fabry perot interfeometer

Please download to get full document.

View again

of 8
All materials on our website are shared by users. If you have any questions about copyright issues, please report us to resolve them. We are always happy to assist you.
Similar Documents
Information Report



Views: 53 | Pages: 8

Extension: PDF | Download: 0

IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 46, NO. 10, OCTOBER 2008 3115 Differential Radiometers Using Fabry–Perot Interferometric Technique for Remote Sensing of Greenhouse Gases Elena M. Georgieva, William S. Heaps, and Emily L. Wilson Abstract—A new type of remote-sensing radiometer based upon the Fabry–Perot (FP) interferometric technique has been devel- oped at NASA’s Goddard Space Flight Center and tested from both ground and aircraft platforms. The
  IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 46, NO. 10, OCTOBER 2008 3115 Differential Radiometers Using Fabry–PerotInterferometric Technique for RemoteSensing of Greenhouse Gases Elena M. Georgieva, William S. Heaps, and Emily L. Wilson  Abstract —A new type of remote-sensing radiometer based uponthe Fabry–Perot (FP) interferometric technique has been devel-oped at NASA’s Goddard Space Flight Center and tested fromboth ground and aircraft platforms. The sensor uses direct orreflected sunlight and has channels for measuring the columnconcentration of carbon dioxide at 1570 nm, oxygen lines sensitiveto pressure and temperature at 762 and 768 nm, and water vapor(940 nm). A solid FP etalon is used as a tunable narrow bandpassfilter to restrict the measurement to the gas of interest’s absorp-tion bands. By adjusting the temperature of the etalon, whichchanges the index of refraction of its material, the transmissionfringes can be brought into nearly exact correspondence with theabsorption lines of the particular species. With this alignmentbetween absorption lines and fringes, changes in the amount of a species in the atmosphere strongly affect the amount of lighttransmitted by the etalon and can be related to gas concentration.The technique is applicable to different chemical species. We haveperformed simulations and instrument design studies for CH 4 , 13 CO 2 isotope, and CO detection.  Index Terms —Absorbing media, atmospheric measurements,Fabry–Perot (FP) interferometers, optical interferometry, remotesensing. I. I NTRODUCTION P RECISE detection of carbon dioxide in the atmosphereis of great interest due to its impact on trapping thelong wavelength radiation emitted from the Earth’s surface.There are different scenarios about the detailed causes for theglobal warming of the Earth’s atmosphere. The net impact of anthropogenic aerosol emissions and greenhouse gases is acomplex study [1]. Recent climate models focus more on theeffects of aerosol, which can absorb solar energy and warmthe atmosphere. However, the indirect effects also occur whenaerosols change the cloud optical properties and reflect thesolar energy back into space [2]–[5]. Studies have shown thataerosoland greenhouse gas forcing arealmostofthesameorderof magnitude [1], [3]. However, whereas the anthropogenicaerosols have short lifetime, the carbon dioxide is a long-lived Manuscript received September 28, 2007; revised March 13, 2008. Currentversion published October 1, 2008. This work was supported in part by theInstrument Incubator Program, NASA Earth Sun-System Technology Officeunder Grant NRA-01-OES-01 and NASA IRAD program.E. M. Georgieva is with the Goddard Earth Science and Technology Center,University of Maryland Baltimore County, Baltimore, MD 21228 USA, andalso with the NASA Goddard Space Flight Center, Greenbelt, MD 20771 USA.W. S. Heaps is with the Instrument Systems and Technology Division,NASA/Goddard Space Flight Center, Greenbelt, MD 20771 USA.E. L. Wilson is with the Laser and Electro-Optics Branch, NASA/GoddardSpace Flight Center, Greenbelt, MD 20771 USA.Digital Object Identifier 10.1109/TGRS.2008.921570 constituent. The concentrations of the principal anthropogenicgreenhouse gases (CO 2 , CH 4 , and N 2 O) have substantiallyincreased during the industrial period. The carbon dioxideconcentration has increased by more than 95 ppmv in thelast 150 years due to the emissions of the modern industrialrevolution, and the mixing ratio has become 380 ppmv [6], [7].In this same time frame, the concentration of atmosphericmethane has increased by 150% from approximately 700 to1745 ppbv [8], [9]. Although found in lower concentrationsthan carbon dioxide, methane is about 20 times more effectivein absorbing infrared radiation. While the majority of CO 2 variability occurs in the lower atmosphere ( ∼ 1000–800 mbar),satellite instruments measure the total atmospheric column.Since sources and sinks at the surface represent a small per-turbation to the total column, a precision of better than 1% isrequired [10], [11]. That number comes from the transport in-version model experiments, which indicate that global columnmeasurements with a precision of better than 1% (3 ppmv onthe 380 ppmv background) on a time scale of one month isthe science requirement to improve the surface flux estimatesbeyond the capability of the existing network [11]. The naturalgeographic distribution and temporal variability of CO 2 sourcesand sinks are still not well understood, and more monitoringinstrumentsareneeded toquantify thecarbon dioxidedynamicsand to help predict climate change [12], [13]. The water vaporconcentration resulting from the evaporation of water from landand ocean is enhanced by carbon dioxide. As carbon dioxideabsorbs radiative heat from the Earth’s surface (warming theatmosphere), the atmospheric water vapor increases becausethe warmer air can hold more water [14]. The increased watervapor leads to an increase in the greenhouse effect and furtherincrease in temperature; the cycle is repeated until equilibrium.For climate research, high-precision ( error < 5%) water vapormeasurements are needed with global coverage [15], [16].Our knowledge of atmospheric carbon processes comes fromsatellites (Atmospheric Infrared Sounder, SCIAMACHY on-board ENVISAT satellite [17]) and ground network measure-ments. The latter consists of two main sources, i.e., the longterm in situ CO 2 measurement program led by the NationalOceanic and Atmospheric Administration (NOAA) ClimateMonitoring and Diagnostics Laboratory and the TransCom 3transport/flux estimation experiment [18], [19]. These in situ measurements are very precise (uncertainties on the order of 0.1 ppm) and accurate, but are necessarily limited in time andspace. Rayner and O’Brien [11] predicted that the requiredaccuracy for monthly averaged CO 2 column data needs to bebetter than 2.5 ppmv on 8 0 × 10 0 footprint for comparable 0196-2892/$25.00 © 2008 IEEE Authorized licensed use limited to: IEEE Xplore. Downloaded on November 5, 2008 at 12:37 from IEEE Xplore. Restrictions apply.  3116 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 46, NO. 10, OCTOBER 2008 performance to a moderate surface network. Such a precisionis difficult to achieve.The instrument that we have developed and tested on twoflight campaigns detects the absorption of CO 2 , O 2 , and H 2 Ogases using direct or reflected sunlight [20]–[24]. The sensormakes use of two features of the Fabry–Perot (FP) interferome-ter to achieve high sensitivity: high spectral resolution resultingfrom matching the width of an atmospheric absorption featureto the instrumental band pass, and high optical throughputenabled by the simultaneous use of multiple spectral lines.For any species that one wishes to measure, this first featureis available, whereas the use of multiple spectral features canonly be employed for species with suitable spectra and freedomfrom interfering species in the same wavelength region. Thethroughput of an interferometer can be as much as two ordersof magnitude larger than a spectrometer with a grating with thesame resolution, which means larger signals or better signal-to-noise ratio ( S/N  ) . Our sensor records data every 0.1 s, has highspectral selectivity, and has only one moving part (chopper),which makes it an ideal candidate for an airborne or satellitemission. It can also work as a ground-based sensor using directlight from the Sun. The current goal is to develop a preciseinexpensive ground-based device sensitive enough to measurecolumn-averaged CO 2 and suitable for wide deployment asa validation instrument for the Orbiting Carbon Observatory(OCO) satellite [25]–[28]. This will fill the gaps of the sparseground network and help to understand the global distributionof CO 2 and its role in climate change. Parallel to that, we aredeveloping a new sensor for methane detection and working onsimulations for systems to measure CO and 13 CO 2 isotope.II. D ESCRIPTION OF I NSTRUMENT AND B ASIC T HEORY The incoming light is split into three wavelength bands witha dichroic beam splitter and directed into CO 2 , O 2 pressure-sensitive, and O 2 temperature-sensitive channels. The opticalschematic for the CO 2 channel of the FP instrument is shownin Fig. 1. In this design, the incoming light is focused ontoa 2-mm-diameter aperture to define the field of view (FOV).The light is modulated at ∼ 400 Hz with a chopper and thenrecollimated as it emerges from the aperture. The incidentlight in the CO 2 channel is prefiltered at a central wavelengthof 1570 nm. The light then splits between the FP and Ref-erence subchannels with 90% of the light going to the FPsubchannel. In the FP subchannel, the light passes through theFP etalon mounted in a temperature-controlled oven for finefree spectral range (FSR) tuning. By adjusting the temperatureof the etalon, which changes the index of refraction of itsmaterial, the transmission fringes can be brought into nearlyexact correspondence with the absorption lines of the particularspecies (Fig. 2). A set of off-axis parabolic (OAP) gold mirrorsfocuses light onto InGaAs detectors for the CO 2 subchannelsand Si detectors for the two oxygen channels. For bringing lightinto the instrument, we use fiber bundles, which intentionallyscramble the incident light to remove spatial information. TheFP etalons in this instrument are solid-fused silica etalons, andthey have different thicknesses for each instrument channel.For the CO 2 channel, the etalon has an FSR of 0.306 nm, arefractive index of 1.443 at λ = 1571 nm, and a clear aperture Fig. 1. Optical and electronic schematic for the CO 2 channel of the FPinstrument. Incoming light is collimated by two OAP gold mirrors. The light isdivided into FP (detector 2) and reference (detector 1) subchannels, measuringchanges in CO 2 absorption and solar flux, respectively. The ratio of thesesignals can then be related to changes in the CO 2 column.Fig. 2. Alignment of etalon transmission bands (blue) with CO 2 absorptionlines (pink) can be adjusted by changing the etalon temperature. An etalontemperature of 54 ◦ C (top) shows better alignment than at 42 ◦ C (bottom). of 50 mm with highly reflective coatings on both surfaces. Thelight that passes through the etalon undergoes multiple reflec-tions on each inside surface, creating an interference pattern of equidistant fringes as a function of frequency. The width of the Authorized licensed use limited to: IEEE Xplore. Downloaded on November 5, 2008 at 12:37 from IEEE Xplore. Restrictions apply.  GEORGIEVA et al. : DIFFERENTIAL RADIOMETERS USING FABRY–PEROT INTERFEROMETRIC TECHNIQUE 3117 Fig. 3. FP to Reference channel ratio is plotted as a function of local timefor an O 2 instrument (yellow), a CO 2 instrument (blue), and a water vaporinstrument (red).The principal variation is that arising from the change in airmass as the sun rises and sets throughout the day. The measurement is done onJune 30, 2006, and data were collected every 0.1 s and averaged for every 120 s. pass bands depends on the coatings quality and on the flatnessand parallelism of the surfaces. The Fresnel formalism gives thereflected and transmitted amplitude components for the lightwave as functions of the optical constants of the two media andthe angle of incidence. The resulting intensity follows the Airyfunction distribution pattern. The ideal FP etalon with perfectlyflat surfaces transmits a narrow spectral band, and the energytransmission coefficient I  T  is given by [29], [30] I  T  = T  2 (1 − R ) 2  1 +4 R (1 − R ) 2 sin 2  2 πnd cos θλ  − 1 (1)where λ is the wavelength, n is the refractive index, d is thethickness of the etalon, θ is the angle of incidence withinthe cavity, T  is the intensity transmission coefficient for eachcoating, and R is the intensity reflection coefficient. The trans-mission is periodic.III. E XPERIMENTAL  A. Data Taken With the Flight Version of the FP Radiometer  Carbon dioxide and oxygen column measurements using theFP radiometer have been demonstrated in laboratory, ground-based, and airborne experiments. Water vapor channel wasrecently added and has been tested in the lab and ground-basedmeasurements.TheFPradiometeristakingdailymeasurementsat Goddard as a ground-based sensor, and the CO 2 , O 2 , andH 2 Ocolumnsaremeasuredthroughabsorptionoflightbythosespecies in the atmosphere directly between the sun and theinstrument. We accomplished this by collecting light with asmall telescope fixed to an equatorial mount, aligned to track the sun throughout the day. An optical fiber coupled at the rearof the collimator brings light into the instrument. The initialtests indicate that the instrument can detect changes in the CO 2 column as small as 2.3 ppm with a 1 s average and better Fig. 4. Here, the ratios from the previous figure are plotted against thecalculated air mass. The A . M . and P . M . values for O 2 are virtually the sameand cannot be distinguished. The CO 2 channel shows some slight variationthroughout the day, probably arising from consumption by plants. The watervapor channel shows considerable variation consistent with typical changes inhumidity throughout a day.Fig. 5. Water vapor data taken on June 29, 2006 at Goddard using FPradiometer (blue dots) compared to Annapolis water vapor data (red dots). TheFPradiometerdataandtheNOAAdatausingGPStechnologyshowremarkableagreement. than 1 ppm in less than 10 s of averaging. The precision of the oxygen column pressure changes is as small as 0.88 mbar.The water vapor precision is 2% for 1 s of averaging. Fig. 3illustrates a typical data set collected over the course of a clearday with the instrument. The ratio of FP to reference signalsis inversely proportional to absorption for CO 2 and H 2 O, anddirectly proportional for O 2 , which is plotted on an invertedscale. Fig. 4 shows the same data plotted as a function of thecalculated air mass (the path length of the sunlight through theatmosphere between the sun and the instrument). The air massdecreases all morning until local noon and then increases untilsunset. The air mass m is expressed by the Chapman functionin the Interactive Data Language program for data analysis, so Authorized licensed use limited to: IEEE Xplore. Downloaded on November 5, 2008 at 12:37 from IEEE Xplore. Restrictions apply.  3118 IEEE TRANSACTIONS ON GEOSCIENCE AND REMOTE SENSING, VOL. 46, NO. 10, OCTOBER 2008 Fig. 6. Instrument response to altitude in meters at airborne mission. (a) CO 2 ratio of FP to Reference signals is shown compared to changes in altitude ona May 14, 2004 flight. As the aircraft descends, the photon path length getsshorter so the absorption in the FP subchannel decreases. (b) Ratio for thepressure-sensing channel ( O 2 ) and air mass as a function of time is shown. Theembedded graph shows the blown images of the measured Ratio and altitudefor 4 min. The standard deviation is calculated to be 0.00393. we can account for the sphericity of the Earth as m i =1 τ  i ∞   z 0 σ i ( z ) dz  1 −  n 0 z 0 nz sin θ 0  2  1 / 2 (2)where m is the air mass, τ  is the optical depth, σ is theextinction coefficient as a function of altitude z , θ is the zenithangle of the sun, and n is the atmospheric index of refraction.For very small solar zenith angles, the air mass is equal tothe secant of the zenith angle, and that is the plane-parallelapproximation.The O 2 data essentially lie on top of one another because theatmospheric pressure (and hence the O 2 column) was constanton this day. The CO 2 channel shown in blue has some slightvariability. CO 2 is expected to very slightly change on a diurnalbasis because of the consumption of CO 2 by plants during theday. CO 2 also has a seasonal variability with maxima in springand minima in the late summer [31]. Those variations reflectchanges in the biological sinks and can range from 1 ppmv inthe Southern Hemisphere to more than 20 ppmv in the NorthernHemisphere.The variation in the water vapor column is shown to be realin a comparison with the total column measurements by NOAAusing Global Positioning System (GPS) technology. This resultverifies that instruments of this type properly function and Fig. 7. Representation of the physical design of the small-size CO 2 instrument.Fig. 8. Laser scan of the two filters for the small CO 2 radiometer. The CO 2 absorption lines from the HITRAN database are also shown. The narrow bandversus wide band subchannel ratio will be sensitive to CO 2 changes in theatmosphere. can precisely be calibrated. Fig. 5 shows this intercomparisonof H 2 O measurement by the FP instrument system and by aNOAA GPS-based sensor located in Annapolis, MD, for fivedays in late June and early July. The FP radiometer only oper-ates during the day. Agreement is consistent with the variabilityexpected for sites ∼ 30 miles apart.For airborne or satellite measurements, the light passingthrough the atmosphere reflects on the Earth’s surface be-fore entering the instrument platform. The flight-hardened ver-sion of the instrument was tested at two flight campaigns atthe National Aeronautics and Space Administration (NASA)Dryden Research Center and at New Hampshire for the PolarAura Validation campaign on NASA’s DC-8 research airplane.Flights were conducted over a variety of surfaces (vegetation,water, and snow) and under different atmospheric conditions.An in situ instrument provided CO 2 profiles up to the flightaltitude. Temperature and density profiles were available from Authorized licensed use limited to: IEEE Xplore. Downloaded on November 5, 2008 at 12:37 from IEEE Xplore. Restrictions apply.
We Need Your Support
Thank you for visiting our website and your interest in our free products and services. We are nonprofit website to share and download documents. To the running of this website, we need your help to support us.

Thanks to everyone for your continued support.

No, Thanks

We need your sign to support Project to invent "SMART AND CONTROLLABLE REFLECTIVE BALLOONS" to cover the Sun and Save Our Earth.

More details...

Sign Now!

We are very appreciated for your Prompt Action!