sola

The goal of sola is to provide tools for solar resource assessment and utilization. This package implements the function of calculating solar radiation based on sunshine hours, latitude, and other related parameters. Compared with other existing packages, the main advantage of sola is the use of vectorization operations, allowing for efficient calculation of large datasets. The package is continuously updated to include more functions relevant to solar energy resource assessment.

Installation

You can install the development version of sola from GitHub with:

# install.packages("devtools")
devtools::install_github("renliang1996/sola")

Example

Basic Calculation of Solar Radiation

Here is a simple example of calculating solar radiation:

library(sola)

# Calculate solar radiation for March 15, 2023, at latitude 35 with 8 hours of sunshine
calc_Rs(lat = 35.0, date = as.Date("2023-03-15"), ssd = 8)
#> [1] 17.39019

Vectorization calculation

To address the limitation of certain packages that can only input a single value, we can input a vector for computation:

latitudes <- c(35, -15, 50)
dates <- as.Date(c("2023-03-15", "2024-06-21", "2023-11-01"))
sunshine_hours <- c(8, 10, 6)

calc_Rs(lat = latitudes, date = dates, ssd = sunshine_hours)
#> [1] 17.390195 18.596638  7.233635

Performance Testing with Large Datasets

The following example demonstrates how sola can handle a dataset with 10,000 records efficiently:

set.seed(123)
test_lat <- runif(10000, -90, 90)
test_date <- as.Date("2023-01-01") + sample(1:365, 10000, replace = TRUE)
test_ssd <- runif(10000, 0, 24)

system.time({
  results <- calc_Rs(lat = test_lat, date = test_date, ssd = test_ssd)
})
#> 用户 系统 流逝 
#> 0.15 0.00 0.15

head(results)
#> [1] 37.187776 33.641994 23.242994  1.107624  0.000000  4.139500

Advanced Usage

Calculating Solar Radiation from NetCDF Data

The radNC function computes daily solar radiation for each pixel in a SpatRaster object using sunshine duration data from a NetCDF file, applying the Angstrom-Prescott model. Here’s a step-by-step guide:

library(terra)
#> terra 1.7.78

# Assuming 'ssd.nc' is a NetCDF file with the required data
filename <- system.file("extdata", "ssd.nc", package = "sola")
ssd <- rast(filename)
plot(ssd)

Before running the computation, ensure that the necessary libraries are loaded and data is correctly formatted.

#Set the negative values of SSD to 0, and if na_neg is TRUE, then set it to NA.
result_raster <- radNC(ssd, na_neg = FALSE) 
#>   |                                                                              |                                                                      |   0%  |                                                                              |==============                                                        |  20%Estimated remaining time: 2.8 secs  |                                                                              |============================                                          |  40%Estimated remaining time: 2.2 secs  |                                                                              |==========================================                            |  60%Estimated remaining time: 1.5 secs  |                                                                              |========================================================              |  80%Estimated remaining time: 0.73 secs  |                                                                              |======================================================================| 100%Estimated remaining time: 0 secs
#> 
#> Total computation time: 3.8 secs

# Plot the result
plot(result_raster)

Note that the running time depends on the size of your file.

This example demonstrates how to use the radNC function to compute solar radiation from NetCDF data. Make sure to have the terra package installed and the appropriate NetCDF file available for this to work.