initial results of Data collected by the ApL D2p radar ... · initial results of Data collected by the ApL D2p radar Altimeter over Land and sea ice Carl J. Leuschen and R. Keith

114 JohnshopkinsApLTechnicALDigesT,VoLume26,number2(2005)

c. J. Leuschen and r. k. rAneY

M

initialresultsofDatacollectedbytheApLD2pradarAltimeteroverLandandseaice

Carl J. Leuschen and R. Keith Raney

easuringchangesinthecryospherefromsatellite-basedaltimetershasandwillcontinuetoprovidedatathatareessentialtoourunderstandingofthelong-termtrendsoftheearth’sicecover.overthepastyears,ApLhasplayedamajorroleinairbornefieldactivities observing continental ice sheets, outlet glaciers, and sea ice using the delay-Dopplerphase-monopulse(D2p)airborneradaraltimeter,designedandbuiltbystaffatApLaspartofthenAsAinstrumentincubatorprogram.TheD2paltimeterhasbeendeployedduringfieldcampaignsinboththeArctic(2002and2003)andAntarctic(2003and2004).Duringtheseairbornedeployments,radardatawerecollectedalongwithsimul-taneousandcoincidentallaseraltimetermeasurements.Theresultingcomparisonsshowthatthetwoaltimetersrespondquitedifferentlytothevariouscryosphericgeologies.ThisarticlebrieflyreviewstheD2psystemandcomparesradarandlaseraltimeterdataselectedfromthecampaigns.

INTRODUCTIONover the last decade, the mean sea level has been

increasing at an appreciable rate (≈1.8mm/year).1 Amajorsourceofthisrise,otherthanthermalexpansion,is thought tobecontributions fromtheearth’smeltinglandicecover.measurementsofthecurrenticemassandrateofchangeof theearth’s ice inventoryareessentialdata for predicting climate trends andhelping to guidepoliciesaimedatmitigatingenvironmentalchanges.

ice sheets and glaciers—mainly those on Antarc-ticaandgreenland—constitutemorethan75%oftheearth’s total freshwater supply.2 Their ice cover holdsenough water to raise the oceans by ≈80 m.3 even a

smallriseinsealevelcouldhavesignificanteffectsonhumanhabitationincoastalareas.substantialincreasesinfreshwater frommeltingicesheetsarepredictedtoupset major ocean circulation patterns (e.g., the ther-mohalinepolar–equatorialcurrents),anddecliningseaicecoveragefacilitatesenergylossfromtheoceanintotheatmosphere.eitherofthesevariationswouldimposemajorchangesontheglobalclimate.clearly,observingand subsequently understanding trends in land or seaicecoverisofcriticallong-termimportance.

Among the several sensors used to monitor theearth’s cryosphere, airborne- and satellite-based laser

JohnshopkinsApLTechnicALDigesT,VoLume26,number2(2005) 115

D2prADArALTimeTerresuLTs

and radar altimetershaveplayedandwill continue toplay important roles. measurements made from laseraltimetershaveexposedregionsnearthesoutherncoastofgreenland that are experiencing elevationchangesexceeding –100 cm/year.4 radar altimeter measure-ments from european remote sensing satellites (ers1 and 2) showed that the elevation of Antarctica’sinterior decreased by approximately 0.9 cm/year from1992to1996.1ThegeoscienceLaserAltimetersystem(gLAs)aboardtheice,cloud,andlandelevationsatel-lite(icesat),at1064-and532-nmopticalwavelengths,is currently providing detailed surface elevation dataovertheicesheetstoenableprecisemonitoring.5Thecryosatmission(partoftheeuropeanspaceAgency’sLiving planet programme), whose payload is the ku-band (2-cm wavelength) synthetic Aperture radar(sAr)/interferometric radar Altimeter (sirAL), wasthefirst of theearthexploreropportunitymissions.6Theobjectiveofcryosat’s3-yearmissionwastomea-surechangesintheearth’sinventoryoflandandseaice.(Themissionexperiencedalaunchfailureon8october2005.Areplacementmissionisbeingconsidered.)

OVERVIEWicesheetschangeslowly,eveninresponsetomarked

increases inclimatic temperature, soanaccurate, reli-ableestimationoftherateofchangeoficesheetheightrequiresameasurement record that spansmanyyears.Therefore,themeasurementrecordmustbemuchlongerthanthelengthoftimeduringwhichanyoneobserva-tionplatformcan reasonablybe expected to continueoperation.conventionalradaraltimetershavebeencol-lectingdataformorethan25years.recentlaserssuchas gLAs on icesat eliminate surface-slope–inducederrors over continental ice sheets by providing beam-limited measurements. in addition, improved radaraltimeterssuchasthesirALoncryosatarecontinu-ing to be developed to focus on the cryosphere. bothlaserandradaraltimetersoffertheirownspecificben-efitsintermsofmeasurementaccuracy,resolution,cov-erage,andlifespan.

comparisonsbetweenthemeasurementsofthesetwoinstruments(laserandradaraltimeters)areessentialforpreservingthecontinuityofaltimetrydataoversnowandiceandalsoforprovidingtheopportunitytoextractaddi-tional dataproducts through the combinationof thosemeasurements.Tomakethesecomparisons, two funda-mentalquestionsarise:towhatextent,andunderwhatconditions, might ice sheet height measurements fromradars and lasers be expected to be comparable? ApLhasplayedamajor role inearlyexperiments toaddressthesequestions.

Aradaraltimeter’sresponseovericeisnotalwayswellunderstood,especiallywhentheremaybesnowcoverinthemeasurementfootprint.electromagneticpenetrationof dry, cold snow is significant at radar frequencies.7,8

more importantly, it is not yet fully understood whataspectsofaradarwaveform(whichrespondsprimarilytointerfacesoflargedielectriccontrast,suchasthetopoftheice)maybecomparedtoheightsmeasuredbyalaser(which responds primarily to the upper optical surfaceat the topof the snow). if snowand iceconditions arewellknown,thentherespectivelaser-andradar-reflectedwaveforms can be modeled. The inverse problem is farmoredifficult:givenaradar-reflectedwaveform,whatcanbelearnedaboutsurfaceconditions?onewaytoexplorethisissueistostudydatacollectedbyco-locatedlaserandradar altimeters that operate simultaneously. Whereasthatwouldbeimpracticalfromspace,simultaneouslaserandradaraltimetrymaybecollectedoveravarietyofsur-facesfromanairbornesystem.

THE D2P INSTRUMENT inanticipationofcryosat,whichwouldhavebeen

the first implementation of a D2p-class altimeter inspace, an airborne proof-of-concept instrument wasdesignedandbuiltbyApL.TheD2pradaraltimeterisacoherent,interferometricku-bandsystem.9inDecem-ber1998,staffatApLbegantheproject,fundedundernAsA’s instrument incubator program, to build thisinstrument.TheD2pdevelopmentproject lasted for3yearsandincludedaseriesofflighttestsoversoutherngreenlandinJune2000.

conventionalradaraltimetersareconsidered“pulse-limited” instruments and, as opposed to their “beam-limited” laser altimeter counterparts, have footprintswhose locationandsizearedeterminedbythesurfaceslopeandsystemresolution,respectively.simplyput,aconventionalradaraltimeter“sees”theclosestreflectivesurface.over land ice, the closest surface may not bedirectlybelowthealtimeter,whichimposesanerroronthe altimeter’s height measurement. sirAL and D2psharesystemdesignfeaturesthatovercomethisfunda-mentallimitation.

AsshowninFig.1a,whenthemeasurementlocationisassumedtobeatnadiroverslopedterrain,theradaraltimeterheightestimatewillbeincorrectinbothvalueandgeophysicallocation.Thiserrorissignificantevenovericesheetswhosemeanslopesareassmallas1ºorless. To mitigate slope-induced measurement errors intheacross-trackdirection,theD2p(andsirAL)designusestworeceivechannelswiththeir respectiveanten-nasseparatedintheacross-trackdirectionbyabaselineb,asshowninFig.1b.Whenthetworeceivechannelsare combined interferometrically, the phase differenceis invertible to estimate across-track surface slope.10,11in the along-track direction, slope-induced errors aremitigatedbycollectingdataatasufficientlyhighpulserepetitionfrequencytosynthesizeasetofbeam-limitedmeasurements using the delay-Doppler method,12,13 asshown in Fig. 2. The delay-Doppler method has beenshown to reduce range errors caused by along-track



surfaceslopes,improvealong-trackresolution,andpro-vide additional processing gain that in turn reducesmeasurementerror(andlowerstransmitpowerrequire-ments).since the synthesizedmeasurementsarebeamlimited,thesurfaceslopecanbeestimatedfromthepeakDoppler bin. Together, these across- and along-trackimprovements increase surface height measurementaccuracyandminimizegeophysicallocationerrorsthatsloping ice sheets induce on measurements from con-ventionalpulse-limitedradaraltimeters.

TheD2pradarsystemineffectisanairborneproto-type for the sirAL altimeter aboard cryosat.14 bothoperate at ku-band, have similar bandwidths, includetwo receivechannels, andproducecoherentdata thatmaylaterbeprocessedbythedelay-Doppleralgorithm.Table1liststhesignificantparametersandcomparestheD2pandsirALsystems.recognizingthesesimilarities,

the european space Agency (esA) co-sponsored theD2paltimetertoparticipateinseveralprelaunchcryo-satcalibrationandvalidationfieldactivities.15,16

The D2p altimeter was designed to accommodatebothlowandhighaltitudes,asnotedinthetable.The0.2-kmminimumaltitudehasproventobeveryusefulfor simultaneous operation with airborne laser altim-eters,which typicallymustbenomore than≈0.5kmabovethesurface.

FIELD ACTIVITIES

Arcticin may 2002, the D2p radar altimeter was flown

aboard the nAsA p-3 airplane along with the Air-borneTopographicmapper (ATm) laser altimeters inanexperimentco-fundedbynAsAandesAtocollectsimultaneouslaserandradaraltimeter(theLarAproj-ect)measurementsoverlandandseaice.15Theprinci-palobjectiveofthatexperimentwastoprovideinsightintothedifferencesbetweentheestimatedheightvaluesextracted from coincidental laser (optical) and radar(rF)measurementsovervarious ice and snow surfaceconditions.Therespectiveheightmeasurementsofthetwo instruments were cross-calibrated to a few centi-metersusingrunwayoverflightsatthenAsAWallopsFlight Facility. primarily because of the tight calibra-tion, the remaining differences in the measured sur-faceheightswereattributedtothespecificcryosphericcharacteristics of the scene, such as the loss factorsinfluencing optical or microwavepenetration into thesnowand ice.Figure3 shows the radar systemand itsinstallationonthenAsAp-3duringtheLarAcam-paign. The upper image shows the aircraft parked inahanger atThule,greenland.The lower left inset is

Estimatednadirlocation

(a) (b)

Actualmeasurementlocation

b

r r

yh

rα

α

∆∆

∆

Figure 1. Radar altimeter geometry in the across-track plane. (a) The measurement occurs at a location normal to the surface. When the surface is sloped and the measurement is assumed at nadir, the height estimate will be incorrect. (b) By using two receive channels, with the antennas separated by a baseline, the angle can be determined from the range (phase) difference be-tween the two channels.

(a) (b)

Measurementarea

Measurementareas

Pulsewidth

Platform velocity

r

Figure 2. Radar altimeter geometry in the along-track plane. (a) The measurement area of a conventional radar altimeter is pulse limited. (b) Through Doppler processing, a set of synthesized nar-row antenna beams is formed to provide beam-limited measure-ments, which correct the errors in height estimates over a sloped surface. Table 1. D2P and SIRAL comparison.

parameter D2p sirAL

operatingfrequency(ghz) 13.900 13.575systembandwidth(mhz) 360 320pulsewaveform LinearFm LinearFmpulselength(mm) 0.3–3.0 51.0pulserepetition frequency(khz) 1.00–1.75 1.97–17.80rangeresolution(m) 0.42 0.47peakpoweroutput(W) 5 25Antenna(deg) Along-track 4.0 1.0 Across-track 8.0 1.2Antennabaseline(m) 0.15 1.20operationalaltitudes(km) 0.2–10.0 720.0platformvelocity(km/s) 0.15 7.00



anenlargedviewoftheinstrumentbaywheretheD2pantennacanbeseenintheforwardstarboardportion.The lower right inset shows the two D2p equipmentracksinstalledsidebysideinsidetheaircraft.

ThefieldgeometryisillustratedinFig.4.Thelaserandradarcollectedbothcoincidentalandsimultaneousmeasurements.Dependingontheaircraftaltitude,theD2pfootprintrangedfrom5to20minthealongtrackand 30 to 100 m in the across track, while the coni-calscanninglaserswathsufficientlycoveredtheentireD2pfootprintwithmanylasershotsofanapproximately1-mradius.

Asafollow-ontotheLarAcam-paign, the D2p system was flownagain in 2003 under joint nAsAandesAsponsorshipaspartofthecryosat Validation experiment(cryoVex) field campaign. As in2002, simultaneous laserandradaraltimeter measurements were col-lected in the Arctic.16 in additionto data collection, the cryoVexmission was an experimental “testrun”toexerciseproceduresplannedforfuturecalibrationandvalidationactivitiestobeconductedbyesAinsupportofcryosat.cryoVexwasamajor multi-agency enterprise. In situ snowand icemeasurementsaswellashelicopter-basedelectromag-neticinductionicethicknesssound-ings were collected by researchersaboardtheAlfredWegnerinstitutePolarsternicebreakerresearchvesselfromgermany.

Figure 5 shows some images ofthesysteminstalledonthegreen-

Figure 3. D2P installation on the NASA P-3 aircraft: (clockwise from top) aircraft, D2P equipment racks, and bomb bay–mounted antenna.

Laserfootprint

D2Pfootprint

Figure 4. Field geometry. The scanning laser covers a wider swath than the D2P footprint. Each D2P footprint encompasses a number of laser measurements.

land Air Twin otter during the cryoVex campaign.Theupperleftinsetshowstheinteriorinstallation.Theextrafueltankcanbeseenontheleft,theD2pracksontherightplacedfronttoback,andthelaserrackinthebackgroundbehindthefueltank.Theupperrightinsetisabottomviewoftheaircraft.TheD2pantennaappears as the brownish square just behind the rearwheelsandbeforetherearwing.Thebottomimageistheaircraftparkedatsvalbardinnorthernnorway.TheD2p antenna can just be seen behind the ladder andbelowtheopendoor.Figure6showsthecoverageoftheLarAcampaigninredandcryoVexinblue.

Antarctic in late summer 2003, the D2p radar altimeter was

deployed again on the nAsA p-3 during the Antarc-ticAmsr-eseaice(AAsi)calibrationandvalidationfieldcampaign.TheAmsr-e(earthobservingsystemAdvanced microwave scanning radiometer) instru-mentaboardtheAquasatelliteretrievedseaiceconcen-trationdatausingpassiveradiometrictemperaturemea-surementsoveravarietyofmicrowavefrequencybands.TheroleoftheD2paltimeterduringthiscampaignwastoprovideadditionalestimatesofseaiceconcentration,freeboard, and possible snow cover from the precisionheight measurements. unfortunately, the 2003 cam-paign was canceled after aircraft difficulties occurredduring the first data flight. The AAsi campaign wascompleted in 2004 on a naval research Laboratoryp-3. Figure 7 shows the flight tracks during both the2003and2004AAsicampaigns.



EXAMPLE DATA someexampledatagatheredover

anorthernsectionofthegreenlandicesheetarepresentedinFig.8.Thehorizontalaxisrepresentsgeophysi-cal location, and the vertical axiscontains the radar profiles relativetotheWorldgeodeticsystem1984(Wgs84)referenceframe.Thepro-file waveforms are coded into therange window using a hue-satura-tion-valuecolorscheme.Theradarpowerismappedintothevalueandthecross-channelphase ismappedinto the hue. cross-channel phaseis ameasureof the incident angle,shownbythecolorbar.Theimageisfullysaturated.inareaswheretheaircraftisrolling,theimageshowsabroader range of hues since, underthose anomalous conditions, theantenna is illuminating a largerand range-varying swath on the

Figure 5. D2P installation on the Greenland Air Twin Otter aircraft: (clockwise from bot-tom) aircraft, D2P equipment racks, and bottom view of aircraft showing the antenna just aft of the rear wheels.

Figure 6. Flight tracks during the LaRA campaign (red) and Cry-oVEx (blue).

Figure 7. Flight tracks during the AASI campaign in 2003 (red) and 2004 (blue).



surface.Theradarsurfaceandotherinternalreflectionsarelabeledintheimage.

Thefeaturesinthisimagecanbeexplainedbycon-sideringthewaveformgeneratedfromalayeredmedium.Figure9illustratesasimplescenarioconsistingofthreeinterfaces (layers).here, thebeampatternof theD2pantennapointsatapproximately2.5ºstarboard,andthecolorspectrumoftheincidentanglescorrespondstothecolor bar inFig. 8.The surface response consists of a

Figure 8. Example data over the northern Greenland ice sheet. A profile of the radar waveforms is shown in color as a function of height (left axis) and geolocation. The light blue line labeled laser-derived height shows the laser measurement overlaid on the radar profile. The radar-tracking height is the blue contour just below the laser height. Internal layering is also annotated. The white plot near the top (right) shows the difference between the laser and radar measurements using the scale on the right.

Hue vs.angle

Surface

Layer 1

Layer 2

Sur

face

resp

onse

Laye

r 1

resp

onse

Laye

r 2

resp

onse

Tota

lre

spon

se

Figure 9. D2P waveform over a layered medium (not to scale). The response of a single interface shows a blue peak tapering off in amplitude and progressing in color according to the angle of in-cidence. The superposition of many interfaces (layers) is indicated by the plot on the right.

peakvalueatnadir,whichhasabluehuecorrespondingto0.0º.Theangleonthesurfaceincreaseswithreceiverdelay,causingtheamplitudeofthesurfaceresponsetodecrease and the colors to progress through the spec-trum.The response for theother two layers is similarexcept that they are lower in amplitude and delayed.Thetotalresponseisthecombinationofthethreewave-forms,wherethecolorofthelargestamplitudeataspe-cificdelaytendstodominate.intheexample,thecolorsof the total responsevarybetweenblueandmagenta,similar tothose illustratedbythewaveforms inFig.8.ThissimpleexampleshowshowthebluehuesinFig.8arethe initial responseof thesurfaceandconsecutivelayerinterfaces,whilethemagentaandotherhuesareduetooff-nadirbackscatterfromtheseinterfaces.

referring again to Fig. 8, the white line near theuppermost signal return is the laserheightderived foreachradarfootprint.Thewhite-on-blackplotnearthetopoftherangewindowdisplaysthemagnifieddiffer-encebetweenthe laserandradar trackingheights.Tomakemeaningfulcomparisonsbetweenthetwoinstru-ments,theirrespectivemeasurementsmustbemappedinto similar footprints and resolutions. The higher-resolution laser shots are mapped into the radar foot-printby takinganaverageof all the laserheightesti-mateswithintheradarfootprintweightedbytheradarantennapattern.AsshowninFig.4,eachcircularlaserfootprint is much smaller than the radar footprint;



however,theconicalscanofthelaserinstrumentcoversawiderswaththanthefootprintoftheradar.Theradarfootprintisarectanglelimitedinthealong-trackplanebythedelay-Dopplerprocessingandpulselimitedintheacross-trackplane.

penetrationintothesurfacebytheradar2-cmwave-length radiation is evident by the many near-surfaceaccumulation layers that standout in theprofile.The≈1.8-mdifferencebetweenthelaserandradarmeasure-mentsmostlikelycorrespondstotheoverlyinglayerofsnowandisinagreementwiththeannualwaterequiva-lentaccumulation(≈500mm/year)inthisareaestimatedfromicecores,firncores,andadditionalobservations.17Thismeasurementdifference implies that the “surfaceheight” detected by the radar is not actually the air/snowsurface,asdefinedbythelaser,butmorelikelytheinterface between the recently accumulated snow andtheice.Thisisatypicaloccurrenceforaradaroperat-ingoverdry snow.8 (ground truth sufficient to justifythis interpretationisnotavailableatthis location.)inaddition,suchdifferenceswillberegionallyandseason-allyvariable.understandingthesechanges isessentialiflong-termdatarecordsfromlaserandradarmeasure-mentsaretobecompared.

Figure10showsa radarwaveformcorrespondingtothe geolocation in Fig. 8 at n:77.066, e:301.466. Thegreen amplitude plot of the cross-channel power usesthelinearscaleontheright,theblackpowerplotusesthelogarithmic(db)scaleontheleft,andtheredlineisthesurfaceheightasmeasuredbythelaser.Thepeakpoweratapproximately1935.95misatalowersurfaceheightthanthelaserestimate(1937.74m).Thisdiscrep-ancycorrespondstothe1.8-mdifferenceillustratedinFig.8.Thereisaweakerpeakjustbeforethemaximumpowerthatislessinamplitudebyabout50%(or3db).Thisearlierradarresponsecoincideswellwiththelaser-measuredheight.Theamplitudedifferenceisconsistent

Figure 10. Waveform at N:77.066, E:301.466 (see Fig. 8). Heights are relative to the WGS84 reference frame. The logarithmic (dB) scale is in black, the linear scale in green, and the laser height in red. The radar measurement is offset about 1.8 m from the laser measurement.

withthereflectioncoefficientsofair/snowandsnow/iceinterfacescalculatedusingappropriatemixingformulasforsnowandice.18Thisplotreaffirmstheinterpretationthat thepositive laser–radardifference indicates snowcover.Thesemeasurementsalsosuggestthatadetailedanalysisofradarwaveformsmay(underappropriatecir-cumstances)beinvertedtoprovidesurfaceheightdatathatwouldbeconsistentwithdatafromlasers.

Figure 11 shows data from sea ice collected at lowaltitude (≈250 m) over a horizontal extent of about1 km. The white profile near the top shows a profileof the sea icederived from the laser and radarheightvalues.Thelabelsbetweentheradarresponseandtheseaiceprofiledivideregionsoftheimageintonewice,openwater,andseaice.newiceisidentifiedbyaspec-ular waveform and level tracking. For a typical quasi-roughsurface,radarenergyisbackscatteredatallangles,but the surfaceofnew icewithin a lead is so smoothandlevelthatonlynear-perfectverticalreflectionsaregenerated from the radar’s illumination.The resultingspecularresponseconsistsofthenadirreflection(bluehue), but lacks the usual off-nadir components appar-ent at later time delays observed in other waveforms.Thespecularinterpretationisalsoreinforcedbyamuchlarger signal power (not apparent in this data formatin which all waveforms have beennormalized). openwaterisidentifiedbyarougherwaveform(broaderrangeofhues)andaslightlylowerheightcomparedtothenewice.seaiceisidentifiedbyvaryingwaveformroughnessandheightvalues.Theseaicealsoappearstobehighlyfracturedbymanysmallerleads.

The ice profile near the top also shows a compari-sonbetweenthelaserheight(red)andtheradarheight(green).inregionswherethelaservalueishigher,thegap between the two values is blue, while in regionswheretheradarishigher,thehuetendstowardyellow.Theblueregionsareexplainedtobesnowcoveraswasillustratedinthepreviousexample.Anexplanationoftheyellowregionsismoreproblematic;aneedformorecomprehensive geophysical descriptions of the localsnowandiceisindicated.

Figure 12 shows two radar waveforms correspond-ingtotheleadinFig.11atn:–67.690,e:283.730.Thetopwaveformshowstheradarresponsetobe“specular,”havingverylowoff-nadirresponse,indicativeofsmoothnewice.Thelinear-scalewaveformconsistsofasinglepeakmatchingthelaserheight.Thedecibel-scalewave-formshowssubtlesidelobesassociatedwiththewindow-ingfunctionusedduringtheprocessingandalsoexposesminornonlinearitiesinthesystemresponse.Thebottomwaveformillustratestheradarresponseoveropenwater.Thisresponseshowsaslightrolloffonthetrailingedgeofthemainpeakasaresultoftheroughersurfaceofthewatercomparedtothenewice.Thesmall–30-dbpeaknear theendof thewaveforms is theresultofa slightDcbiasintheanalog-to-digitalconverter.



Figure 11. Example data over Antarctic sea ice. A profile of the radar waveforms is shown as a function of height (left axis) and geolocation. The image is separated into zones corresponding to sea ice, open water, and new ice. A profile of the ice (derived from the LaRA measurements) is displayed near the top using the scale on the right. Regions where the laser is higher than the radar indicate snow cover.

Figure 12. Waveforms at N:–67.690, E:283.730. Heights are relative to the WGS84 reference frame. The logarithmic scale is in black, the linear scale in green, and the laser height in red. The laser and radar show minimal offset. Top: Specular re-sponse over newly formed ice. Bottom: rougher response over open water.

CONCLUSIONS TheairborneApLD2pradaraltimeter,whenflown

in company with a co-located laser altimeter over avariety of ice sheets, has collected data that illustratethefrequentoccurrenceofdifferentialsbetweensurfaceheightsmeasuredbythetwotypesofinstruments.Thesedifferencesareduelargelytothepenetrationoftheradarpulsethroughthesnowcovertotheicesurface,incon-trasttothelaserwhichrespondsprimarilytothetopofthesnow.Theexamplespresentedherehaveshownthatinregionswheremicrowavepenetrationislow,suchasopenwater,orwhereseaiceoricesheetshaveminimalsnowcover,thelaser-andradar-derivedheightsagree.conversely,theresultsdifferinareaswherepenetrationis significant, such as ice sheets covered by relativelydeep,drysnow.Theseinterpretationsarefurthercom-plicatedinregionswherethelaserseemstoshowpen-etrationwithrespecttotheradar.closeexaminationofwaveformsoverdeepersnowshowsasmallpeakbeforethemainpeakintheradarwaveformthatoftencorre-spondstothesnowsurfaceheightobservedbythelaser.Thegeneralizabilityofthissignal,anditsapplicabilitytosurfaceheightmeasurementsfromspace,willbethesubjectoffutureinvestigations.

TheresultsobtainedfromtheD2pinstrumentoverthelastfewyearshaveprovidedsomeinitialinsightintothe nature of the radar waveform over varying cryo-sphericscenes,butamorerobustinterpretationofthe



THE AUTHORS

carlJ.Leuschen

r.keithraney

Carl J. LeuschenisaseniorprofessionalstaffmemberofthespaceDepartment’soceanremotesensinggroupofApL.Dr.Leuschen’sinterestsincluderadaraltimetry,radarsounding,andground-penetratingradar.heiscurrentlyaparticipatingscientistforthemArsisradarsounderaboardtheeuropeanspaceAgency’smarsexpressmissionandisalsocontributingtotheshArADsounderaboardthenAsAmarsreconnaissanceorbiter.in2002,hereceivedanAsAnewinvestiga-torAwardtoinvestigatesignalprocessingalgorithmsforicesoundingradars.hehastheprimaryroleforthedeploymentoftheApLD2pairborneradaraltimeterandhasparticipatedinnumerousfieldexperimentsatpolarregions.beforejoiningApL,Dr.LeuschenwasagraduatestudentattheuniversityofkansasradarsystemsandremotesensingLaboratory.in

1997,hereceivedanAsAgraduatestudentresearchFellowshiptodevelopaground-penetrat-ingradarformars.hegraduatedwithhighestdistinctionforhisb.s.in1995andwithhonorsforhis ph.D. in 2001, for which he received the richard and Wilma moore Thesis Award. he is amemberoftheieeeandtheTaubetapiengineeringhonorsociety.R. Keith Raney istheAs-sistantsupervisoroftheoceanremotesensinggroupofApL.currently,Dr.raneyisdevelopingadvancedradaraltimeterandradaricesoundingconceptsandisonthescienceAdvisorygroupforesA’scryosatradaraltimeterearthexplorerproject.heisaguestmemberof theLunarradar

orbiterscienceWorkinggroup,representingthemini-rFradarimager,atechnicaldemonstrationinstrumentforthemission.Whilewiththecanadacentreforremotesensing(1976–1994),Dr.raneyhelpedtoinitiatetheradarsatmission.Astheradarsatprojectscientist,hewasresponsiblefortheconceptualdesignofthesArsystem.hecontributedtothedesignofnAsA’s magellan Venus mapping radar, the esA’s ers-1 sAr, and theshuttleimagingradarsir-c.healsoservedontheeuropaorbiterradarsounderTeam(nAsA/JpL).Dr.raneyistheprincipalinventorfortheu.s.patentonthechirpscalingsArprocessingalgorithm,holdsthepatentonthedelay/Dopplerradaraltimeter,andhasapatentonanicesoundingra-dar.hewasonthefoundingboardofAssociateeditorsfortheInternational Journal of Remote Sensing and serves as anAssociateeditor (radar) for theIEEE Transactions on Geoscience and Remote Sensing.hismanyawardsincludea group Achievement Award for the magellan radar science Team, thegoldmedalofthecanadianremotesensingsociety,andthemillenniummedal 2000 from the ieee, of which he is a Fellow. For further informa-tionontopicscoveredinthisarticle,[email protected].

resultswouldrequirein-depthknowledgeoflocalsnowandicecharacteristics.Additionalcalibrationandvali-dationactivitiesarecurrentlybeingplannedtoacquirethismuch-neededknowledgeofthecoincidentalin situsnowproperties.TheD2pinstrumenthasalreadypro-videdvaluabledataandshouldplayasignificantroleinfuturefieldseasons.

reFerences 1Wingham, D. J., ridout, A. J., scharroo, r., Arthern, r. J., and

schum,c.k.,“Antarcticelevationchangefrom1992to1996,”Sci-ence282,456–458(1998).

2evans, D. L., Apel, J., Arvidson, r., bindschadler, r., carsey, r.,etal.,Spaceborne Synthetic Aperture Radar: Current Status and Future Directions, Technical memorandum 4679, nAsA (1995); http://southport.jpl.nasa.gov/nrc/.

3Drewry, D., Jordan, s. r., and Jankowski, e., “measured proper-ties of the Antarctic ice sheet: surface configuration, ice Thick-ness,Volumeandbedrockcharacteristics,”Ann. Glaciol.3, 83–91(1982).

4krabill,W.,Frederick,e.,manizade,s.,martin,c.,sonntag, J., etal.,“rapidThinningofpartsofthesoutherngreenlandicesheet,”Science283,1522–1524(1999).

5Zwally,h.J.,schutz,b.,Abdalati,W.,Abshire,J.,bentley,c.,etal.,“icesat’sLasermeasurementsofpolarice,Atmosphere,ocean,andLand,”J. Geodynamics34,405–445(2002).

6Drinkwater,m.r.,ratier,r.F.g.,andWingham,D.J.,“Theeuro-pean space Agency’s earth explorer mission cryosat: measuringVariabilityinthecryosphere,”Ann. Glaciol.39,313–320(2004).

7rignot,e.,echelmeyer,k.,andkrabill,W., “penetrationDepthofinterferometricsyntheticApertureradarsignalsinsnowandice,”Geophys. Res. Lett.28(18),3501–3504(2001).

8Laxon,s.,peacock,n.,andsmith,D.,“highinterannualVariabil-ityofseaiceThicknessintheArcticregion,”Nature425,947–950(2003).

9raney,r.k.,andJensen,J.r.,The New Generation of Radar Altim-eters: Proof of Concept. NASA Instrument Incubator Program (IIP) Final Report,sro-01-16,Jhu/ApL,Laurel,mD(2001).

10Jensen, J. r., “Design and performance Analysis of a phase-mono-pulse radar Altimeter for continental ice sheet monitoring,” inIEEE Int. Geoscience and Remote Sensing Symp. Proc.2,Florence,italy,pp.865–867(1995).

11Jensen, J. r., “Angle measurement with a phase monopulseradar Altimeter,” IEEE Trans. Antennas Propagat. 47, 715–724(1999).

12raney,r.k.,“ADelay/DopplerradarAltimeterforicesheetmoni-toring,” in IEEE Int. Geoscience and Remote Sensing Symp. Proc. 2,Florence,italy,pp.862–864(1995).

13raney, r. k., “The Delay/Doppler radar Altimeter,” IEEE Trans. Geosci. Remote Sens.36,1578–1588(1998).

14raney,r.k.,andJensen,J.r.,“AnAirbornecryosatprototype:TheD2p radar Altimeter,” in IEEE Int. Geoscience and Remote Sensing Symp. Proc.3,Toronto,canada,pp.1765–1767(2002).

15raney,r.k.,andLeuschen,c.,Technical Support for the Deployment of Radar and Laser Altimeters During LaRA 2002. Final Report,sro-03m-09,Jhu/ApL,Laurel,mD(2003).

16raney, r. k., and Leuschen, c., “simultaneous Laser and radarAltimeter measurements over Land and sea ice,” in IEEE Int. Geoscience and Remote Sensing Symp. Proc. 1, Anchorage, Ak,pp.676–678(2004).

17Dethloff,k.,schwager,m.,christensen,J.h.,kiilsholm,s.,rinke,A.,etal.,“recentgreenlandAccumulationestimatedfromregionalclimate model simulations and ice core Analysis,” J. Climate 15,2821–2832(2002).

18ulaby,F.T.,moore,r.k.,andFung,A.k.,Microwave Remote Sens-ing: Active and Passive, Vol. 3: From Theory to Applications, Artechhouse,inc.,norwood,mA(1986).

Documents

initial results of Data collected by the ApL D2p radar ... · initial results of Data collected by the ApL D2p radar Altimeter over Land and sea ice Carl J. Leuschen and R. Keith