NighttimeLights.jl
National Oceanic and Atmospheric Administration (NOAA) releases nighttime lights images produced using the Visible Infrared Imaging Radiometer Suite (VIIRS) since April 2012. Nighttime lights data had emerged as a useful tool to measure economic activity. Many researchers have established a correlation between prosperity and the brightness of a region. In many situations, nighttime lights generates measures with accuracy, latency and geographical resolution that are superior to conventional methods of measurement, such as GDP.
Using nighttime lights for economic analysis require cleaning of data over regions of interest. This is the first open source implementation of these procedures for nighttime lights.
This package was a foundation for a research paper, But clouds got in my way: bias and bias correction of VIIRS nighttime lights data in the presence of clouds. The paper diagnoses a source of bias in the data and responds to this problem with a bias correction scheme. Along with other mainstream methods of data cleaning, this method is also implemented in the package.
Foundations for nighttime lights data analysis is the research paper associated with this software package. Please cite it if you use the package.
@techreport{patnaik2022foundations,
title={Foundations for nighttime lights data analysis},
author={Patnaik, Ayush and Shah, Ajay and Thomas, Susan},
year={2022}
}The package is built on top of Rasters.jl, which provides cutting edge features to study raster data and perform routine tasks such as zonal statistics and plotting.
Installation
pkg> add NighttimeLightsGetting help
To get inquire about the Julia package, you can also email one of the authors of the package, Ayush Patnaik.
For questions regarding nighttime lights data, you can email EOG
Objects used in this documentation
In the following objects are used throughout the documentation.
1. radiance_datacube
48×101×131 Raster{Union{Missing, Float32},3} unnamed with dimensions:
X Mapped{Float64} Float64[72.78333535560002, 72.78750202230002, …, 72.97500202380003, 72.97916869050003] ForwardOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Y Mapped{Float64} Float64[19.266666220799998, 19.262499554099996, …, 18.8541662175, 18.8499995508] ReverseOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Ti Sampled{DateTime} DateTime[2012-04-01T00:00:00, …, 2023-01-01T00:00:00] ForwardOrdered Irregular Points
extent: Extent(X = (72.78125202225002, 72.98125202385003), Y = (18.84791621745, 19.26874955415), Ti = (DateTime("2012-04-01T00:00:00"), DateTime("2023-01-01T00:00:00")))
missingval: missing
crs: EPSG:4326
mappedcrs: EPSG:4326
parent:
[:, :, 1]
19.2667 19.2625 19.2583 … 18.8625 18.8583 18.8542 18.85
72.7833 3.36 1.46 1.2 0.52 0.43 0.41 0.38
72.7875 2.75 1.54 1.51 0.56 0.47 0.5 0.52
72.7917 3.2 2.19 1.71 0.53 0.46 0.5 0.5
72.7958 4.89 3.29 1.83 0.53 0.46 0.51 0.47
⋮ ⋱ ⋮
72.9625 25.05 22.82 10.87 1.72 1.79 1.72 1.88
72.9667 32.95 32.63 20.81 1.71 1.49 1.18 1.05
72.9708 28.36 29.57 38.93 1.46 1.28 1.39 1.22
72.975 15.14 23.27 27.99 1.49 1.26 1.12 1.28
72.9792 16.96 22.11 24.48 … 1.67 1.33 1.12 1.24
[and 130 more slices...]This is a datacube of radiance around Mumbai city in India. It has been created by cropping the global monthly nighttime lights data using a shapefile of the city. It contains the images from 2012-04 to 2023-01. It can be loaded using load_example().
2. ncfobs_datacube
48×101×131 Raster{UInt16,3} unnamed with dimensions:
X Mapped{Float64} Float64[72.78333535560002, 72.78750202230002, …, 72.97500202380003, 72.97916869050003] ForwardOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Y Mapped{Float64} Float64[19.266666220799998, 19.262499554099996, …, 18.8541662175, 18.8499995508] ReverseOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Ti Sampled{DateTime} DateTime[2012-04-01T00:00:00, …, 2023-01-01T00:00:00] ForwardOrdered Irregular Points
extent: Extent(X = (72.78125202225002, 72.98125202385003), Y = (18.84791621745, 19.26874955415), Ti = (DateTime("2012-04-01T00:00:00"), DateTime("2023-01-01T00:00:00")))
crs: EPSG:4326
mappedcrs: EPSG:4326
parent:
[:, :, 1]
19.2667 19.2625 … 18.8583 18.8542 18.85
72.7833 0x000b 0x000b 0x000a 0x000a 0x000a
72.7875 0x000b 0x000b 0x000b 0x000b 0x000b
72.7917 0x000c 0x000c 0x000a 0x000b 0x000b
72.7958 0x000c 0x000c 0x000a 0x000b 0x000b
⋮ ⋱ ⋮
72.9625 0x000c 0x000c 0x000a 0x0009 0x0009
72.9667 0x000c 0x000c 0x000a 0x0009 0x0008
72.9708 0x000c 0x000c 0x0009 0x000a 0x0009
72.975 0x000c 0x000c 0x0009 0x0009 0x000b
72.9792 0x000c 0x000c … 0x0009 0x0009 0x000b
[and 130 more slices...]This is a datacube of the number of cloud-free observations around Mumbai city in India. It has also been created by cropping the global monthly nighttime lights data using a shapefile of the city. It has the same dimensions as radiance_datacube. It also contains the images from 2012-04 to 2023-01. It can be loaded using load_example().
3. radiance_image
48×101 Raster{Union{Missing, Float32},2} unnamed with dimensions:
X Mapped{Float64} Float64[72.78333535560002, 72.78750202230002, …, 72.97500202380003, 72.97916869050003] ForwardOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Y Mapped{Float64} Float64[19.266666220799998, 19.262499554099996, …, 18.8541662175, 18.8499995508] ReverseOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG
and reference dimensions:
Ti Sampled{DateTime} DateTime[DateTime("2012-12-01T00:00:00")] ForwardOrdered Irregular Points
extent: Extent(X = (72.78125202225002, 72.98125202385003), Y = (18.84791621745, 19.26874955415))
missingval: missing
crs: EPSG:4326
mappedcrs: EPSG:4326
parent:
19.2667 19.2625 19.2583 … 18.8625 18.8583 18.8542 18.85
72.7833 1.52 0.84 0.64 0.09 0.06 0.07 0.06
72.7875 1.63 1.35 0.86 0.08 0.07 0.05 0.06
72.7917 3.3 2.28 1.26 0.09 0.08 0.07 0.11
⋮ ⋱ ⋮
72.9708 31.7 37.09 42.36 1.23 1.06 0.97 0.93
72.975 21.98 28.94 38.1 1.19 0.95 0.86 0.94
72.9792 19.52 26.22 28.3 1.35 1.01 0.87 0.8This is the image of Mumbai's radiance in 2012-12. It can be loaded from radiance_datacube by subsetting it.
using Rasters
using Dates
radiance_image = radiance_datacube[Ti=At(Date("2012-12"))]4. ncfobs_image
48×101 Raster{UInt16,2} unnamed with dimensions:
X Mapped{Float64} Float64[72.78333535560002, 72.78750202230002, …, 72.97500202380003, 72.97916869050003] ForwardOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Y Mapped{Float64} Float64[19.266666220799998, 19.262499554099996, …, 18.8541662175, 18.8499995508] ReverseOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG
and reference dimensions:
Ti Sampled{DateTime} DateTime[DateTime("2012-12-01T00:00:00")] ForwardOrdered Irregular Points
extent: Extent(X = (72.78125202225002, 72.98125202385003), Y = (18.84791621745, 19.26874955415))
crs: EPSG:4326
mappedcrs: EPSG:4326
parent:
19.2667 19.2625 … 18.8583 18.8542 18.85
72.7833 0x000d 0x000d 0x000f 0x000f 0x000e
72.7875 0x000d 0x000d 0x000e 0x000e 0x000e
72.7917 0x000d 0x000d 0x000e 0x000e 0x000e
⋮ ⋱ ⋮
72.9708 0x000c 0x000c 0x000d 0x000e 0x000d
72.975 0x000c 0x000c 0x000e 0x000e 0x000d
72.9792 0x000c 0x000c 0x000d 0x000d 0x000eThis is the image of Mumbai's number of cloud-free observations for radiance in 2012-12. It can be loaded from ncfobs_datacube by subsetting it.
using Rasters
using Dates
ncfobs_image = ncfobs_datacube[Ti=At(Date("2012-12"))]5. radiance_pixel_ts
131-element Raster{Union{Missing, Float32},1} unnamed with dimensions:
Ti Sampled{DateTime} DateTime[2012-04-01T00:00:00, …, 2023-01-01T00:00:00] ForwardOrdered Irregular Points
and reference dimensions:
X Mapped{Float64} Float64[72.78333535560002] ForwardOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Y Mapped{Float64} Float64[19.266666220799998] ReverseOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG
extent: Extent(Ti = (DateTime("2012-04-01T00:00:00"), DateTime("2023-01-01T00:00:00")),)
missingval: missing
parent:
2012-04-01T00:00:00 3.36
2012-05-01T00:00:00 2.34
2012-06-01T00:00:00 2.04
2012-07-01T00:00:00 0.0
⋮
2022-11-01T00:00:00 2.88
2022-12-01T00:00:00 4.05
2023-01-01T00:00:00 3.1This is the timeseries of radiance of pixel [1,1] extracted from radiance_datacube. It contains radiance observations between 2012-04 and 2023-01. It can be loaded using by subsetting radiance_datacube
radiance_pixel_ts = radiance_datacube[1, 1, :]6. ncfobs_pixel_ts
131-element Raster{UInt16,1} unnamed with dimensions:
Ti Sampled{DateTime} DateTime[2012-04-01T00:00:00, …, 2023-01-01T00:00:00] ForwardOrdered Irregular Points
and reference dimensions:
X Mapped{Float64} Float64[72.78333535560002] ForwardOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG,
Y Mapped{Float64} Float64[19.266666220799998] ReverseOrdered Explicit Intervals crs: EPSG mappedcrs: EPSG
extent: Extent(Ti = (DateTime("2012-04-01T00:00:00"), DateTime("2023-01-01T00:00:00")),)
parent:
2012-04-01T00:00:00 0x000b
2012-05-01T00:00:00 0x000f
2012-06-01T00:00:00 0x0004
2012-07-01T00:00:00 0x0000
⋮
2022-11-01T00:00:00 0x000c
2022-12-01T00:00:00 0x0012
2023-01-01T00:00:00 0x000fThis is the timeseries of number of cloud-free observations of the pixel [1,1] extracted from ncfobs_datacube. It contains radiance observations between 2012-04 and 2023-01. It can be loaded using by subsetting ncfobs_datacube
ncfobs_pixel_ts = ncfobs_datacube[1, 1, :]