atmospheric soundings, surface properties, clouds the bologna lectures paul menzel noaa/nesdis/ora
TRANSCRIPT
Atmospheric Soundings, Surface Properties, Clouds
The Bologna Lectures
Paul Menzel
NOAA/NESDIS/ORA
Relevant Material in Applications of Meteorological Satellites
CHAPTER 6 - DETECTING CLOUDS6.1 RTE in Cloudy Conditions 6-16.2 Inferring Clear Sky Radiances in Cloudy Conditions 6-26.3 Finding Clouds 6-3
CHAPTER 7 - SURFACE TEMPERATURE 7.2. Water Vapor Correction for SST Determinations 7-37.3 Accounting for Surface Emissivity in the Determination of SST 7-67.4 Estimating Fire Size and Temperature 7-6
CHAPTER 8 - TECHNIQUES FOR DETERMINING ATMOSPHERIC PARAMETERS 8.1 Total Water Vapor Estimation 8-18.3 Cloud Height and Effective Emissivity Determination 8-88.6 Satellite Measurement of Atmospheric Stability 8-13
Earth emitted spectra overlaid on Planck function envelopes
CO2
H20
O3
CO2
Profile Retrieval from Sounder Radiances
ps
I = sfc B(T(ps)) (ps) - B(T(p)) [ d(p) / dp ] dp .
o
I1, I2, I3, .... , In are measured with the sounderP(sfc) and T(sfc) come from ground based conventional observations(p) are calculated with physics models
First guess solution is inferred from (1) in situ radiosonde reports, (2) model prediction, or (3) blending of (1) and (2)
Profile retrieval from perturbing guess to match measured sounder radiances
Sounder Retrieval Products
Directbrightness temperatures
Derived in Clear Sky20 retrieved temperatures (at mandatory levels)20 geo-potential heights (at mandatory levels)11 dewpoint temperatures (at 300 hPa and below)3 thermal gradient winds (at 700, 500, 400 hPa)1 total precipitable water vapor1 surface skin temperature2 stability index (lifted index, CAPE)
Derived in Cloudy conditions3 cloud parameters (amount, cloud top pressure, and cloud top temperature)
Mandatory Levels (in hPa)sfc 780 300 701000 700 250 50950 670 200 30920 500 150 20850 400 100 10
Remote Sensing Regions
Windows to the atmosphere (regions of minimal atmospheric absorption) exist near 4 m and 10 m; these are used for sensing the temperature of the earth surface and clouds.
CO2 absorption bands at 4.3 m and 15 m are used for temperature profile retrieval; because these gases are uniformly mixed in the atmosphere in known portions they lend themselves to this application.
The water vapor absorption band near 6.3 m is sensitive to the water vapor concentration in the atmosphere; as H20 is not uniformly mixed in the atmosphere, measurements in this spectral region are used to infer moisture distribution in the atmosphere.
The ozone absorption band at 9.7 m reveals locations of O3 concentration in the upper atmosphere.
GOES Sounder Spectral Bands: 14.7 to 3.7 um and vis
GOES Imager -- Spectral Coverage
Comparison of GOES-8 PW with microwave retrievals
GOES Sounder total precipitable water (PW) values compare well with co-located microwave radiometer measurements at the CART site (Lamont, OK). “Flat” first-guess trace is adjusted by sounder radiances to capture the trend
and range of total moisture.
Comparison of GOES-8 PW with microwave retrievals
A scatter plot comparing MWR integrated water vapor values and GOES-8 first guess/physical retrieval values at the CART site . RMS and bias values for all matches are quantified in the lower right hand corner.
VALUE ADDEDVALUE ADDED THERMODYNAMIC DIAGRAMS SHOWING THERMODYNAMIC DIAGRAMS SHOWINGGOES VERTICAL TEMPERATURE AND MOISTUREGOES VERTICAL TEMPERATURE AND MOISTUREPROFILES DERIVED FROM PROFILES DERIVED FROM GOES SOUNDERGOES SOUNDER
Generated atGenerated athundredshundreds of ofsites across USsites across USevery hourevery hour
Used extensivelyUsed extensivelyby NWS fieldby NWS fieldforecasters forforecasters forsevere weathersevere weatherforecastingforecasting
Product can beProduct can belooped over a 24looped over a 24hour period tohour period toshow show trendstrends of ofatmosphericatmosphericstabilitystabilityhttp://orbit-net.nesdis.noaa.gov/goes/soundings/skewt/html/skewtus.html
UNSTABLEUNSTABLE
STABLESTABLE
Interactive Viewing of GOES Sounder DPI Time SeriesUW/Madison/CIMSS NOAA/NESDIS/ORA/ARAD/ASPT
A new interactive web site (http://cimss.ssec.wisc.edu/goes/realtime/gdpiviewer.html) allows users to view time series of GOES derived product imagery (Lifted Index (LI), Precipitable Water (PW), and Convective Available Potential Entergy (CAPE)) by clicking on a desired location within the latest Derived Product Image (DPI).
(graph lines: thick when sounding available, thin when cloud obscured)
(example from Dec. 2, 1999)(de-stabilizing with time)
Detecting Clouds (IR)
IR Window Brightness Temperature Threshold and Difference Tests
IR tests sensitive to sfc emissivity and atm PW, dust, and aerosolsBT11 < 270BT11 + aPW * (BT11 - BT12) < SSTBT11 + bPW * (BT11 - BT8.6) < SST aPW and bPW determined from lookup table as a function of PW BT3.9 - BT11 > 12 indicates daytime low cloud coverBT11 - BT12 < 2 (rel for scene temp) indicates high cloudBT11 - BT6.7 large neg diff for clr sky over Antarctic Plateau winter
CO2 Channel Test for High Clouds
BT13.9 < threshold (problems at high scan angle or high terrain)
Detecting Clouds (vis)
Reflectance Threshold Test
r.87 > 5.5% over ocean indicates cloudr.66 > 18% over vegetated land indicates cloud
Near IR Thin Cirrus Test
r1.38 > threshold indicates presence of thin cirrus cloudambiguity of high thin versus low thick cloud (resolved with BT13.9)problems in high terrain
Reflectance Ratio Test
r.87/r.66 between 0.9 and 1.1 for cloudy regionsmust be ecosystem specific
Snow Test
NDSI = [r.55-r1.6]/ [r.55+r1.6] > 0.4 and r.87 > 0.1 then snow
aa
MODIS cloud mask exampleMODIS cloud mask example(confident clear is green, probably clear is blue, uncertain is red, cloud is white)
1.6 µm image 0.86 µm image 11 µm image 3.9 µm image cloud mask
Snow test(impacts choice of tests/thresholds)
VIS test(over non-snowcovered areas)
3.9 - 11 BT test
for low clouds
11 - 12 BT test (primarily for
high cloud) 13.9 µm
high cloud test(sensitive in cold regions)
AVIRIS Movie #2
AVIRIS Image - Porto Nacional, Brazil20-Aug-1995
224 Spectral Bands: 0.4 - 2.5 mPixel: 20m x 20m Scene: 10km x 10km
MODIS identifies
cloudclasses
Clouds separate into classes when multispectral radiance information is viewed
Multispectral data reveals improved information about ice / water clouds
Cloud Composition
Ice Cloud
Water Cloud
Contrails
Contrails
Infrared Temperature Difference - 8.6 m (Band 29) - 11.0 m (Band 31)
Infrared Temperature Difference - 11.0 m (Band 31) - 12.0 m (Band 32)
Image Over Kansas - 21 April 1996
Tri-spectral IR thermodynamic phase algorithm
• 8.6-11 vs 11-12
• when slope > 1 then ice
• when slope < 1 then water
ice cloudApril 1996Success
water cloudJan 1993TOGA/COARE
Strabala, Menzel, and Ackerman, 1994, JAM, 2, 212-229.Baum et al, 2000, JGR, 105, 11781-11792.
Water phase clouds with 238K < Tc < 253K Water phase clouds with 238K < Tc < 253K
RTE in Cloudy Conditions
Iλ = η Icd + (1 - η) Ic where cd = cloud, c = clear, η = cloud fraction λ λ
oIc = Bλ(Ts) λ(ps) + Bλ(T(p)) dλ . λ ps
pc
Icd = (1-ελ) Bλ(Ts) λ(ps) + (1-ελ) Bλ(T(p)) dλ
λ ps
o + ελ Bλ(T(pc)) λ(pc) + Bλ(T(p)) dλ
pc
ελ is emittance of cloud. First two terms are from below cloud, third term is cloud contribution, and fourth term is from above cloud. After rearranging
pc dBλ
Iλ - Iλc = ηελ (p) dp .
ps dp
Techniques for dealing with clouds fall into three categories: (a) searching for cloudless fields of view, (b) specifying cloud top pressure and sounding down to cloud level as in the cloudless case, and (c) employing adjacent fields of view to determine clear sky signal from partly cloudy observations.
Cloud Clearing
For a single layer of clouds, radiances in one spectral band vary linearly with those of another as cloud amount varies from one field of view (fov) to another
Clear radiances can be inferred by extrapolating to cloud free conditions.
RCO2
RIRW
cloudy
clear
x x
x xx
xxx
xpartly cloudy
N=1 N=0
Paired field of view proceeds as follows. For a given wavelength λ, radiances from two spatially independent, but geographically close, fields of view are written
Iλ,1 = η1 Iλ,1cd + (1 - η1) Iλ,1
c ,
Iλ,2 = η2 Iλ,2 cd + (1 - η2) Iλ,2
c ,
If clouds are at uniform altitude, and clear air radiance is in each FOV
Iλcd = Iλ,1
cd = Iλ,2 cd
Iλc = Iλ,1
c = Iλ,2c
cd c c
η1 (Iλ - Iλ ) η1 Iλ,1 - Iλ = = η* = ,
cd c c
η2 (Iλ - Iλ) η2 Iλ,2 - Iλ
where η* is the ratio of the cloud amounts for the two geographically independent fields of view of the sounding radiometer. Therefore, the clear air radiance from an area possessing broken clouds at a uniform altitude is given by
c Iλ = [ Iλ,1 - η* Iλ,2] /[1 - η*]
where η* still needs to be determined. Given an independent measurement of surface temperature, Ts, and measurements Iw,1 and Iw,2 in a spectral window channel, then η* can be determined by
η* = [Iw,1 - Bw(Ts)] / [Iw,2 - Bw(Ts)]
and Iλc for different spectral channels can be solved.
Cloud Properties
RTE for cloudy conditions indicates dependence of cloud forcing (observed minus clear sky radiance) on cloud amount () and cloud top pressure (pc)
pc
(I - Iclr) = dB .
ps
Higher colder cloud or greater cloud amount produces greater cloud forcing; dense low cloud can be confused for high thin cloud. Two unknowns require two equations.
pc can be inferred from radiance measurements in two spectral bands where cloud emissivity is the same. is derived from the infrared window, once pc is known. This is the essence of the CO2 slicing technique.
Moisture
Moisture attenuation in atmospheric windows varies linearly with optical depth. - k u = e = 1 - k u
For same atmosphere, deviation of brightness temperature from surface temperature is a linear function of absorbing power. Thus moisture corrected SST can inferred by using split window measurements and extrapolating to zero k
Ts = Tbw1 + [ kw1 / (kw2- kw1) ] [Tbw1 - Tbw2] .
Moisture content of atmosphere inferred from slope of linear relation.
Early SST algorithms
* IRW histogram of occurrence f of observed brightness temperatures T
f(T) = fs exp [ -(T - Tsfc)2/22 ]
instrument noise / scene variability produce Gaussian distribution; warm
side of histogram reveals Ts = T(d2f/dT2=0) -
* Three point methodcombinations of (Ti, fi), (Tj, fj), and (Tk, fk) on the warm side of the histogram enable 3 equations / 3 unknowns, hence a histogram of Ts solutions.
* Least squares method
ln (f(T)) = ln (fs) - Ts2/22 + TsT/2 - T2/22
has the form
ln (f(T)) = Ao + A1T + A2T2 ,
so
Ts = - A1/(2A2) .
Histograms of infrared window brightness temperature in cloud free and cloud contaminated conditions
Water Vapor Correction for SST Determinations
* Water vapor correction (T)
Ts = Tb + T
ranges from 0.1 C in cold/dry to 10 K in warm/moist atmospheres for 11 um
IRW observations.
* Water vapor correction is highly dependent on wavelength.
* Water vapor correction depends on viewing angle.
* In IRW for small water vapor concentrations,w = e-Kwu ~ 1 - Kwu
so that Ts = Tbw1 + [ Kw1 / (Kw2- Kw1) ] [Tbw1 - Tbw2] .
linear extrapolation to moisture free atmosphere
* Regression of clear sky IRW obs and collocated buoys create current
operational algorithm
Ts=A0+A1* Tbw1 +A2*( Tbw1-Tbw2)+A3*(Tbw1-Tbw2)2
a quadratic term helps account for occasional large water vapor concentrations.
Cloud Detection
Several multispectral methods have evolved to detect clouds in the area of interest.
Input data are vis, T3.9, T11, and T12 , T11@30min, and SST guess.
General tests include:
T11 > 270 K ocean rarely frozenT11 > T12 + 4 K clouds affect moisture correctionvis < 4% clouds reflect more than ocean sfcT11 - T3.9 > 1.5 K subpixel clouds T11 < 0.3 K SST over 1 hr small-2 K < SST- guess < 5 K SST over days bounded
Advantages of Geostationary SST Estimates
* 10 times more observations of a given location
* multispectral cloud detection supplemented by temporal persistence checks
* clear sky viewing enhanced by persistence (e.g. can wait for clouds to move through)
* daily composite provides good spatial coverage
* can discern diurnal excursions in SST
* can track SST motions as estimates of ocean currents
GOES daily composite SST reveals small scale features in oceans
GOES detects diurnal SST excursions of 2-3 C in calm waters
Accounting for surface emissivity
When the sea surface emissivity is less than one, there are two effects that must be considered: (a) the atmospheric radiation reflects from the surface, and (b) the surface emission is reduced from that of a blackbody. The radiative transfer can be written
ps
I = B(Ts) (ps) + B(T(p))d(p) o
ps
+ (1-) (ps) B(T(p))d‘(p) o
where ‘(p) represents the transmittance down from the atmosphere to the surface. This can be rewritten
I = B(ps)(ps) + B(TA)[1 - (ps) - (ps)2 + (ps)2] .
Note that as the atmospheric transmittance approaches unity, the atmospheric contribution expressed by the second term becomes zero.
HIS and GOES radiance observations plotted in accordance with the radiative transfer equation including corrections for atmospheric moisture, non-unit emissivity of the sea surface, and reflection of the atmospheric
radiance from the sea surface. Radiances are referenced to 880 cm-1. The intercept of the linear relationship for each data set represents a retrieved surface skin blackbody radiance from which the SST can be retrieved.
Comparison of ocean brightness temperatures measured by a ship borne interferometer (AERI), by an interferometer (HIS) on an aircraft at 20 km altitude, and the geostationary sounder (GOES-8). Corrections for
atmospheric absorption of moisture, non-unit emissivity of the sea surface, and reflection of the atmospheric radiance from the sea surface have not been made.
GOES 3 by 3 FOVs (30 km)
MODIS 5 by 5 FOVs (5 km)
11 micron
30km resolution GOES
500hPa T
5km resolution MODIS
MODIS
GOES vs. MODIS 2000/06/30 1600 UTCTotal Precipitable Water (mm)
GOES 30 km resolution
MODIS 5 km resolution
TPW
TPW
GOES vs. MODIS 2000/06/30 1600 UTCTotal Precipitable Water (mm)
MODIS total precipitable water vapor shows a wet bias wrt GOES;bias 1.5 mm and rms of 3 mm; bias will be removed after more validation
MODIS 2000/09/05-08 Daytime Total Precipitable Water (cm)
values over land not shown to facilitate comparison with AMSU
NOAA-15 AMSU-A 2000/09/05Daytime Total Precipitable Water (mm)
MODIS TPW: Upper panel; AMSU TPW: Lower panel
GOES 30 km resolution
MODIS 5 km resolution
Ozone
Ozone
MODIS ozone is very close to the GOES ozone (over North America); rms of about 10 Dobsons; polar extreme ozone values will be improved
MODIS Ozone
TOMS Ozone
Early Estimates of UW MODIS Product Quality
MODIS IR window radiances agree to within 1 C with GOES and ER-2 MAS/SHIS
Cloud mask has demonstrated advantages of new multispectral approachsun glint, desert, and polar problems diminished
MODIS cloud top pressures compare well with HIRS;aircraft validation is better than 50 mb.
MODIS cloud phase determinations are revealing interesting patterns;first ever global day/night ice/water cloud determinations; validations pending.
MODIS tropospheric temperatures compare well with AMSU;rms better than 1 C, both within 2 C of radiosonde observations
MODIS total precipitable water vapor shows a wet bias wrt GOES;bias 1.5 mm and rms of 3 mm; bias will be removed after more validation.
MODIS ozone is very close to the GOES ozone (over North America); rms of about 10 Dobsons; polar extreme ozone values will be improved