)
# Number of rows: 8760 = 365 days * 24 hours
nrow(wind_temp_2007)## [1] 8760
# Check first and last records
head(wind_temp_2007)## # A tibble: 6 x 12
## datetime latitude longitude temperature_10m windspeed_10m
## <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 2007-01-01 00:00:00 40.7 -74.0 278. 1.42
## 2 2007-01-01 01:00:00 40.7 -74.0 278. 1.19
## 3 2007-01-01 02:00:00 40.7 -74.0 278. 1.22
## 4 2007-01-01 03:00:00 40.7 -74.0 279. 1.04
## 5 2007-01-01 04:00:00 40.7 -74.0 279. 1.05
## 6 2007-01-01 05:00:00 40.7 -74.0 280. 1.91
## # ... with 7 more variables: winddirection_10m <dbl>,
## # temperature_40m <dbl>, windspeed_40m <dbl>, winddirection_40m <dbl>,
## # temperature_80m <dbl>, windspeed_80m <dbl>, winddirection_80m <dbl>
tail(wind_temp_2007)## # A tibble: 6 x 12
## datetime latitude longitude temperature_10m windspeed_10m
## <dttm> <dbl> <dbl> <dbl> <dbl>
## 1 2007-12-31 18:00:00 40.7 -74.0 278. 6.65
## 2 2007-12-31 19:00:00 40.7 -74.0 278. 6.67
## 3 2007-12-31 20:00:00 40.7 -74.0 278. 6.62
## 4 2007-12-31 21:00:00 40.7 -74.0 278. 5.60
## 5 2007-12-31 22:00:00 40.7 -74.0 277. 4.10
## 6 2007-12-31 23:00:00 40.7 -74.0 277. 4.00
## # ... with 7 more variables: winddirection_10m <dbl>,
## # temperature_40m <dbl>, windspeed_40m <dbl>, winddirection_40m <dbl>,
## # temperature_80m <dbl>, windspeed_80m <dbl>, winddirection_80m <dbl>
# Plots
plot_subtitle <- glue("2007, New York (lat {latitude}, lon {longitude})")
## Daily min/max temperature at 10m
wind_temp_2007 %>%
## Convert temperature from Kelvin to Celsius
mutate(temperature_10m = temperature_10m - 273.15) %>%
## Get the min and max temperature for each day
mutate(date = as_date(datetime)) %>%
group_by(date) %>%
summarize(min_temperature_10m = min(temperature_10m),
max_temperature_10m = max(temperature_10m)) %>%
## Plot timeseries
ggplot() +
geom_line(aes(x = date,
y = min_temperature_10m),
color = "blue") +
geom_line(aes(x = date,
y = max_temperature_10m),
color = "red") +
labs(x = "Date",
y = "Temperature (°C)",
title = "Daily min/max temperature at 10m",
subtitle = plot_subtitle,
caption = "Blue = daily min - Red = daily max") +
theme_classic()
## Daily average wind speed at 80m
wind_temp_2007 %>%
## Get the average of windspeed for each day
mutate(date = as_date(datetime)) %>%
group_by(date) %>%
summarize(avg_windspeed_80m = mean(windspeed_80m)) %>%
## Plot timeseries
ggplot() +
geom_line(aes(x = date,
y = avg_windspeed_80m)) +
labs(x = "Date",
y = "Wind speed (m/s)",
title = "Daily average wind speed at 80m",
subtitle = plot_subtitle) +
theme_classic()
## Histogram of hourly wind speed at 80m
ggplot(wind_temp_2007) +
geom_histogram(aes(x = windspeed_80m),
fill = "white",
color = "black") +
labs(x = "Wind speed (m/s)",
y = "Count",
title = "Histogram of hourly wind speed at 80m",
subtitle = plot_subtitle) +
theme_classic()
Please note that this project is released with a Contributor Code of Conduct. By participating in this project you agree to abide by its terms.
Copyright (c) 2018, Fernando Flores
License TBD.