Using Polar2Grid software to display historical MODIS data

May 19th, 2020 |

Terra MODIS Bands 2 (0.86 µm), 6 (1.62 µm) and 26 (1.38 µm) from 28 January 2004 (Click to enlarge)

The satellites Terra (launched in 1999) and Aqua (launched in 2002) both carry the Moderate-Resolution Imaging Spectroradiometer (MODIS) instrument, an imager with 26 channels at wavelengths that range from 0.41 µm to 14.1 µm. There are simple ways to create useful imagery with this historical data with Polar2Grid software that was developed at CIMSS as part of the Community Satellite Software Package (CSPP). This blog post will show you how to create imagery as shown above (0.86 µm, 1.62 µm and 1.38 µm) over Mt. Everest on 28 January 2004.  (Similar imagery from 29 January 2004:  0.86 µm, 1.62 µm, 1.38 µm, or toggles between 28/29 January at 0.86 µm, 1.62 µm and 1.38 µm)

The self-contained Polar2Grid software package can be downloaded from this link. (You may have to register your email address before accessing the site; registration is free). Once on the website, scroll down to find “Polar2Grid V2.3 Reprojection Software for Linux” (note that the version number will occasionally increment!) and download the gzipped tarfile. You should also download the documentation (it’s a pdf file) at that site. This will tell you what to do before you can successfully run the software: for example, the POLAR2GRID_HOME variable must be set:
export POLAR2GRID_HOME=/path/to/softwarebundle.

Next, order archived MODIS data. These data are available at the NASA LAADS (Level-1 and Atmosphere Archive and Distribution System) DAAC (Distributed Active Archive Center) at this link. If you click on ‘Find Data’ at that website, a long list of possible products will be displayed. MODIS data that are compatible with Polar2Grid are Level 1b Calibrated Radiances: MOD02 files, and for this example I chose 1-km and half-kilometer resolution (that is, MOD021KM, MOD02HKM). Geolocation files (MOD03) must also be selected.

Polar2Grid includes software to create a grid onto which the data will be projected; for the example above, I first ran the Polar2Grid script ./p2g_grid_helper.sh asia 87.0 28.0 500 -500 6000 6000 > myasiagrids.txt.

This creates a grid centered at 28 N, 87 E (west longitudes are negative) with a 500-m grid spacing in both x- and y-directions; the grid has a size of 6000×6000. If you don’t create a grid, the satellite data are placed on the native satellite grid, a grid that changes from day to day for a polar orbiter.

Once the MODIS data has been placed on your local machine, you are ready to use Polar2Grid to query what products can be created using this command

./polar2grid.sh modis gtiff --list-products -f /data-hdd/AckFriendData/MODIS/MOD02_03/day028/;

the -f flag identifies the directory holding the MODIS data and modis gtiff identifies the data type and output files to be created. The result of this is a (sometimes lengthy) list of products that can be created given the input. The following command creates geotiff:

./polar2grid.sh modis gtiff -p vis02 vis06 vis26 --grid-configs /home/scottl/Polar2Grid/polar2grid_v_2_3/bin/myasiagrids.txt -g asia -f /data-hdd/AckFriendData/MODIS/MOD02_03/day028/

This command creates Bands 2, 6 and 26 GeoTiffs, and the data are placed on the ‘asia’ grid defined above (and placed in the myasiagrids.txt file). The grids created do not have georeferencing embedded within the image; that is added with the add_coastlines shell script:
./add_coastlines.sh --add-grid --add-borders --borders-resolution=f --borders-outline='red' terra_modis_vis06_20040129_031000_asia.tif

The add-grid flag inserts lat/lon lines; add-borders includes country borders (with the outline color defined, and the resolution specified; for more control flags, refer to the documentation).

Other CIMSS blog posts that discuss Polar2Grid software are here , here and here.

Eruption of the Taal Volcano in the Philippines

January 12th, 2020 |

Himawari-8

Himawari-8 “Red” Visible (0.64 µm, left) and “Clean” Infrared Window (10.4 µm, right) images [click to play animation | MP4]

The Taal Volcano erupted in the Philippines around 0850 UTC on 12 January 2020. JMA Himawari-8 “Red” Visible (0.64 µm) and “Clean” Infrared Window (10.4 µm) images (above) displayed the volcanic cloud during the initial 3 hours post-eruption. Note the presence of a pronounced “warm wake” (red enhancement) downwind (north) of the summit of Taal — this appeared to be an Above-Anvil Cirrus Plume (AACP), as seen in a toggle between the Visible and Infrared images at 1910 UTC (below).

Himawari-8 "Red" Visible (0.64 µm) and "Clean" Infrared Window (10.4 µm) images at 1910 UTC [click to enlarge]

Himawari-8 “Red” Visible (0.64 µm) and “Clean” Infrared Window (10.4 µm) images at 1910 UTC [click to enlarge]

The warmest Himawari-8 10.4 µm brightness temperatures within the Above-Anvil Cirrus Plume were around -60ºC (red enhancement), which corresponded to approximately 21 km on data from 3 rawinsonde sites in the Philippines (Legaspi, Mactan and Laoag) (below).

Plots of rawinsonde data from Legaspi, Mactan and Laoag in the Philippines [click to enlarge]

Plots of rawinsonde data from Legaspi, Mactan and Laoag in the Philippines [click to enlarge]

The TROPOMI detected SO2 at altitude of 20km on 13 January:


A longer animation of Himawari-8 Infrared imagery revealed the intermittent presence of the warm wake feature until about 1400 UTC. The coldest 10.4 µm cloud-top brightness temperature was -89.7ºC.

Himawari-8 "Clean" Infrared Window (10.4 µm) images [click to play animation | MP4]

Himawari-8 “Clean” Infrared Window (10.4 µm) images [click to play animation | MP4]

A large-scale view of Himawari-8 Infrared images (below) showed that the volcanic cloud was advected a great distance north-northeastward.

Himawari-8 "Clean" Infrared Window (10.4 µm) images [click to play animation | MP4]

Himawari-8 “Clean” Infrared Window (10.4 µm) images [click to play animation | MP4]

A toggle between NOAA-20 VIIRS Day/Night Band (0.7 µm) and Infrared Window (11.45 µm) images (below) showed the volcanic cloud at 1649 UTC.

NOAA-20 VIIRS Day/Night Band (0.7 µm) and Infrared Window (11.45 µm) images at 1648 UTC (credit: William Straka, CIMSS) [click to enlarge]

NOAA-20 VIIRS Day/Night Band (0.7 µm) and Infrared Window (11.45 µm) images at 1648 UTC (credit: William Straka, CIMSS) [click to enlarge]

In a sequence of Split Window Difference (11-12 µm) images (Terra MODIS, NOAA-20 VIIRS and Suomi NPP VIIRS) from the NOAA/CIMSS Volcanic Cloud Monitoring site (below), there was only a subtle ash signature (blue enhancement) immediately downwind of the Taal summit — due to the large amount of ice within the upper portion of the volcanic cloud, the infrared spectral ash signature was significantly masked.

Split Window Difference (11-12 um) images from Terra MODIS, NOAA-20 VIIRS and Suomi NPP VIIRS [click to enlarge]

Split Window Difference (11-12 µm) images from Terra MODIS, NOAA-20 VIIRS and Suomi NPP VIIRS [click to enlarge]

Of interest was the fact that Manila International Airport (RPLL) reported a thunderstorm at 15 UTC — there was a large amount of lightning produced by Taal’s volcanic cloud.

===== 14 January Update =====

GOES-17 SO2 RGB images [click to play animation | MP4]

GOES-17 SO2 RGB images [click to play animation | MP4]

2 days after the eruption, the leading edge of Taal’s SO2-rich volcanic plume (brighter shades of yellow over areas of cold clouds) began to appear within the far western view of GOES-17 (GOES-West) Full Disk SO2 Red-Green-Blue (RGB) images (above), about 1000 miles southeast of Japan. There were also some thin filaments of SO2 (brighter shades of white over warm ocean areas) moving southward, about 1500 miles west of Hawai’i.

Severe turbulence over coastal South Carolina

November 15th, 2019 |

GOES-16 Upper-level Water Vapor (6.2 µm) images, with plots of pilot reports and SIGMET boundaries [click to play animation | MP4]

GOES-16 Upper-level Water Vapor (6.2 µm) images, with pilot reports of turbulence and SIGMET boundaries [click to play animation | MP4]

GOES-16 (GOES-East) Upper-level Water Vapor (6.2 µm) images (above) revealed the presence of elongated W-E oriented billow clouds, many of which exhibited small-scale ripples that were oriented N-S along the billow cloud tops, over coastal areas of South Carolina and North Carolina on 15 November 2019. An initial SIGMET (November 1) was issued covering airspace over Georgia and South Carolina — Severe Turbulence (plotted in red) was reported at 41,000 feet and at 35,000 feet. A second SIGMET (November 2) was later issued covering airspace over South Carolina and North Carolina.

The same GOES-16 Water Vapor images which include isotachs of RAP40 model maximum wind (at any level) are shown below — most of the Moderate to Severe turbulence reports were occurring within the speed gradient along the poleward (left) edge of a SW-NE oriented jet stream flowing parallel to the coast.

GOES-16 Upper-level Water Vapor (6.2 µm) images, with plots of pilot reports, SIGMET boundaries, and isotachs of RAP40 model maximum wind [click to play animation | MP4]

GOES-16 Upper-level Water Vapor (6.2 µm) images, with pilot reports of turbulence, SIGMET boundaries, and isotachs of RAP40 model maximum wind [click to play animation | MP4]

More detailed views of the billow-top ripples were provided by a Terra MODIS Visible image at 1600 UTC, and NOAA-20 VIIRS True Color Red-Green-Blue (RGB) and Infrared images as visualized using RealEarth (below).

Terra MODIS Visible (0.65 µm) image, with plots of pilot reports and SIGMET boundaries [click to enlarge]

Terra MODIS Visible (0.65 µm) image, with pilot reports of turbulence and SIGMET boundaries [click to enlarge]

NOAA-20 VIIRS True Color RGB and Infrared Window (11.45 µm) images, with pilot reports of turbulence [click to enlarge]

NOAA-20 VIIRS True Color RGB and Infrared Window (11.45 µm) images, with pilot reports of turbulence [click to enlarge]

Lake-effect, river-effect and bay-effect cloud bands producing snowfall

November 13th, 2019 |

GOES-16

GOES-16 “Red” Visible (0.64 µm), “Clean” Infrared Window (10.35 µm) and Day Cloud Phase Distinction RGB images on 07 November [click to play animation | MP4]

During the course of multiple intrusions of arctic air across the Lower 48 states during early November 2019, a variety of lake-effect, river-effect and bay-effect cloud features were generated — many of which produced varying intensities of snowfall. GOES-16 (GOES-East) “Red” Visible (0.64 µm), “Clean:” Infrared Window (10.35 µm) and Day Cloud Phase Distinction Red-Green-Blue (RGB) images on 07 November (above) showed lake-effect clouds streaming south-southeastward across Lake Superior. The Day Cloud Phase Distinction RGB images (in tandem with the Infrared images) helped to highlight which cloud features had glaciated and were therefore more capable of producing moderate to heavy lake-effect snow; the dominant band yielded 5-10 inches of snowfall in the central part of northern Michigan.

On 11 November, GOES-16 Nighttime Microphysics RGB images (below) displayed lake-effect clouds originating from the still-unfrozen waters of Fort Peck Lake in northeastern Montana — these clouds did produce a brief period of light snowfall downstream at Glendive (KGDV). On this particular morning, the lowest temperature in the US occurred in north-central Montana, with -30ºF reported north of Rudyard.

GOES-16 Nighttime Cloud Phase Distinction RGB images on 11 November [click to play animation | MP4]

GOES-16 Nighttime Microphysics RGB images on 11 November [click to play animation | MP4]

On 12 November, cold air moving southward across the Lower Mississippi Valley produced horizontal convective roll clouds which were evident in GOES-16 Nighttime Microphysics RGB and subsequent Visible images after sunrise (below) — one of these narrow cloud bands was likely enhanced by latent heat fluxes as it passed over the comparatively-warm waters of the Mississippi River, and produced accumulating snowfall in downtown Memphis. Note that since Memphis International Airport KMEM was located just east of the cloud band, no accumulating snow was reported there (only a brief snow flurry around 1430 UTC).

GOES-16 Nighttime Microphysics RGB and "Red" Visible (0.64 µm) images on 12 November [click to play animation | MP4]

GOES-16 Nighttime Microphysics RGB and “Red” Visible (0.64 µm) images on 12 November [click to play animation | MP4]

Aqua MODIS Sea Surface Temperature values along parts of the Mississippi River were as warm as the mid-40s F (below).

MODIS Sea Surface Temperature product at 1848 UTC on 12 November; rivers are plotted in red [click to enlarge]

Aqua MODIS Sea Surface Temperature product at 1848 UTC on 12 November; rivers are plotted in red [click to enlarge]


On 13 November, as the cold air was moving off the US East Coast, GOES-16 Infrared images (below) revealed bay-effect cloud plumes which developed over Chesapeake Bay and Delaware Bay — the Chesapeake Bay plume produced brief periods of light snow at Oceana Naval Air Station in Virginia Beach KNTU from 06-10 UTC (and possibly contributed to snowfall farther south at Elizabeth City, North Carolina KECG).

GOES-16 "Clean" Infrared Window (10.35 µm) images on 12 November [click to play animation | MP4]

GOES-16 “Clean” Infrared Window (10.35 µm) images on 12 November [click to play animation | MP4]

Terra MODIS Sea Surface Temperature values in Chesapeake Bay and Delaware Bay were in the lower to middle 50s F where the bay-effect cloud plumes were originating (below).

Terra MODIS Sea Surface Temperature product and Visible (0.65 µm) image at 1613 UTC [click to enlarge]

Terra MODIS Sea Surface Temperature product and Visible (0.65 µm) image at 1613 UTC [click to enlarge]