In MazamaScience/MazamaSatelliteUtils: Utilities for Working with Model and Satellite Data as Rasters

library(MazamaSatelliteUtils)
library(MazamaSpatialUtils)

setSatelliteDataDir("~/Data/Satellite")
setSpatialDataDir("~/Data/Spatial")

loadSpatialData("NaturalEarthAdm1")

knitr::opts_chunk$set(echo = TRUE)
cellSize <- 0.25

The GOES-16 and GOES-17 satellites together cover the east and west regions of the continental US with their ABI scanners, but what exactly do their coverage areas look like? Do they leave any regions out? This article will try to answer these questions by plotting scans from each satellite at different times and data quality levels.

GOES-16 Coverage

The GOES-16 satellite operates at the GOES East position of 75.2°W. Let's plot one of its recent scans from 2021 with DQF levels of 3, 2, 1, 0 (highest to lowest quality).

goesaodc_downloadScanFiles(
  satID = "G16",
  datetime = "2021-05-12 09:00",
  timezone = "America/Chicago",
  verbose = TRUE
)

scanFileG16 <- goesaodc_listScanFiles(
  satID = "G16",
  datetime = "2021-05-12 09:00",
  timezone = "America/Chicago"
)

# Create G16 scan rasters

scanRasterG162021dqf3 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 3
)

scanRasterG162021dqf2 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 2
)

scanRasterG162021dqf1 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 1
)

scanRasterG162021dqf0 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 0
)

# Plot G16 scan rasters

goesaodc_plotScanRaster(
  raster = scanRasterG162021dqf3,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF3 Coverage (2021-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG162021dqf2,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF2 Coverage (2021-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG162021dqf1,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF1 Coverage (2021-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG162021dqf0,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF0 Coverage (2021-05-12)"
)

It looks like GOES-16's coverage depends on the given DQF level. When a DQF of 3 or 2 is requested, then the entire CONUS is covered. If the requested DQF is lower than that, then most of the western states are cut out either partially or completely: Washington, Oregon, and Idaho have no coverage, and California, Nevada, Montana, Utah, Wyoming, and the Dakotas are only semi-covered.

Is this true across all of GOES-16's scans though? Since MazamaSatelliteUtils stores GOES scans going back to March 2019, let's take a look at a GOES-16 scan from 2 years prior to the one we just plotted.

goesaodc_downloadScanFiles(
  satID = "G16",
  datetime = "2019-05-12 09:00",
  timezone = "America/Chicago",
  verbose = TRUE
)

scanFileG16 <- goesaodc_listScanFiles(
  satID = "G16",
  datetime = "2019-05-12 09:00",
  timezone = "America/Chicago"
)

# Create G16 scan rasters

scanRasterG162019dqf3 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 3
)

scanRasterG162019dqf2 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 2
)

scanRasterG162019dqf1 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 1
)

scanRasterG162019dqf0 <- goesaodc_createScanRaster(
  filename = scanFileG16,
  cellSize = cellSize,
  dqfLevel = 0
)

# Plot G16 scan rasters

goesaodc_plotScanRaster(
  raster = scanRasterG162019dqf3,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF3 Coverage (2019-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG162019dqf2,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF2 Coverage (2019-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG162019dqf1,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF1 Coverage (2019-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG162019dqf0,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-16 DQF0 Coverage (2019-05-12)"
)

Yup, looks like GOES-16's coverage has stayed consistent -- for these two instances at least.

GOES-17 Coverage

The GOES-17 satellite operates at the GOES West position of 137.2°W. Again, let's plot one of its recent scans from 2021 with DQF levels of 3, 2, 1, 0 (highest to lowest quality).

goesaodc_downloadScanFiles(
  satID = "G17",
  datetime = "2021-05-12 09:00",
  timezone = "America/Chicago",
  verbose = TRUE
)

scanFileG17 <- goesaodc_listScanFiles(
  satID = "G17",
  datetime = "2021-05-12 09:00",
  timezone = "America/Chicago"
)

# Create G17 scan rasters

scanRasterG172021dqf3 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 3
)

scanRasterG172021dqf2 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 2
)

scanRasterG172021dqf1 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 1
)

scanRasterG172021dqf0 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 0
)

# Plot G17 scan rasters

goesaodc_plotScanRaster(
  raster = scanRasterG172021dqf3,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF3 Coverage (2021-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG172021dqf2,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF2 Coverage (2021-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG172021dqf1,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF1 Coverage (2021-05-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG172021dqf0,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF0 Coverage (2021-05-12)"
)

GOES-17 only covers the western portion of the CONUS even when the requested DQF level is 3. A DQF set to 1 or lower restricts the satellite's range so that portions of Montana and Wyoming are cut off and the Dakotas are removed entirely.

MazamaSatelliteUtils only has GOES-17 scans going back to September 2019, so we can't look quite as far back to check for consistent coverage.

goesaodc_downloadScanFiles(
  satID = "G17",
  datetime = "2019-09-12 09:00",
  timezone = "America/Chicago",
  verbose = TRUE
)

scanFileG17 <- goesaodc_listScanFiles(
  satID = "G17",
  datetime = "2019-09-12 09:00",
  timezone = "America/Chicago"
)

# Create G17 scan rasters

scanRasterG172019dqf3 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 3
)

scanRasterG172019dqf2 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 2
)

scanRasterG172019dqf1 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 1
)

scanRasterG172019dqf0 <- goesaodc_createScanRaster(
  filename = scanFileG17,
  cellSize = cellSize,
  dqfLevel = 0
)

# Plot G17 scan rasters

goesaodc_plotScanRaster(
  raster = scanRasterG172019dqf3,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF3 Coverage (2019-09-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG172019dqf2,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF2 Coverage (2019-09-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG172019dqf1,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF1 Coverage (2019-09-12)"
)

goesaodc_plotScanRaster(
  raster = scanRasterG172019dqf0,
  naColor = "transparent",
  legendLimits = c(-0.5, 5.5),
  stateCodes = datasets::state.abb,
  title = "GOES-17 DQF0 Coverage (2019-09-12)"
)

GOES-17's coverage was even more restricted back in 2019. For DQF 3 and 2, the only difference is that some of the Pacific Ocean was cut off. However, for DQFs lower than 2, every state on the west coast is completely removed and only some of Arizona, Utah, Idaho, and Montana were in range.

Combined GOES Coverage

Let's overlay the GOES satellites' ranges on top of each other to see if there are any regions that aren't covered at all. Since the views appear to only shrink when DQF < 2, we only need to take a look at DQF 3 and 1.

# 2021 DQF3 combined coverage
AirFirePlots::plot_base(
  title = "GOES-16/17 2021 DQF3"
) +
  AirFirePlots::layer_raster(
    raster = scanRasterG162021dqf3,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_raster(
    raster = scanRasterG172021dqf3,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_states(
    stateCodes = datasets::state.abb
  ) +
  ggplot2::scale_fill_gradient(
    low = rgb(0, 0, 1),
    high = rgb(0, 0, 1),
    guide = FALSE
  )

# 2021 DQF1 combined coverage
AirFirePlots::plot_base(
  title = "GOES-16/17 2021 DQF1"
) +
  AirFirePlots::layer_raster(
    raster = scanRasterG162021dqf1,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_raster(
    raster = scanRasterG172021dqf1,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_states(
    stateCodes = datasets::state.abb
  ) +
  ggplot2::scale_fill_gradient(
    low = rgb(0, 0, 1),
    high = rgb(0, 0, 1),
    guide = FALSE
  )

In 2021, the combined range of the GOES satellites covers the entire CONUS for DQF 3. When DQF is lower than 2, a gap in their coverage appears over eastern Montana and western North Dakota.

# 2019 DQF3 combined coverage
AirFirePlots::plot_base(
  title = "GOES-16/17 2019 DQF3"
) +
  AirFirePlots::layer_raster(
    raster = scanRasterG162019dqf3,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_raster(
    raster = scanRasterG172019dqf3,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_states(
    stateCodes = datasets::state.abb
  ) +
  ggplot2::scale_fill_gradient(
    low = rgb(0, 0, 1),
    high = rgb(0, 0, 1),
    guide = FALSE
  )

# 2019 DQF1 combined coverage
AirFirePlots::plot_base(
  title = "GOES-16/17 2019 DQF1"
) +
  AirFirePlots::layer_raster(
    raster = scanRasterG162019dqf1,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_raster(
    raster = scanRasterG172019dqf1,
    varName = "AOD",
    naRemove = TRUE,
    alpha = 0.3
  ) +
  AirFirePlots::layer_states(
    stateCodes = datasets::state.abb
  ) +
  ggplot2::scale_fill_gradient(
    low = rgb(0, 0, 1),
    high = rgb(0, 0, 1),
    guide = FALSE
  )

Like 2021, in 2019 the combined range of the GOES satellites covered the entire CONUS for DQF 3. When DQF is lower than 2, however, there is much less coverage. Washington and Oregon are cut off completely, and only portions are available for: California, Nevada, Idaho, Montana, Wyoming, Utah, and North Dakota.

Conclusion

It appears that over the course of their operation, the combined ranges of the GOES satellites cover the entire CONUS when DQF is 2 or 3. However, when the requested DQF drops below these values, no AOD readings will be available for eastern Montana or western North Dakota. Earlier scans will also be missing data for many of the western states.

MazamaScience/MazamaSatelliteUtils documentation built on Dec. 17, 2021, 3:20 a.m.