Applications of Passive Remote Sensing Using Emission: Remote Sensing of Precipitation

Lecture 16
Applications of passive remote sensing using emission: Remote sensing of precipitation

Objectives: 1. Classification of remote sensing techniques to measure precipitation. 2. Visible and infrared remote sensing techniques to measure precipitation. 3. Sensing precipitation in the microwave region.
Required reading: G: 7.4 Additional/advanced reading: Kidder and Vonder Haar (1995): Chapter 9
1. Classification of remote sensing techniques to measure precipitation.

Importance: Precipitation is a key element of the Earths hydrological cycle.
Only a small fraction of clouds produce rain => need to separate raining from nonraining clouds
Classification of remote sensing techniques: Passive remote sensing: (i) Visible and infrared techniques (ii) Microwave techniques Active remote sensing: Radar (e.g., TRMM radar; see Lecture 21)
Other techniques: Meteorological Weather Stations (rain gauges)
2. Visible and infrared remote sensing techniques to measure precipitation.

Basic principles: Clouds are not transparent in the IR and visible (i.e., rain drops can not be sensed directly), thus one needs to relate the independent measurements of rainfall to the properties of a cloud measured by IR and visible remote sensing (called indirect measurements of precipitation)
Main problem: a lack of ground truth data to establish a correlation between satellite data and rainfall.
Techniques Cloud Indexing: developed by Barrett (1970). Principle: assign a rate rain to each cloud type
Rr = ri fi
i
where Rr is the rainfall rate, ri is the rain rate assigned to cloud type i , fi is the fraction of time (or fraction of area covered) by cloud type i.
Cloud Visible Reflection: developed by Kilonsky and Ramage (1970). Principle: tropical oceanic rainfall dominates by deep clouds which are highly reflective in the visible. Highly reflective clouds are more likely to precipitate that dark clouds because reflection is related to optical depth and hence to cloud thickness => relate the frequency of highly reflective clouds to precipitation
Parameterization by Garcia (1981) for tropical oceanic rainfall
Rr = 62.6 + 37.4ND
Rr is the monthly rainfall (in mm), Nd is the number of days during the month that the location was covered by highly reflective clouds (e.g., analysis of GOES visible channel).
OLR (outgoing longwave radiation): developed by Arkin (1979) to estimate precipitation for climatological studies. Principle: clouds that are cold in the IR are more likely to precipitate than warm clouds because cold clouds have higher tops (exception, cirrus clouds) GOES Precipitation Index (GPI) for the tropical Atlantic : GPI = 3Act where GPI is the mean rainfall (in mm), Ac is the fractional area (unitless, from 0 to 1) of cloud colder that 235 K in 2.50x2.50 box, and t is the time period (hours) for which Ac was determined.
Bispectral techniques Principle: clouds that have the high probability to produce rain are both cold (IR brightness temperature) and bright (high reflection in visible)
Cloud model techniques Principle: use cloud models to relate satellite visible and IR observations to precipitation
3. Sensing precipitation in the microwave region.

Advantages: microwave radiation penetrates clouds because cloud droplets only weakly interact with microwave radiation; rain-size drops interacts strongly with microwave radiation Disadvantages: microwave radiometers have poor spatial resolution contamination from ice crystal scattering
Main principles: Ice crystals scatter but do not absorb microwave radiation. Rain liquid drops both scatter and absorb, but absorption dominates => relate the optical depth associated with the emitting rain drops and brightness temperature measured by a passive microwave radiometer.
3
Recall the Marshall-Palmer precipitation size distribution (Lecture 8, Eq.8.5).
N(r) = N0 exp(2 r)
where N0= 8x103m-3mm-1, but, in general, N0 depends on rain type; = 4.1 Rr -0.21 mm-1, Rr is the rainfall rate (mm/hour) . Thus the volume extinction coefficient is
ke, rain = N 0 r 2Qe exp( 2 r )dr

r1
r2
Figure 16.1 Volume absorption (top) and scattering coefficients (bottom) calculated with Mie theory for the Marshall-Palmer precipitation size distribution of water and ice spheres at three frequencies 19.35, 37, 85.5 GHz).
Recall the radiative transfer equation in the microwave region (Lecture 14, Eq.[14.13])
p Tb,~ = ~ Tsur exp(( * / ) + Tatm ( ) exp( / )d / 0
*
+ R exp( / ) Tatm ( ) exp(( * ) / )d /

p ~ * 0
Lets assume that Tatm is constant in the rain layer and that the volume absorption coefficient is nearly zero except the rain layer.
*
T atm ( ) exp( / ) d / T atm [1 exp( * / )]
We can re-write the above equation for the microwave brightness temperature observed by a nadir looking microwave radiometer in the following form
p Tb,~ = ~ Tsur exp( * ) + Tatm [1 exp( * )] p + (1 ~ ) exp( * )Tatm exp( * )
where * is the optical depth associated with the emitting/absorbing rain drops:
* = k a,rain z rain
z rain is the depth of the rain layer. Re-arranging the terms in the above equation, we have
p Tb,~ = Tatm[1 + ~ (
Tsur P 1) exp( * ) (1 ~ )(exp( * ))2 ] Tatm

[16.1]
Eq.[16.1] helps to understand the brightness temperature-rain rate relationships: No rain (* = 0) =>
T b ,~ = ~p T sur T b ,~ T atm Therefore, for water surfaces,
( is small for water surfaces, and = 0.9 for dry land) Rain increases (* increases) =>
the brightness temperature strongly increases with the increases of rain rate => raining areas are easily detected over the oceans => over the dry land, the changes in brightness temperature are small with increasing rain rate: not useful for rainfall estimations
Figure 16.2 Brightness temperature versus rain rate for three frequencies.
MSPPS (Microwave Surface and Precipitation Products System) project: Uses new NOAA microwave radiometers: Advanced Microwave Sounding Unit (AMSU) AMSU-A and AMSU-B: launched on NOAA 15, 1998
AMSU-A: 15-channel cross track scanning microwave radiometer; mixed polarization AMSU-B: 5-channel cross track scanning microwave radiometer
Table 16.1 Microwave sensing of precipitation: AMSU vs. SSM/I AMSU Frequency SSM/I Frequency Microwave processes Controlled by absorption/emission by cloud water: - large drops/high water content - medium drops/moderate water content - small drops/low water content Controlled by ice-cloud scattering Retrieved product
31 GHz
9 GHz
cloud water and rainfall over oceans cloud water and rainfall over oceans non-raining clouds over oceans rainfall over the land and ocean
50 GHz
37 GHz
89 GHz 89 GHz
85 GHz 85 GHz

Applications of Passive Remote Sensing Using Emission: Remote Sensing of Precipitation

Uploaded by

Copyright:

Available Formats

Applications of Passive Remote Sensing Using Emission: Remote Sensing of Precipitation

Uploaded by

Document Information

Original Title

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

Applications of Passive Remote Sensing Using Emission: Remote Sensing of Precipitation

Uploaded by

Copyright:

Available Formats

Lecture 16

Applications of passive remote sensing using emission: Remote sensing of precipitation

1. Classification of remote sensing techniques to measure precipitation.

Other techniques: Meteorological Weather Stations (rain gauges)

2. Visible and infrared remote sensing techniques to measure precipitation.

Parameterization by Garcia (1981) for tropical oceanic rainfall

3. Sensing precipitation in the microwave region.

Recall the Marshall-Palmer precipitation size distribution (Lecture 8, Eq.8.5).

ke, rain = N 0 r 2Qe exp( 2 r )dr

+ R exp( / ) Tatm ( ) exp(( * ) / )d /

T atm ( ) exp( / ) d / T atm [1 exp( * / )]

Tsur P 1) exp( * ) (1 ~ )(exp( * ))2 ] Tatm

T b ,~ = ~p T sur T b ,~ T atm Therefore, for water surfaces,

You might also like