metadata

license: mit
tags:
  - climate
  - cloud
  - physics
  - meteorology
  - satellite
pretty_name: Cloud-Temperature correlation Data
size_categories:
  - 100K<n<1M

Why was it made?

It was made by collocating data from two satellites, INSAT-3DR and CLOUDSAT, dedicated for meteorology.
INSAT-3DR is a geostationary satellite over Indian region, while CLOUDSAT is a polar satellite dedicated for precise cloud observation.
Since CLOUDSAT is a polar satellite (altitude 450 KM), it can only look at one minute region at a time. But because of this we get precise cloud data (Cloudy-Clear distinction, Cloud Height and Cloud Thickness).
Since INSAT-3DR is a geostationary satellite (36,000 KM), it scans almost half of the globe every 30 minutes, with two pixel sizes (4KM × 4KM, and 1KM × 1KM).
Idea is to combine the power of both to approximate cloud data almost anywhere on earth at any given time.

How was it made?

INSAT-3DR (L1B) data and CLOUDSAT (2B-CLDCLASS-LIDAR) data were collected for the whole year of 2017 and collocated to get this final collocated dataset.
Collocation was done by taking readings from both satellites if their pixels were close to each other in both space (<= 4KM) and time (15 minutes).

Etymology and Information

BT (Brightness Temperature) - Measured in Kelvin, it's a measure of how hot a particular object is in a particular wavelength range. The lower the BT, the cooler the emission in that specific wavelength range.
VIS - Visible Region (0.55-0.75 µm). From this reading, Albedo is derived.
SWIR - Short Wave Infrared Region (1.55-1.70 µm). From this reading, radiance is derived.
MIR - Midwave Infrared Region (3.80-4.00 µm)
TIR1 - Thermal Infrared 1 Region (10.3-11.3 µm)
TIR2 - Thermal Infrared 2 Region (11.5-12.5 µm). From all the last 3 reading, brightness temperatures are derived.
The Brightness Temperature measured by satellites is the brightness temperature from the uppermost layer of the pixel. So if a pixel has clouds, its uppermost layer will be at an altitude (say 2-3 km from ground), thus lowering its Brightness Temperature as compared to a cloudless pixel.
Albedo - Fraction of the incident reflected by an object. Always less than 1. In general, if a pixel has clouds, it'll have higher albedo because of cloud's reflective properties.
Thickness - The total thickness of all cloud layers in a particular pixel. For example, if a pixel has 3 distinct cloud layers (1 km thick each), the total cloud thickness would be 3 km. This doesn't take the cloud's optical density into account, so caution is advised.
InsatCorTir - Coordinates of INSAT pixel's TIR1/TIR2 system.
InsatCorVis - Coordinates of INSAT pixel's VIS system.
CsatCorTir - Coordinates of Cloudsat's pixel that was used to collocate with InsatCorTir. The distance between InsatCorTis and CsatCorVis is VisOffset.
CsatCorVis - Coordinates of Cloudsat's pixel that was used to collocate with InsatCorVis. The distance between InsatCorVis and CsatCorTir is TirOffset. The lesser the offsets, the more reliable the readings.
LandSeaFlag - 1- Land, 2-Ocean, 3-Coast
SolarElevation - 90 degrees when sun is overhead, 0 when sun is rising or setting.

SatElevation - 0 degree when satellite is overhead. ± 90 degrees when satellite is on the horizon.
Elevation - Height above Sea Level. -9999 indicates ocean. 9999 indicates error in height calculation.
Profile - A profile is equivalent to one pixel for CLOUDSAT. In one orbit, it makes approximately 40K profiles, accounting to a length of 1.12KM/profile along the satellite's orbit. Profile number is just the number of profile, starting from 0 when CLOUDSAT starts a new orbit.
Orbit - Total number of orbits completed by CLOUDSAT at that time.
A single slice of atmosphere of length and breadth 1 KM, can have at max 5 separate cloud layers. It is very rare to have more layers than that, although not impossible. So CLOUDSAT is designed to measure data of 5 cloud layers only.
c1type, c2type, c3type, c4type, c5type - Types of clouds in 1st, 2nd, 3rd, 4th and 5th cloud layer.

cloud base height, top height are the height of cloud base and cloud top from the ground. Thus for any given cloud, top height > base height.
All times are GST based. Time = 415 means, 04:15AM GST. Local time can be derived by using any of the coordinates.

What is the purpose of the dataset?

It can be used to evaluate cloud-data (cloudy/clear distinction, cloud top height etc.) from brightness temperatures. Brightness temperatures are readily available to us every 30 minutes.