National Aeronautics and Space Administrationwww.nasa.gov

NASA: Supporting Earth System Science 2005

NASA: Supporting Earth System Science 2005Applications of Earth System Science Data from the NASA Science Mission Directorate

National Aeronautics and Space Administrationwww.nasa.gov

NASA Distributed Active Archive Centershttp://nasadaacs.eos.nasa.gov

ii

Cover Images

Front

Clockwise from top left:

1. Students line up for school meals, made with locally produced foods,at Bar Sauri Primary School in the Millennium Village at Sauri, Kenya.(Image courtesy of Glenn Denning)

2. El Elegante Crater, part of the Pinacate Biosphere Reserve in north-western Sonora, Mexico, is a maar crater about 1.6 kilometers (1 mile) indiameter and 244 meters (800 feet) deep. (Image courtesy of JimGutmann)

3. The Slender Loris (Loris tardigradus) from Sri Lanka is assessed asendangered and is on the World Conservation Union's 2004 Red List ofThreatened Species. Habitat fragmentation over the years has seriouslyreduced the area available for this species. Between 1956 and 1993, SriLanka lost more than 50% of its forest cover to human activities, fol-lowed by a similar rate of decline in the remaining forest cover between1994 and 2003. (Image copyright K.A.I. Nekaris)

4. The Qinghai-Xizang railroad, scheduled to be completed in 2007, willtraverse some of the most desolate landscape on the Tibetan Plateau.(Image courtesy of Richard Armstrong)

5. An Anopheles mosquito takes blood from a human host. This mosquitois a vector for malaria, which kills more than a million people each yearin over 100 countries and territories. (Image courtesy of James Gathanyand the Centers for Disease Control and Prevention)

Back

The Multi-angle Imaging SpectroRadiometer (MISR) instrument cap-tured this image of an actinoform cloud on November 16, 2001. Theidentification of this unusual cloud formation, which looks very muchlike a leaf, prompted a search for more examples of this cloud type in theMISR data. (Image courtesy of NASA/GSFC/LaRC/JPL, MISR Team)

iii

Acknowledgements

We extend our gratitude to the Earth Science Data and InformationSystem (ESDIS) project for its support of this publication; to the DAACmanagers and User Services personnel for their direction and reviews;and to the DAAC scientists who alerted us to the research and investiga-tions that made use of DAAC data in 2005. A special thanks goes tothe investigators whose accomplishments we are pleased to highlighthere.

This publication was produced at the National Snow and Ice DataCenter DAAC under NASA contract No. NAS5-32392.

Supporting Earth System Science 2005 is also available online in PDF formatat: http://nasadaacs.eos.nasa.gov/articles/index.html.

For additional print copies of this publication, please contact:nasadaacs@eos.nasa.gov.

Researchers working with NASA DAAC data are invited to contact theeditor of this publication at daaceditor@nsidc.org to explore possibilitiesfor developing a future article.

Writing, Editing, and Design

Editor: Laurie J. SchmidtWriters: Amanda Leigh Haag, Kelly Kennedy, Laura Naranjo, Laurie J.Schmidt, and Evelyne YoheCover Design: Laura NaranjoInterior Design: Laura Naranjo

The design featured at the end of the articles throughout this edition isone of many African symbols, called adinkra. The symbols are prevalentin the country of Ghana, located on the Atlantic coast of Africa betweenthe Cote d’Ivoire and Togo. The adinkra symbols can be found in cloth,pottery, and on walls, and they reflect a system of human values thatembrace family, integrity, tolerance, harmony, determination, and protec-tion. The Denkyem (crocodile), the symbol featured in this edition, rep-resents adaptability, based on the fact that the crocodile lives in thewater, yet breathes the air, demonstrating an ability to adapt to chang-ing circumstances.

Using newly-updated population data sets archived at the NASADAACs, scientists are working to develop solutions to problems thatthreaten human populations in Africa, including hunger and poverty andthe spread of malaria (see Malaria by the Numbers, page 6, and War onHunger, page 31).

Symbol courtesy of:West African Wisdom: Adinkra Symbols & Meaningshttp://www.welltempered.net/adinkra/index.htm

Earth Science Data and Information Systems ProjectNational Aeronautics and Space AdministrationGoddard Space Flight CenterGreenbelt, MD 20771

Dear Friends and Colleagues:

From space, we can view our Sun and Earth as a whole system, observe the results of complex interactions, and begin to understand how both naturaland human-induced changes can impact life here on Earth.

NASA's Earth Observing System Data and Information System (EOSDIS) generates more than 2,400 Earth system science data products and pro-vides associated services to researchers involved in interdisciplinary studies. EOSDIS manages and distributes these products through the DistributedActive Archive Centers (DAACs). These nine data centers process, archive, document, and distribute data from NASA’s past and current EarthSystem Science research satellites and field programs. Each data center serves one or more specific Earth science disciplines and provides its science,government, industry, education, and policymaker communities with data information, data services, and tools unique to its particular science disci-pline.

This publication attempts to bring satellite science down to Earth. NASA: Supporting Earth System Science 2005 showcases recent research conduct-ed using data from the NASA EOSDIS data centers. These unique articles incorporate data from multiple sensors and across data centers. We focuson what scientific results mean with a global perspective—now, and in the long term—for Earth and its inhabitants. With these stories, we hope toacquaint you with the wealth of resources available through the DAACs.

Jeanne BehnkeEarth Science Data and Information Systems ProjectNASA Goddard Space Flight Center

iv

Making Waves in Tsunami Research 2Data from Jason and TOPEX/Poseidon give scientists the first detailed profile of a major tsunami event.Amanda Leigh Haag

Malaria by the Numbers 6Scientists use a combination of field research, census data, and population projections to address the malaria problem in Africa.Kelly Kennedy

Bloom or Bust: The Bond Between Fish and Phytoplankton 10Ocean color data from the SeaWiFS and MODIS sensors enable researchers to examine the link between phytoplankton blooms and fish and bird health.Laurie J. Schmidt

Cloudy with a Chance of Drizzle 13By analyzing data from the MISR instrument, scientists discover that a unique type of cloud formation is much more prevalent than was previously believed.Amanda Leigh Haag

Riding the Permafrost Express 17Researchers use innovative techniques to protect the newly constructed Qinghai-Xizang railroad across the Tibetan Plateau from permafrost.Evelyne Yohe and Laurie J. Schmidt

Silent Signals 20Synthetic aperture radar data aid scientists in detecting the ground deformation associated with earthquakes and volcanic eruptions.Kelly Kennedy

Color Data Image Section 23

A Rainforest Divided 27The Amazon rainforest is losing ground as human activity fragments the forests.Laura Naranjo

War on Hunger 31Scientists use geospatial analysis to identify pockets of hunger and the geographic and physical causes of hunger.Amanda Leigh Haag

Matter in Motion: Earth’s Changing Gravity 35A new satellite mission sheds light on Earth’s gravity field and provides clues about changing sea levels. Laura Naranjo

In Search of Martian Craters 39Craters on the Earth’s surface provide scientists with clues about how to identify craters on Mars.Laurie J. Schmidt

Checking Earth’s Vital Signs 44The Millennium Ecosystem Assessment is the first-ever attempt to take stock of the health of the planet as a whole. Amanda Leigh Haag

SPoRTscast 48Weather forecasters improve short-term predictions using data from NASA’s Earth Observing System satellites.Laura Naranjo

NASA: Supporting Earth System Science 2005

It was pure coincidence that in the early morn-ing light of December 26, 2004, within hoursof the 9.0 earthquake that shook the floor ofthe Indian Ocean, two joint NASA/FrenchSpace Agency satellites watched from highoverhead while tsunami waves silently racedacross the Bay of Bengal. Half a world away,U.S. Geological Survey geophysicist PeterCervelli came home from a Christmas dinnerwith friends to find seismometer readingsheralding ominous news.

Sadly, scientists had no way to warn officials orthe public about the deadly force that wasminutes away from surging ashore. More than220,000 lives were lost to the tsunami.

Today, scientists are gathering data from a vari-ety of sensors in an effort to reconstruct theevent and see what lessons they can learn fromit. Serendipitously, as the tsunami waves wererolling toward the shore, the “Jason” and“TOPEX/Poseidon” satellites recorded theheight changes of the waves as they formed—the first detailed measurements of their kindduring a major tsunami.

While satellites cannot provide an early warn-ing, their data hold great promise for helping

2

Physical Oceanography DAAC by Amanda Leigh Haag

The rushing waters of the 2004 tsunami approach theshore in Thailand. (Image courtesy of David Rydevik,Stockholm, Sweden)

Making Waves inTsunami Research

3

scientists improve computer models of wavebehavior during tsunamis. Scientists say thatbetter models will be the first line of defenseagainst the havoc tsunamis can cause on coastalareas. The data obtained from Jason andTOPEX/Poseidon—archived at NASA’sPhysical Oceanography Distributed ActiveArchive Center (PO.DAAC) in Pasadena,California—are enabling scientists to look atthe mechanisms that produced the killer wavesin the Indian Ocean.

Jason and TOPEX/Poseidon are assembling aglobal, long-term record of sea surface height,which is helping scientists better understandocean circulation and climate variability. Seasurface height reflects the storage of heat in theocean: when the ocean warms, it expands, thusraising the sea level, explained Lee-Lueng Fu,chief scientist of the satellite missions fromNASA’s Jet Propulsion Laboratory in Pasadena.

But detection of the recent tsunami is a goodexample of their secondary benefits, said Fu.“This happened by chance, because the satel-lites were not designed to make observations ofwaves moving as fast as a tsunami, whichattain the speed of a jet plane in the openocean,” said Fu. “At 500 miles per hour, thewaves are moving very quickly and are veryhard to detect.”

Since the satellites make 13 revolutions aroundEarth each day, the probability of catching atsunami in the way that Jason andTOPEX/Poseidon did is about one chance in50, said Fu. But their ability to detect minutechanges in sea surface height enabled them tospot subtle changes in the wave behavior.

When the tsunami passed through the Bay ofBengal, the satellites picked up a sea heightchange of half a meter, which is a huge signal,according to Fu. “This kind of first-handknowledge is helping researchers better under-stand how waves propagate in the ocean,” hesaid, “so they can refine their ability to pin-point where the wave is going to crash overbeaches and with how much energy.”

To make their measurements, Jason andTOPEX/Poseidon continually bounce radarpulses off the sea surface and record the time ittakes for the signal to return. Each radar pulsegives a measure of the satellites’ exact locationand altitude above the sea surface. Using thesemeasurements, scientists can also calculate thevelocity at which ocean currents are moving.

Jason is the satellite credited with making suchprecise measurements of the recent tsunamievent, according to Fu. Its observations includ-ed detailed ripples—on the order of 10 cen-timeters (about 4 inches) high—spread over a200-kilometer (124-mile) area. Previously sci-entists believed tsunamis to be single, fast-moving elevations of the sea surface over aspan of several hundred kilometers, accordingto Fu. Such movement would be akin to thesimple rhythmic rise and fall of taking in adeep breath and expelling it. But scientists

have learned from Jason’s recent observationsthat tsunami waves in the open ocean are morecomplicated than that. “This kind of measure-ment is telling us that a tsunami has a muchmore complex structure than we used tothink,” said Fu.

Fu is quick to point out, however, that satel-lites will never be able to provide a warningsystem from space because the cost of deploy-

The black areas along the axis labeled "Jason track"represent sea surface height measured by the Jasonsatellite two hours after the initial magnitude 9 earth-quake hit the region southeast of Sumatra (shown inred) on December 26, 2004. The data were taken by aradar altimeter onboard the satellite along a track tra-versing the Indian Ocean when the tsunami waves hadjust filled the entire Bay of Bengal. The maximum heightof the leading wave crest was about 50 cm (1.6 ft), fol-lowed by a trough of sea surface depression of 40 cm(1.3 feet). The arrows indicate the directions of wavepropagation along the satellite track. (Image courtesy ofNASA/JPL/CNES/National Institute of AdvancedIndustrial Science)

“This kind of first-hand knowledge is helping researchers better understand howwaves propagate in the ocean, so they canrefine their ability to pinpoint where thewave is going to crash over beaches and withhow much energy.” - Lee-Lueng Fu

4

ing enough satellites to be at any given pointover the ocean within half an hour is prohibi-tive. “I don’t want to give people the impres-sion, ‘oh, we caught this tsunami, therefore weshould launch more satellites to catchtsunamis.’ We just cannot afford to do that,”said Fu. Moreover, since Jason and TOPEX/Poseidon data take a minimum of five hours toprocess, there is a low likelihood of using theirdata in time to warn coastal residents of anapproaching tsunami. But in the rare case thatthe satellites do record the profile of a tsunami,they provide excellent hindsight because theyhave a continuous profile of the sea surfaceheight change. “There’s no other measurementthat can produce such a record,” said Fu.Ocean buoys, which are often separated bymore than 500 kilometers (about 310 miles),don’t record a continuous profile.

The key for early warning is to have more bot-tom-mounted pressure sensors in the ocean,according to Fu. But experts agree that evenhaving such scientific instrumentation in placewon’t be sufficient if the communication andeducation components of a warning system aremissing, as was the case on December 26.Cervelli, who’s based at the Alaska VolcanoObservatory in Anchorage, recalled themoment on Christmas night when the seis-mometer readings came in. “I remember think-ing to myself, ‘this is going to create a largetsunami that is going to kill a lot of people,’”he said. “Hundreds of people around theworld—when they saw the information—knewthat to be the truth, but there was nothingthat we could do,” he said.

Satellites carrying specialized radar instrumentsare making it possible for scientists, likeCervelli, to understand the geologic processesthat could lead to undersea landslides andpotentially devastating tsunamis in thefuture—perhaps in time to see them comingand to adequately prepare.

Cervelli, who has worked extensively on under-standing the volcanic processes of KilaueaVolcano, noted that in Hawaii, the governmenthas made a concerted effort to educate thepublic about what to do in case of a tsunami.Beginning in kindergarten, children are taughtin school and through the newspaper aboutwhat to do if they hear a tsunami warningsiren. “All of these things are now secondnature to most Hawaiian residents, but Hawaiiis a relatively small population in a first worldcountry,” said Cervelli. “The most challenging

thing for establishing a warning system in theIndian Ocean is going to be the education andcommunication components.”

Installing more buoys around the islands hithard by the recent tsunami would not be toocostly or difficult to do, Cervelli said. “But allof that is useless if there is no way to get thatinformation to officials in the countries thatwill be affected, and it’s also useless if you getthe information to an official, but the officialsare powerless to do anything about it. If peoplearen’t educated about what to do when thetsunami horn goes off, if there even is a tsuna-mi horn, then it doesn’t really matter,” saidCervelli.

Cervelli’s research is one example of the waythat scientists are using satellite data to under-stand tsunami-forming processes before theywreak havoc. The south flank of KilaueaVolcano on the island of Hawaii (the “BigIsland”) has been moving towards the sea at arate of 6 centimeters (2.3 inches) per year for atleast a decade. The fact that there’s been somuch motion has left some people to wonderwhether the south flank is a candidate for “cat-astrophic flank failure,” which would probablylead to a very large tsunami, Cervelli said. If itdid occur, it would threaten the HawaiianIslands, and under some models, the west coastof the United States, South America, and possi-bly Japan.

Some evidence suggests that very largetsunamis caused by massive undersea landslideshave hit the Hawaiian Islands in the not-too-distant geologic past. Researchers have found

A building destroyed during the Hilo, Hawai'i, tsunamion April 1,1946. Every house on the main street facingHilo Bay was washed across the street and smashedagainst the buildings on the other side. (Image courtesyof NOAA and the U.S. Army Corps of Engineers)

anomalous corals and marine shells depositedhundreds of meters high above the shoreline ofsome islands. If it has happened before, someexperts wonder whether giant tsunamis mayplague the Hawaiian Islands from time totime.

So scientists like Cervelli are using a techniqueknown as Synthetic Aperture Radar Interfer-ometry (InSAR) to understand tectonic systemsand, hopefully, identify high-risk areas.Satellites carrying InSAR instruments beam aradar signal down onto a given location of theEarth’s surface at two different points in time.The second measurement reveals whether theground has shifted either toward or away fromthe satellite, explained Cervelli.

While InSAR data can provide a glimpse intothe geologic future, they are by no means acrystal ball. “I’m not suggesting that we canpredict earthquakes. That, so far, has provenvery difficult, if not impossible. But we cangive you an idea that ‘well, this fault is slippingor it’s accumulating strain, so eventually it’sgoing to result in an earthquake,’” saidCervelli. “The trick is telling you on what dayand at what time it’s going to happen.”

So for now, scientists will take what they canget. Such fine resolution and detail from satel-lite data will help fortify the defense againsttsunamis in the future. Computer modelers willbe better equipped to compute the strength

and pattern of the tsunami waves, explainedFu. “That’s a critical link,” he said. “Our datawill aid researchers by improving their under-standing of tsunami dynamics in order to makebetter models. So that’s the benefit of quite aserendipitous measurement.”

References:Cervelli, Peter. 2004. The Threat of Silent

Earthquakes. Scientific American.Fu, L.-L., and A. Cazenave, editors, 2001.

Satellite Altimetry and Earth Sciences: A Handbook of Techniques and Applications.San Diego, CA: Academic Press.

For more information, visit the followingweb sites:Physical Oceanography (PO) DAAC

http://podaac.jpl.nasa.gov/Ocean Surface Topography from Space

http://sealevel.jpl.nasa.gov/USGS Earthquake Hazards Program

http://earthquake.usgs.gov/eqinthenews/2004/usslav/

USGS Hawaiian Volcano Observatoryhttp://hvo.wr.usgs.gov/

Pacific Tsunami Museumhttp://www.tsunami.org/

5

Peter Cervelli is a U.S. Geological Survey research geo-physicist at the Alaska Volcano Observatory inAnchorage, and an adjunct professor in the Departmentof Geology and Geophysics at the University of Alaska,Fairbanks. His research focuses on deformation processesat Kilauea Volcano on the “Big Island” of Hawaii.Cervelli is credited with discovering the first “silentearthquake” to occur at a volcano after he and colleaguesobserved that Kilauea’s south flank had imperceptiblyshifted 10 centimeters over a course of 36 hours inNovember 2000. He holds a PhD in geophysics fromStanford University.

Lee-Lueng Fu is a senior researchscientist at the Jet PropulsionLaboratory (JPL) at the CaliforniaInstitute of Technology, and aNASA project scientist for the jointU.S./French TOPEX/Poseidon andJason missions. He is also head ofthe Ocean Science Research Element

of the Division of Earth and Space Sciences at JPL. Since1988, Fu has led an international team of oceanogra-phers and geophysicists in the development of precisionmeasurements of ocean surface topography fromTOPEX/Poseidon and its follow-on instrument, Jason.He holds a PhD in oceanography from MassachusettsInstitute of Technology and Woods Hole OceanographicInstitution.

6

When Bob Snow of the Kenya MedicalResearch Institute/Wellcome TrustCollaborative Programme in Nairobi travelsthrough Africa, he sees firsthand the effects ofmalaria. Children suffer from fevers, chills,seizures, and anemia, and then often lapse intocomas and die. Sometimes the victims get themedicine they need, but more often, parentswho can’t afford food or school fees must gowithout the drugs that could save their chil-dren.

The World Health Organization (WHO) esti-mates that more than a million people in Africadie from malaria every year, including 3,000children each day. “It’s a huge problem,” saidSnow, who is also a professor of tropical publichealth at the University of Oxford. “The NewYork Times reported 150,000 deaths from the

recent tsunamis in Indonesia, but that numberis only the margin of error for malaria deaths inAfrica.” Despite 20 years of research in Africaon ways to combat the burden posed by malar-ia, Snow said he hasn’t seen much progress inthe fight against the disease.

Malaria is caused by a parasite carried by thefemale mosquito Anopheles. Mosquitoes trans-mit the parasite to humans through their bite,and they also contract it by biting already-infected humans. The parasite travels first tothe liver, where it reproduces, then proceeds tothe bloodstream, where it reproduces again anddestroys the red blood cells. In Africa, a partic-ularly dangerous strain of malaria, known asPlasmodium falciparum, resists most drugs andoften kills its victims.

Problems in Africa are compounded by the factthat malaria helps create a “circle of poverty.”Because of high infant mortality, women tendto have more children to ensure that some sur-vive. With many children to care for, motherscan’t leave the home to help provide for theirfamilies. So, malaria burdens individual house-holds, as well as local health services.

In the 1980s, scientists found that using bednets significantly decreased the chance of get-ting malaria, yet fewer than 5 percent ofAfrican children sleep under them, accordingto Snow. In addition, limited funding leads tothe use of cheaper drugs to which the malariaparasite has developed resistance, which makesthem ineffective. “Malaria victims are beingtreated for a life-threatening disease with drugsthat don’t work,” he said.

Malariaby theNumbersSocioeconomic Data and ApplicationsCenter (SEDAC)

by Kelly KennedyAn Anopheles mosquito takes blood from a human host. This mosquito is a vector for malaria, which kills more thana million people each year in over 100 countries and territories. (Image courtesy of James Gathany and the Centersfor Disease Control and Prevention)

7

Targeting bed nets and effective drugs to spe-cific populations could cut the spread of malar-ia in half, Snow said, but researchers must haveaccurate population data to determine exactlywhere they should concentrate their efforts. Forexample, Snow said the United Kingdom gavea non-governmental organization (NGO) $20million to market bed nets. The NGO market-ed the nets in a city where people were lesslikely to contract malaria and, therefore, theproject failed. “It’s a complete mess on theinternational level,” he said. “There’s a lot ofrhetoric and a lot of hand-holding and a lot ofmeetings, but that doesn’t translate to solu-tions.”

Andrew Tatem, a zoology researcher at theUniversity of Oxford, said that before scientistscan make headway in fighting malaria, theymust first understand how the disease affectsdifferent populations. “There is so much wedon’t currently know about malaria,” he said.“The basic numbers must be understood.”The WHO, along with the United NationsDevelopment Program and the World Bank,plans to “Roll Back Malaria” by half by 2010.But Tatem said no one knows exactly what“half ” is. If malaria researchers had accuratenumbers for specific locations, they might beable to attack the problem in a proactive way.Data, Snow said, could provide solutions.

Deborah Balk, lead project scientist for theSocioeconomic Data and Applications Center(SEDAC) at Columbia University, and her col-leagues recently released a grid that showspopulation densities for specific regions notonly in Africa, but in the rest of the world aswell. The Gridded Population of the World(GPW) data set, version 3, converts censusinformation from more than 375,000 adminis-trative subdivisions, such as countries andprovinces, into a series of longitude/latitudegrids that provide estimates of where thosepopulations live. The data set allowsresearchers to easily analyze population datawith other geographically referenced data, suchas land-use patterns, geophysical hazards, andclimate information.

Version 3 of the data set includes more inputunits (individual sets of data for the adminis-trative units) than the first two versions ofGPW, allowing Balk to create smaller grids formany countries. “We did a massive updatewhen the year 2000 census updates came out,”Balk said. “In addition, we collected a differentstream of census data—population estimates ofhuman settlement and urban areas—andmatched it with night-time-lights satellitedata.”

The third version of GPW spawned two newprojects: the Global Rural Urban Mapping

Project (GRUMP) and the Gridded Populationof the World, Future Estimates 2015. GRUMPadds a new dimension to the gridded popula-tion data by consistently defining how manypeople live in urban versus rural areas. “Thishas never been done before,” said Balk. “Itallows us to ask, ‘What geophysical and envi-ronmental features define city location?’ and‘Why are coastal areas disproportionatelyurban?’”

Balk, who works at the Center forInternational Earth Science InformationNetwork (CIESIN) at Columbia University,where SEDAC is based, said that she and hercolleagues have already begun using these newdata to investigate where people live relative toecosystems and urban areas. For example, theyfound that while coastal ecosystems in Africaare densely populated and disproportionatelyurban, close to 45 percent of Africans live incultivated zones, with only about 40 percent ofthose residents being urban dwellers. Withthese data, scientists can look at where malaria-

“We now have maps of malaria risk withaccuracy to within 1 kilometer. Once youhave the numbers, you can come up witha plan and goals to work out how you’regoing to attack the problem.” - Andrew Tatem

The Gridded Population of the World (GPW) data sethas undergone a substantial improvement in the num-ber of input units and in the target years of estimation.This table describes these improvements, and alsorefers to the new Global Rural-Urban Mapping Project(GRUMP) data collection, which shows how many peo-ple live in urban versus rural areas. (Information cour-tesy of Socioeconomic Data and Applications Center)

8

ridden mosquitoes live in relation to people,and learn about the effects of populationgrowth on the environment.

Although field research shows that malaria ismore prevalent in rural areas, the combinedGPW, GRUMP, and satellite data allow scien-tists to better estimate malaria risk and bur-den. “You’re less likely to be bitten by a malar-ious mosquito in an urban area,” Tatem said.So, the new data give scientists a better idea ofwhere to send bed nets and medicine.

Tatem works with Simon Hay, also of theUniversity of Oxford and principal scientist forthe malaria project, to combine the GRUMPpopulation data with maps of areas consideredto be at-risk from malaria. Using the data todetermine disease distribution, they study theeffects of climatic conditions on malaria. “Wenow have maps of malaria risk with accuracy towithin 1 kilometer,” Tatem said. “Once youhave the numbers, you can come up with aplan and work out how you’re going to attackthe problem.”

While the new population data promise to aidresearchers in cutting malaria risk, compilingthe data sets is not without obstacles.Although most nations have been open withtheir data, Balk said it’s not always easy to getinformation. And for many countries, data maynot be redistributable. Such was the case withthe Indonesian census. Aid workers immediate-ly called Balk after the tsunami hit inDecember 2004 to see if she could share herhigh-resolution input data for Aceh Province.Unfortunately, because of perceived securityrisks, Balk said the Indonesian governmentrequired her to sign a contract agreeing not torelease the information to other parties. “Wemade the gridded data and estimates ofexposed population available to the aid agen-cies as soon as possible, but we could notrelease the underlying data,” Balk said. “Thatwas a real shame.”

In addition to financial and political hin-drances, some governments simply don’t havethe information. War-torn countries, such asthe former Yugoslav republics and Afghanistan,don’t collect census data regularly or have hadtheir regular census-taking interrupted. Othernations, such as Rwanda and Uganda, haveseen their populations move to other countriesto seek safety.

There is enough census data in Africa, however,to try to sort out the malaria problem. Doctorsand scientists have solved the malaria problembefore, according to Tatem. “Malaria used to bea problem in the central United States and inNorthern Europe, but global efforts pushed itback to the tropics,” he said.

More than 3,000 children in Africa die from malaria every day. (Image courtesy of Pedro Sanchez, Earth Institute atColumbia University)

9

The WHO sprayed DDT from the 1940s until 1969, when environmen-tal concerns forced them to stop. And, as Europeans improved theirhousing with windowpanes, screens, and doors, they lessened malariarisks there. Growing cities and changing land use also removed manymosquito breeding grounds, but these precautions were often not takenin Africa. “Once the African countries gained independence, funding forcontrolling malaria dried up,” Tatem said.

Snow said solving the malaria problem boils down to understandingwhere people live in relation to the malaria risk, and that’s why theGPW, GRUMP, and population prediction data are so important.Analysis using GPW and UN projected population growth suggests 400million births will occur within the malaria-infested areas of the world inthe next five years, according to Tatem, but even that isn’t a clear indica-tor of future malaria numbers. For example, the population boom willpossibly cause those areas to become more urban, thus reducing mosqui-to habitat.

At the turn of the 20th century, about 77 percent of the world’s popula-tion was at risk of contracting malaria. By 1994, that number had fallento 46 percent. But in 2002, it went back up to 48 percent because ofpopulation growth in at-risk areas. This underscores the need for accu-rate population data, and for a thorough understanding of how popula-tion distribution affects the spread of infectious diseases.

Despite his frustrations, Snow continues to conduct research in Kenyaand work with African governments to control the problem. “Africangovernments can’t fight this alone—it takes money,” he said. “And weneed to get the numbers first.”

“That is why I keep going. It’s a lifetime of work.”

Reference:Malaria is alive and well and killing more than 3000 African children

every day. World Health Organization. Accessed January 10, 2005.http://www.who.int/mediacentre/news/releases/2003/pr33/en/

For more information, visit the following web sites:Center for International Earth Science Network (CIESIN)

http://ciesin.columbia.eduSocioeconomic Data and Applications Data Center (SEDAC)

http://sedac.ciesin.columbia.edu/Gridded Population of the World

http://beta.sedac.ciesin.columbia.edu/gpw/World Health Organization “Roll Back Malaria”

http://www.rbm.who.int/cgi-bin/rbm/rbmportal/custom/rbm/home.do

Deborah Balk is an associate research scientist at ColumbiaUniversity’s Center for International Earth Science InformationNetwork (CIESIN), one of the Earth Institute’s centers, and the leadproject scientist for NASA’s Socioeconomic Data and ApplicationsCenter (SEDAC). She also heads CIESIN’s spatial demographyefforts, where she led development efforts on the GriddedPopulation of the World data set and spearheaded the GlobalUrban-Rural Mapping Project. Currently, Balk also co-leads efforts

at CIESIN to map and conduct spatial analysis in support of the United NationsMillennium Project. She received a PhD in demography from the University ofCalifornia, Berkeley.

Robert Snow is a professor of tropical public health in the Centre forTropical Medicine at the University of Oxford, and head of thePublic Health Group at the KEMRI/Wellcome Trust Programme inNairobi, Kenya. In 1997, Snow was named an honorary fellow atthe Liverpool School of Tropical Medicine, and in 1999, he earnedthe Chalmers Medal for Tropical Medicine from the Royal Society ofTropical Medicine and became a senior research associate at theWellcome Unit for History of Medicine at the University of

Medicine. Snow earned a PhD in epidemiology from the London School of Hygiene andTropical Medicine.

Andrew Tatem is a research officer on a Wellcome Trust-fundedproject aimed at using satellite imagery to map settlements andpopulations across Africa. His most recent research focuses on apply-ing satellite data-based solutions to public health problems, particu-larly the spread of malaria. He earned a PhD in electronics andcomputer science from the University of Southampton, UK.

10

GSFC Earth Sciences Data and Information Services CenterDAAC

by Laurie J. Schmidt

When you’re a juvenile salmon trying to survive into adulthood, timingis everything. If you’re lucky enough to be born in a year when there’s anample supply of zooplankton, then this food source helps keep your pred-ators at bay. If, however, you’re born during a lean plankton season,there’s a good chance you might become a snack for fish like pollock andherring.

Oceanographers have long known that food web dynamics influence fishand bird populations in marine coastal areas. But new evidence has ledsome researchers to suspect that Mother Nature may have a unique wayof sustaining populations by bringing young fish, like salmon, into theworld when food sources are at their maximum.

The marine food chain starts with microscopic plants called phytoplank-ton, which typically float close to the surface where there is sunlight forphotosynthesis. Phytoplankton are eaten by slightly larger, more mobile,herbivores called zooplankton, which range in size from single-celledorganisms to jellyfish. In turn, zooplankton provide food for krill andsome small fish.

Sudden explosive increases in phytoplankton, called “blooms,” occur inthe ocean when nutrient and sunlight conditions are just right. The tim-

Bloom or Bust:

Kachemak Bay, located off the shores of Homer, Alaska, is home to a variety of fishand bird species. (Image courtesy of John Maurer)

The Bond Between Fish and Phytoplankton

11

ing of these blooms plays a large role in main-taining marine ecosystems, and is crucial to thesurvival of certain fish and bird species. “Intheory, you have to have high levels of phyto-plankton to support a high abundance of zoo-plankton, which then supports an abundanceof small feeder fish,” said oceanographer ScottPegau.

Pegau, a researcher at the Kachemak BayResearch Reserve in Homer, Alaska, has beenusing satellite data to look at the possible linkbetween phytoplankton blooms and fisheriesand bird health. By analyzing imagery fromocean color sensors, like the Sea-viewing WideField-of-view Sensor (SeaWiFS) and theModerate Resolution Imaging Spectro-radiometer (MODIS), Pegau and his colleaguescan get a good look at both the timing andlocation of phytoplankton blooms and, hope-fully, identify significant patterns that mightlead to better fisheries management. “All fishare tied back to phytoplankton somewherealong the line,” said Pegau. “So, most of thecoastal fisheries could benefit from understand-ing bloom patterns and being able to makeinferences to survival rates.”

Kachemak Bay is located off the shores ofHomer, Alaska, on the southwest side of theKenai Peninsula, a region dependent on recre-

ational and commercial fishing for its liveli-hood. “Fisheries are extremely important tocoastal Alaska,” said Pegau. “Many of the com-munities are strictly fishing communities—it’sthe sole dimension to their economy.”

But during the past 10 years, several fisherieshave collapsed in the Kachemak Bay area,including king crab, tanner crab, Dungenesscrab, and shrimp. According to Pegau, thesecollapses are due partly to ocean circulationchanges, but they also can be traced to a poorunderstanding of the variables that can influ-ence fish populations in a given year. “We’restill getting millions of pounds of salmon fromthis area, but better fishery management tech-niques could help ensure that the industry as awhole survives,” said Pegau.

Understanding phytoplankton bloom patternsand their effects on fish populations couldspawn new management practices that helpsafeguard the future of the Alaska fishingindustry. But without satellite data, Pegau andhis team wouldn’t be able to see an area largeenough to detect any clear trends. “KachemakBay has 320 miles of coastline, and it’s just asmall dot on the Alaska map,” said Pegau.“There is no way we can monitor what’s hap-pening in coastal Alaska without ocean colorsensor data—SeaWiFS gives us the opportunityto see the big picture.”

In the world’s oceans, color varies according tothe concentration of chlorophyll and otherplant pigments in the water. Satellite sensorsthat detect these subtle changes in ocean colorare critical tools for scientists studying oceanproductivity. By analyzing data from SeaWiFS

and MODIS, Pegau and his team discoveredtwo patterns in the Gulf of Alaska: highchlorophyll concentrations show up inApril/May, and then again in September/October. But the satellite data also show athird bloom that occurs in late June or earlyJuly.

The bloom timing is significant to both fishand birds because it determines when foodsources are available. “If you’re a fish gettingready for winter, then a fall bloom will providea food source that will allow you to fatten upenough to get through the winter,” Pegau said.“Conversely, if you’re a bird, you want thebloom to happen just before your eggs hatch inthe spring. If the bloom happens in late July,then it may be too late to be of any value toyou.”

But exactly what causes the blooms to occurwhen they do is still a mystery to oceanogra-phers. One theory, Pegau explained, is that thetiming may be related to the strength of winterstorms. Like all plants, phytoplankton needlight, and getting light in the ocean means

Homer is one of many communities in coastal Alaskathat are dependent on recreational and commercial fish-ing for their livelihood. (Image courtesy of John Maurer)

“All fish are tied back to phytoplanktonsomewhere along the line. So, most of thecoastal fisheries could benefit from under-standing bloom patterns and being able tomake inferences to survival rates.”- W. Scott Pegau

12

being able to stay near the surface. If an area isplagued by continual storms, then the phyto-plankton keep getting mixed down into deeperwater—away from the light. But at the sametime, they need the storms to bring nutrientsup from the deep and fertilize the ocean sur-face. “It’s a tricky balance,” said Pegau. “Youneed enough storms to bring the nutrients up,but then you also need the storms to ceaseearly enough to allow things to grow.”

The team is also using the ocean color data tolook at the spatial patterns of blooms, whichare believed to have a significant effect on birdpopulations. “If the area that has a lot of phy-toplankton is located far offshore, then the par-ent birds have to fly farther out to get it and,therefore, have less energy to provide for theiroffspring when they return,” he said. “They eatup more of their food flying than they would ifconditions kept the blooms closer to shore.”

Having phytoplankton blooms occur far fromshore poses risks not only to birds, but tosalmon as well. “If food sources are located far-ther offshore, the juvenile salmon are forced to

move away from shore to a place where they’remore vulnerable to predators,” said Pegau.“The farther out they have to swim to findfood, the more likely they are to be taken bypredators.”

While many fish spend their entire lives in theocean, other fish—such as salmon—spend aportion of their lives at sea and then return torivers to spawn. “The ocean provides a muchlarger food source than the rivers and lakeswhere these fish are born,” said Pegau. “Bygoing out into the ocean to feed, they can growmore rapidly and reach the size and maturitythat allows them to spawn. And the faster afish can grow, the sooner it stops being prey.”

Ted Cooney, Professor Emeritus in the Instituteof Marine Sciences at the University of Alaska,Fairbanks, believes predators play a significantrole in the relationship between salmon andplankton. He and his colleagues have beenlooking at survival rates of juvenile salmon inPrince William Sound, a large region of pro-tected waters located on the east side of theKenai Peninsula. “About 800 million juvenilesalmon enter Prince William Sound fromstreams and hatcheries in April and May eachyear,” said Cooney. “These small fish immedi-ately encounter a host of predators, includingwalleye pollock, Pacific herring, cod, and vari-ous seabirds and marine mammals.”

The record in the Prince William Sound areasuggests that when zooplankton stocks arehigh in the spring, pollock and herring derivemost of their energy from that source, Cooneysaid. But when zooplankton stocks are low,they’re forced to derive more of their energy byfeeding on small fishes—like juvenile salmon.

Pegau’s team has taken an important first stepby identifying the fall and spring bloom pat-terns, but more work needs to be done beforelong-term trends can be reported with certain-ty, he said. “Our hunch is that we’re eventuallygoing to find a connection between juvenilesalmon and bird survival rates and the phyto-plankton bloom patterns,” he said.

“The big thing is being able to manage thefisheries more effectively—knowing whichyears you can expect a big return to come in,and which years aren’t going to produce much.You don’t want to end up over-harvesting ifthe ocean isn’t supporting the fish that year.”

For more information, visit the followingweb sites:GSFC Earth Sciences Data and Information

Services Center DAAChttp://daac.gsfc.nasa.gov/

Ocean Color Data Supporthttp://daac.gsfc.nasa.gov/oceancolor/

SeaWiFS Projecthttp://oceancolor.gsfc.nasa.gov/SeaWiFS/

What Are Phytoplankton?http://earthobservatory.nasa.gov/Library/Phytoplankton/

Each spring and summer, more than 16,000 birdsmigrate to Gull Island in Kachemak Bay. (Image cour-tesy of John Maurer)

W. Scott Pegau is an oceanographerat the Kachemak Bay ResearchReserve in Homer, Alaska, where heestablished a research program toaddress ocean circulation and pro-duction issues in Kachemak Bay,Lower Cook Inlet, and the sur-rounding Gulf of Alaska waters. He

holds a PhD in oceanography from Oregon StateUniversity.

13

NASA Langley Atmospheric SciencesData Center DAAC

by Amanda Leigh Haag

Ask just about anyone who’s lived along theSouthern California coast what it’s like to plana picnic or a day at the beach during themonths of May and June. Chances are they’llwarn you about the seemingly endless chain ofdreary, overcast days known locally as “MayGray” and “June Gloom.” The heavy cloudbanks rolling in off the ocean do little morethan spoil an otherwise perfect sunny day anddrop a harmless drizzle from time to time. Tosomeone hoping to spend a day basking in thesun and surf, these clouds are just gray, gloomy,and boring.

But scientists who study cloud behavior arelearning that these dismal clouds hold a lotmore interest than meets the eye. Researchersare finding that the western coasts of conti-nents where May Gray and June Gloom-typecloud systems occur are bountiful huntinggrounds for unique formations called actino-form clouds. Named after the Greek word for“ray” due to their radial structure, these previ-ously overlooked clouds are only now cominginto focus as scientists use satellite data to bet-ter understand their complexity.

Children around the world learn that cloudstake on unique shapes in the sky and thatcloud patterns often foretell the oncomingweather. Yet scientists still know very littleabout what determines the shape of individualclouds. Michael Garay, a graduate student at

Cloudy with aChance of Drizzle

Each year, Southern Californians experience a long cycle of dreary, overcast days known as May Gray and JuneGloom. Scientists are learning that regions where these systems occur are good hunting grounds for actinoformcloud formations. (Image from Photos.com)

14

the University of California, Los Angeles(UCLA), is one of a handful of experts who aretrying to discover what gives actinoform cloudstheir complex shape and what role they play inclimate and weather systems. “In a general

sense, we understand that stratus clouds areformed by one type of process and one type ofatmospheric conditions, while cumulus—thepuffy clouds—are formed by another process,”said Garay. “But when it gets down to the

details of ‘why this cloud looks like a bunnyand this cloud look like a horse,’ that’s reallyhard to understand.”

But the actinoform clouds that Garay and hiscolleagues are studying are not the kind thatyou can see while gazing at the sky. They areactually cloud fields—collections of clouds thatcan be up to 300 kilometers (186 miles) across,an area roughly the size of the state of SouthCarolina and a field of view too large to seewith the naked eye. In addition, they can form“trains” that are up to six times the length ofthe original cloud field, yet they maintain theirown, distinct identity. Using satellite data fromthe Multi-angle Imaging SpectroRadiometer(MISR)—one of five instruments onboardNASA’s Terra spacecraft—scientists are findingthese clouds to be much more prevalent andcomplex than they originally thought.

When viewed in a satellite image, actinoformclouds look like distinct leaf-like or spokes-on-a-wheel patterns that stand out from the restof the low-lying cloud field. Bjorn Stevens, ameteorology professor at UCLA, describesthem as being similar to a “leaf floating in apond.” Embedded within the low-lying cloudsthat give rise to what Southern Californianscall May Gray and June Gloom, actinoformclouds have a distinct pattern that doesn’t fadeaway towards infinity, he said. “Imagine yourgrandmother’s quilt where one of the squaresreally stands out as being something different,”said Stevens. Like the square in the quilt, thecloud will hold its shape and stand out fromthe rest, he explained. “It will move with thewind, but it doesn’t grow or contract verymuch—it just has its own life,” he said.

The MISR instrument captured this image of an actinoform cloud on November 16, 2001. The identification of thiscloud, which looks very much like a leaf, prompted a search for more examples of this cloud type in the MISR data.(Image courtesy of NASA/GSFC/LaRC/JPL, MISR Team)

15

Actinoform clouds are not a recent discovery;they appeared in meteorological literature asfar back as the early 1960s, when the firstweather satellites began sending back images.But until the late 1990s, scientists had dis-missed them as merely a transitional formbetween other more familiar types of clouds,and they were all but forgotten. In fact, theterm “actinoform” isn’t included in the 2000edition of the Glossary of Meteorology, whichis considered to be the comprehensive referencemanual for meteorologists. “There’s also arecent review of all the research that’s beendone in the past 20 years on low-level cloudsand they don’t even mention actinoformclouds,” said Garay. “People just kind of forgotthey were there.”

The NASA Langley Atmospheric Sciences DataCenter (LaRC) played an important role in therediscovery of actinoform clouds. Garayrecalled that his advisor, Roger Davies, now atthe Jet Propulsion Laboratory (JPL) inPasadena, Calif., was looking through imageryfrom the MISR instrument, which is archivedand distributed by the LaRC DistributedActive Archive Center (DAAC).

MISR collects data on the sunlit side of theEarth, using cameras that look in nine differentdirections simultaneously. The change in reflec-tion at different viewing angles enablesresearchers to distinguish different types ofatmospheric particles (aerosols), cloud forms,and land surface covers. Davies noticed a giant,leaf-shaped cloud that took up almost theentire width of one image, recalled Garay.“This sent me on a very interesting chasethrough the meteorological literature looking

for pictures of these clouds, because as good asthese database searches are—and they get bet-ter all the time—it’s hard to put in a searchterm like ‘cloud that looks like a leaf’ and getthe right information, much less come up withthe right picture,” said Garay.

Garay eventually tracked down satellite imagesof actinoform clouds taken by early weathersatellites in the 1960s. But what he uncoveredabout the clouds themselves was quite unex-pected. Actinoform clouds had been thought tobe a relatively uncommon transitional form.However, by sifting through MISR images ofthe western coast of Peru, Garay found thatactinoform-like clouds showed up roughly aquarter of the time as distinct formations with-in the more common stratocumulus clouds inthat region. Closer examination showed thatactinoform clouds occur worldwide in nearlyevery region where marine stratus or stratocu-mulus clouds are common, particularly off thewestern coasts of continents—especially Peru,Namibia in Africa, Western Australia, andSouthern California. Such cloud systems arepersistent year-round off the coast, yet in cer-tain seasons they blow ashore and create thegloomy “May Gray” effect on land.

And to Garay’s astonishment, the closer helooked at the clouds, the more complex hefound their patterns of organization to be.“This is something you wouldn’t expect,” saidGaray. “We don’t have a good understandingof why they have this radial structure to themand why it’s fairly common.”

So what might these elaborate features mean toweather systems and climate patterns?

Unfortunately, there’s no simple answer,according to Garay. “Clouds trace atmosphericmotion, so they’re responding to the atmos-phere in a complicated way,” said Garay.Clouds are complex: they change in response tosmall-scale effects, like the presence of a hill orthe local wind, but they also respond to large-scale weather systems, such as cold fronts andthe presence of the jet stream, said Garay.

Stevens, who studies a closely related cloudform dubbed “pockets of open cells,” or POCs,has a theory that may begin to explain thecomplex interactions between clouds and theclimate system. Both POCs and actinoformclouds have a compact and distinct shapeembedded within low-level marine stratusclouds. But the taxonomy to determine theexact relationship between actinoform cloudsand POCs hasn’t yet been worked out. “Howsimilar they are is an open question,” saidStevens.

The open cells that Stevens studies are one oftwo well-known types of stratocumulus cloudformations: open and closed cells. Open cellsresemble a honeycomb, with clouds around theedges and clear, open space in the middle.Closed cells are cloudy in the center and clearon the edges, similar to a filled honeycomb.Like actinoform clouds, the ‘cells’ in this casedo not refer to a single cloud but to the systemof associated clouds. Previously, researchersbelieved that actinoform clouds represented a

“A recent review of the research on low-levelclouds doesn’t even mention actinoformclouds. People just kind of forgot they werethere.” - Michael Garay

transitional form between open and closed cells, but the recent findingson actinoform clouds show that they are clouds in their own right.

Stevens’ observations from field studies in the Pacific seem to indicatethat when marine stratus clouds exist alone, in the absence of these opencells, the cloud formations are associated with little, or no, drizzle. Yetwhen the open cells are present—and likewise actinoform clouds—thereseems to be a corresponding increase in drizzle. Thus, his data suggestthat the presence of POCs and actinoform clouds is related to the onsetof precipitation. “The shape of the cloud field reorganizes itself when theclouds begin to rain,” said Stevens. In fact, Stevens has been able to mir-ror this reorganization using a computer model of a drizzling cloud field.

Stevens’ findings suggest that clouds play an important role in regulat-ing climate. Clouds reflect incoming sunlight, and they also act as ablanket that slows the upward movement of heat from Earth’s surfaceback into space. This is why cloudy days are generally cooler than sunnydays, but cloudy nights are generally warmer than clear nights. Thus,cloud cover is instrumental in setting Earth’s energy balance and control-ling its internal “body” temperature. “It makes the observation of thesepeculiar cloud forms, which seem to be linked to the development ofprecipitation, more interesting than they would’ve otherwise been,” saidStevens.

So for those of us who spent some of our childhood days lying with ourbacks in the grass and gazing at the sky, the prevalence of unexpectedcloud features in the sky might not come as a particular surprise. But forscientists, it raises the question of whether looking for patterns in theclouds might help unveil some of the deeper mysteries of the climate sys-tem. Either way, it appears that these clouds will continue to puzzle chil-dren and scientists alike. “It’s really kind of fascinating,” said Garay.“You’d expect these clouds to be sort of round or flat. But looking like awheel with spokes coming out of them is just kind of surprising.”

References:Garay, M.J., R. Davies, C. Averill, and J.A. Westphal. 2004. Actinoform

clouds—Overlooked examples of cloud self-organization at the mesoscale. Bulletin of the American Meteorological Society 85(10): 1585-1594.

Stevens, B., G. Vali, K. Comstock, M.C. van Zanten, P.H. Austin, C.S. Bretherton, and D.H. Lenschow. 2005. Pockets of open cells and drizzle in marine stratocumulus. Bulletin of the American Meteorological Society 86(1): 51-57.

For more information, visit the following web sites:MISR (Multi-angle Imaging SpectroRadiometer)

http://www-misr.jpl.nasa.gov/NASA Langley Atmospheric Sciences Data Center DAAC

http://www-misr.jpl.nasa.gov/Clouds are Cooler Than Smoke

http://earthobservatory.nasa.gov/Study/SmokeClouds/Tracking Clouds

http://earthobservatory.nasa.gov/Study/tracking/Clouds in the Balance

http://earthobservatory.nasa.gov/Study/CloudsInBalance/Clouds from a Different Angle

http://nasadaacs.eos.nasa.gov/articles/2004_clouds.html

16

Michael Garay is a graduate student at the University of California, Los Angeles, wherehe earned a MS degree in atmospheric sciences and is currently working on developing aproposal for a PhD in atmospheric sciences. He also works in the Multi-angle ImagingSpectroradiometer (MISR) group at the Jet Propulsion Laboratory (JPL), CaliforniaInstitute of Technology. Garay’s work includes developing cloud masks and assessing theperformance of existing cloud masks for the MISR instrument.

Bjorn Stevens is an associate professor of dynamic meteorology atthe University of California, Los Angeles. His research focuses oncloud climate interactions, particularly those associated with shallowmaritime convection.

17

National Snow and Ice Data CenterDAAC

by Evelyne Yohe and Laurie J. Schmidt

Behind the Himalaya Mountains lies a cold,isolated landscape, where the average elevationis higher than most of the Rocky Mountains inNorth America. Often referred to as “The Roofof the World,” the Tibetan Plateau is thelargest and highest plateau on Earth. Treeless,except for a few river valleys, the Plateau is anexpansive alpine zone, with more than 17,000glaciers covering its surface.

Historically, paths and roads built for tradeconnected the Tibetan people to neighboringregions, but journeys into or out of Tibet werelong and difficult. Travel between Lhasa andtowns within China’s Qinghai or Sichuanprovinces could take six months to a year. Evenin the early 1950s, lack of adequate roads andrailways forced China to use camels to trans-port cargo to Tibet. On average, 12 camelsdied for every kilometer the caravan traveledacross the Tibetan Plateau and over highmountain passes.

Beginning in 2007, the Qinghai-Xizang rail-road will connect Lhasa to the rest of China,providing a major access route into Tibet. Its

pressurized cars will protect passengers fromthe extreme altitude along the route, much ofwhich lies at least 4,000 meters (about 13,000feet) above sea level.

But building a railroad across the highestplateau in the world is laden with constructionhazards. Due to the thin air, unacclimatedworkers risk nosebleeds, blackouts, and evendeath. So, they carry oxygen bags, undergodaily medical monitoring, and work no morethan six hours a day. In addition, workersrotate off the Plateau every few weeks to avoidprolonged exposure to the extreme climateconditions.

In addition to the risks associated with con-struction at high altitude, engineers face thechallenge of building a railroad across anunstable landscape. Half of the Qinghai-Xizang railroad’s 1,118 kilometers (695 miles)of track will lie across permafrost areas, andbuilders must take extraordinary measures toprotect each mile of track from permafrostthawing.

“The Qinghai-Xizang railroad is the mostambitious construction project in a permafrostregion since the Trans-Alaska Pipeline,” saidTingjun Zhang, a scientist at the NationalSnow and Ice Data Center in Boulder, Colo.Zhang studies the effects of climate change onpermafrost areas around the world. “Permafrostis thawing in many regions, and this is signifi-cantly influencing landscapes and engineeredstructures,” he said.

Permafrost refers to perennially (year-round)frozen ground that occurs where temperaturesremain below 0 degrees Celsius for two years orlonger, regardless of the rock and soil particlesin the ground. Permafrost regions occupyabout 20 to 25 percent of the world’s land sur-face, and in parts of northern Siberia, per-mafrost can be up to a mile (1,600 meters)thick.

Scientists and engineers charged with monitor-ing permafrost along the Qinghai-Xizang rail-road’s route are primarily concerned with thelayer that lies directly above permafrost,known as the active layer, which freezes and

Riding the Permafrost Express

The Qinghai-Xizang railroad will traverse some of the most desolate landscape on the Tibetan Plateau. This photowas taken south of the village of TuotuoHeyan near Tangulla Pass at an elevation of 6,070 meters.(Image courtesy ofRichard Armstrong)

18

thaws seasonally. Longer periods of seasonalthaw cause the active layer to become evendeeper, which can result in increased thaw set-tlement during the summer and more frostheaving (distortion of the surface) during thewinter.

When buildings, roads, and railroads are builton permafrost with a deep active layer, seasonalchanges in the soil can wreak havoc on thestructures above. As the ground thaws andfreezes, it contracts and expands, putting stresson foundations and twisting rail lines. In arcticareas where structures have been built overpermafrost, insufficient insulation has led tothe collapse of buildings and twisting of railbeds, due to the movement of thawing per-mafrost. For example, in south-central Alaska,thawing permafrost caused the railroad bed ofthe Copper River and Northwestern Railway tosettle unevenly, resulting in a “roller coaster”buckling effect. Although maintenance and useof the railroad were abandoned in 1938, lateraldisplacement continues today.

To prevent structural damage from thawingpermafrost, engineers have developed variousmethods for maintaining stable temperaturesbelow buildings and roads, including paintingroads to increase surface reflectance, elevatingbuildings on pilings above the ground, andusing thermo siphons—metal tubes placedalong roads or around buildings that helpremove heat from the ground. Thermo siphonshelped alleviate permafrost problems along theAlaska pipeline route, but they are costly toinstall and maintain, and they must be placedalong the entire length of a road or railway.

On the Tibetan Plateau, railroad engineershave even more cause for concern than in thepast. When the Qinghai-Xizang railroad wasfirst designed, researchers predicted that theregion’s air temperatures would increase byonly 1 degree Celsius over the next 50 years.Now, scientists believe that the Plateau’s tem-perature may rise by 2.2 to 2.6 degrees Celsiusin that time period, making the rail bed evenmore susceptible to deformation from frostheaving and thaw settlement. Recent studiesshow that the average annual temperature onthe Tibetan Plateau has risen 0.2 to 0.4degrees Celsius since the 1970s, and accordingto the Chinese Academy of Sciences, some per-mafrost areas on the Tibetan Plateau are 5 to 7meters (16 to 23 feet) thinner now than theywere just 20 years ago.

About half the permafrost under the rail bed is“high-temperature” permafrost, which meansthat the frozen soil is only 1 or 2 degreesCelsius below freezing, according to Zhang.“This high-temperature permafrost is very sus-ceptible to thawing,” he said. “And that’s aproblem, because not only is the climate inthat region slowly warming, but the construc-tion and operation of the railroad itself also cre-ates heat.”

But researchers hope to spare the Qinghai-Xizang railroad from the fate of Alaska’sCopper River Railway. Construction engineers

are using several techniques to stabilize thepermafrost below the roadbed and protect therail line from freeze/thaw hazards. Theseinclude re-routing some sections to avoidunstable areas, erecting overpasses across sensi-tive terrain, and building an insulation layerunder the rail bed to maintain permafrost sta-bility.

Researchers from the Chinese Academy ofSciences found that a layer of crushed rock canbe used to insulate the railbed and keep thefoundation stable. “Like thermal siphons, thecrushed rock can both insulate and cool thepermafrost,” said Zhang. Although installationof crushed rock is labor-intensive, its mainte-nance costs are extremely low.

Through a series of experiments, engineersfound that a 1-meter layer of loose rocks mini-mizes the transfer of heat to the soil under rail-road embankments during warmer months.

“The Qinghai-Xizang railroad is the mostambitious construction project in a per-mafrost region since the Trans-AlaskaPipeline.” - Tingjun Zhang

The abandoned Copper River and NorthwesternRailway near Strelna in south-central Alaska illustrateshow building a railroad on permafrost thaw can causethe railroad bed to settle unevenly. Although mainte-nance and use of the railroad was halted in 1938, lateraldisplacement continues today. (Image courtesy of U.S.Geological Survey)

19

The Cold and Arid Regions Environmental andEngineering Institute in Lanzhou, China, testeda crushed rock layer in a section of railroadembankment that overlaid permafrost. Afterone year, the section was significantly colderthan before the installation of the rock layer.“The rock layer is so effective that it actuallyhelped create a cooling effect over time,” saidZhang. Crushed rock insulation was first inves-tigated as early as the 1960s, but this is thefirst time a large-scale project is using the tech-nique as one of its primary solutions, accordingto Zhang.

Despite the cooling effects of the crushed rock,the fragile permafrost along the Qinghai-Xizang railroad must be routinely inspected todetect any frost heaving or thaw settlementbelow the tracks. But the sheer size and inac-cessibility of the Tibetan Plateau make large-scale monitoring a difficult task.

Over the past 10 years, improved satelliteinstrumentation has enabled researchers tomonitor the freeze/thaw cycles that threaten

structures and railroads. “The advantage ofsatellite data is that it enables us to see theentire plateau,” said Zhang. So, instead of justlooking at changes that are occurring in asmall, isolated area, which may or may notapply to the whole plateau region, Zhang andhis colleagues can now create weekly maps thatshow the timing and extent of near-surface soilfreezing and thawing over the entire TibetanPlateau.

Passive microwave instruments, including theSpecial Sensor Microwave Imager (SSM/I) andthe Advanced Microwave Scanning Radio-meter-Earth Observing System (AMSR-E), candetect surface soil freeze or thaw based onbrightness temperatures—a measure of theradiation emitted by an object. The large con-trast between the brightness temperatures ofwater and those of ice makes it possible for sci-entists to distinguish between freezing andthawing conditions.

“Knowing when the freeze cycle starts inautumn and when thawing begins in springhelps us see whether the thaw season is gettingshorter or longer, which can help us predictchanges in the active layer thickness,” saidZhang. Data from both sensors are archivedand distributed by the National Snow and IceData Center (NSIDC) Distributed ActiveArchive Center (DAAC).

“If current observations are indicative of long-term trends, we can anticipate major changesin permafrost conditions during the next centu-ry,” said Zhang.

“Changing trends in the freeze/thaw cycle indi-cate a warming climate,” said RichardArmstrong, a research scientist and one ofZhang’s colleagues at the National Snow andIce Data Center. “And if there’s a warming cli-mate, then that means more days of thaw—which is a concern for the railroad.”

But as three diesel engines pull each trainacross the “Roof of the World,” a combinationof innovative construction techniques andremote-sensing monitoring will ensure asmooth ride when the Qinghai-Xizang railroadopens in 2007.

Reference:Zhang, T., R.G. Barry, and R.L. Armstrong.

2004. Applications of satellite remote sensing techniques to frozen ground studies.Polar Geography 3: 163-196.

For more information, visit the followingweb site:National Snow and Ice Data Center DAAC

http://nsidc.org/

The Qinghai-Xizang railroad will traverse some of themost desolate landscape on the Tibetan Plateau. Thisphoto was taken south of the village of TuotuoHeyannear Tangulla Pass at an elevation of 6,070 meters.(Image courtesy of Richard Armstrong)

Tingjun Zhang is a research scientist at the National Snow and Ice Data Center (NSIDC) inBoulder, Colorado. His research interests include land surface processes in cold regions/cold seasons;snow, ice, permafrost, and seasonally frozen ground; and the application of satellite remote sensingdata to snow, near-surface soil freeze/thaw status, and northern phenomena. Zhang leads the FrozenGround Data Center at the NSIDC in efforts to rescue, archive, and distribute frozen ground-relat-ed data products worldwide. He is also a lead author for the next IPCC Assessment Report, due in2007. Zhang earned a PhD from the Geophysical Institute at the University of Fairbanks, Alaska.

20

Silent SignalsAlaska Satellite Facility DAAC

by Kelly Kennedy

In the Andes Mountains of South America,people as far back as the Incas learned toaccept devastating volcanoes and earthquakesas simply part of life’s lot. Local lore speaks ofthe anger of ancient gods, which, it wasbelieved, often resulted in violent eruptions andground shaking.

A recent incarnation of Mama Pacha, the EarthGoddess of Incan mythology, killed 2,000 peo-ple in Chile after a magnitude 9.5 earthquakelurched through the region in 1960. But geolo-gist Matthew Pritchard sees the swaying andswelling of the Earth as pure science, and hebelieves that science can potentially save hun-dreds of thousands of people.

“Our long-term research goal is to minimizethe number of surprises we get from planetEarth,” said Pritchard, assistant professor ofearth and atmospheric sciences at CornellUniversity. He and his colleagues hope to reacha point where they can determine whether a

volcano or earthquake poses an immediatethreat—even if they can’t predict them.

For now, Pritchard said he’s motivated by twokey findings he uncovered while working onhis PhD dissertation at the California Instituteof Technology (Caltech). First, about 40 fewervolcanoes in the Andes show activity than waspreviously believed. As a result, scientists onthe ground can spend valuable time andresources evaluating just a few volcanoes in theAndes for dangerous activity, rather thandozens.

Second, some earthquakes make slow andsteady progress over a period of months, ratherthan causing one abrupt shake—whatPritchard refers to as “silent earthquakes.” Thismeans that some faults may never actuallytrigger violent temblors. Other earthquakesPritchard observed in satellite imagery did not

send out jarring seismic signals, but slippedsilently down without discernable movementfrom the ground above. “Why do these occur?”he asked. “Why do we get some earthquakesthat are devastating and others that are not?”

Pritchard said his fascination with geologybegan when he was about 10 years old. “I hada rock collection, and I started reading aboutearth science,” he said. Though he grew up ona “flat and featureless” farm in Illinois,Pritchard said his parents took him to theGrand Canyon and Yellowstone NationalParks, where he got a first-hand look at thelandforms he had read about. “I started readingmore about rocks and wondering why theywere different,” he said.

As Pritchard studied his rocks, he began towonder about other worlds, which led him tostudy planetary science. “I became very inter-

Lascar volcano in Chile is the most active volcano in the central Andes, with several eruptions occurring during the1990s. Pritchard and Simons found no evidence of deformation at Lascar during their survey of 900 volcanoes in theregion. (Image courtesy of Mark Simons)

21

ested in why the Earth behaves as it does andwhy we have certain activities here, like earth-quakes and volcanoes, and not on some otherplanets,” he said. But the movements of hisown planet pulled him back down to Earth.“These events have a big impact on humans,and it’s important to understand the way haz-ards affect society,” he said.

Pritchard’s graduate advisor at Caltech, MarkSimons, suggested that he look at someSynthetic Aperture Radar (SAR) data, some ofwhich are archived at NASA’s Alaska SatelliteFacility Distributed Active Archive Center, thatcovered a small area in the Andes. Through aprocess known as SAR interferometry (InSAR),radar satellite instruments shoot beams of radarwaves towards the Earth and record them afterthey bounce back off the Earth’s surface. If thebackscattered signal differs between twoimages of the same object, taken at two differ-ent times, then the object has moved orchanged. Scientists can also tell what materialthe radar wave hits by how much of it isreflected back. For example, water reflects thewave back differently than rock does.

Think about standing on the edge of a ravine.If you toss a string with a rock tied to the bot-tom over the side of the ravine, you can markthe string at the top, pull it back up, and thenmeasure the distance from the mark to the

rock to see how deep the ravine is. And if youtoss the rock from the same spot a week laterand the measurement decreases by a foot, whatdoes that mean? Either your measurements areinaccurate, or the ground moved.

In the Andes, InSAR technology works partic-ularly well because much of the region is aridand sparsely populated. “We needed a studyarea where the ground properties weren’tchanging,” Pritchard said. “There’s no oneplowing the fields, and there’s little vegetationthere. That helps ensure that we get accuratemeasurements.” Pritchard and Simons begantheir study by looking at volcanoes in a smallarea of the Andes, but they quickly realizedhow difficult it would be to survey the areafrom the ground, given its immense size.“Global Positioning System (GPS) surveyscould do the job, but that’s a very labor-inten-sive process,” Pritchard said. “With 900 volca-noes, it could take years to do what we did in acouple of weeks using satellite data.”

Even if they had the time and resources to sur-vey the Andes from the ground, it wouldinvolve great risk. “Fieldwork there is danger-ous, because of the volcanoes themselves andalso because of the need to cross political bor-ders,” said Pritchard.

So, the researchers used satellite imagery to fillin the gaps from areas that could not be meas-ured from the ground. They found that analy-sis of the InSAR data could reveal up to a 2-centimeter deformation (ground movement) atthe volcanoes. They also found that far fewervolcanoes showed activity than they hadexpected.

“Out of about 900 volcanoes, only four showedsigns of deformation or magma movement thatwe didn’t already know about,” Pritchard said.“I think it’s safe to say that we were expectingmore than four volcanoes—maybe 50 had beenconsidered potentially active. So we realizedthat the life cycles of these volcanoes are a littlemore complicated than we thought.” Now,rather than monitoring all 900 volcanoes in theAndes, Pritchard and his colleagues knowwhich ones they need to watch closely.

Still, Pritchard said, it’s important to remem-ber that volcanic eruptions and earthquakescan’t be predicted. “Volcanoes have very differ-ent personalities,” he said. “Sometimes they dothings we expect, but sometimes they surpriseus. Sometimes magma moves underneath thevolcano, but it doesn’t necessarily lead to aneruption.”

Knowing precisely which volcanoes to monitorcan help researchers concentrate their groundstudies in the areas at most risk. An example ofhow effective monitoring can save lives came inJune 1991 when seismic data revealed thatearthquakes were occurring at Mount Pinatubo

This image shows a tsunami warning sign in the city ofAntofagasta in northern Chile, which was shaken by amagnitude 8.0 earthquake on July 30, 1995. (Imagecourtesy of Mark Simons)

“With a large earthquake, it’s no mysterythat a fault slip occurred. But exactly whicharea slipped and where another slip mightoccur in the future aren’t as apparent.” - Matthew Pritchard

22

in the Philippines. The Philippine Air Force’sAir Base command notified local towns thatthe volcano might erupt, and 60,000 peoplewere evacuated. Officials estimate that 20,000people would have died if the eruption alerthad not been issued.

For the past year, Pritchard has applied InSARdata to earthquakes with the same strikingresults he observed with the volcanoes. “With alarge earthquake—a magnitude 7 or 8—it’s nomystery that a fault slip occurred,” he said.“But exactly which area slipped and whereanother slip might occur in the future aren’t asapparent.” And sometimes, the earth slipsslowly with no quake at all, triggeringPritchard’s newest questions: “Can a silentearthquake accelerate into a violent earth-quake?”

In the Pacific Northwest, GPS recordings showthat a silent earthquake occurs beneathWashington State and British Columbia aboutonce every 14 months, but no one on theground even notices it happening. Pritchard is

looking for the same phenomena in the Andeswhere, unfortunately, there are few continuous-ly operating GPS stations.

Since silent earthquakes may take days, weeks,or even months to occur, they don’t send outseismic waves, Pritchard said. So, he watchesthem to try to learn how to assess potentialhazards. “There appear to be some faults thatnever produce large earthquakes,” he said.“And maybe there are also a few silent earth-quakes that never evolve into a major event.”

Pritchard and his colleagues sometimes jokeabout things seen in satellite images that don’tactually appear on ground, like “the lost city ofAtlantis.” Some have suggested that the conti-nent disappeared into Lake Titicaca in Bolivia,based on descriptions by Plato. But forPritchard, the movements of the Earth holdplenty of fascination without lost cities.

“Some people claim to see evidence of Atlantisin satellite data,” said Pritchard. “But if theysaw the same area from the ground, they’drealize it’s not Atlantis.” But he does agree thatobservations from space often lead to discover-ies that could take years to find from theground alone, and he believes it’s important touse both ground and remote sensing tech-niques.

“In satellite imagery, we get a bird’s-eye viewof where the deformation is happening,” hesaid. “We spend most of our time looking atthose images and using models to interpret thedata, but we still like to get out in the field tosee what’s happening from the ground.”

References:Pritchard, Matthew E. and Simons, M. 2004.

An InSAR-based survey of volcanic deformation in the southern Andes. Geophysical Research Letters 31, L15610, doi:10.1029/2004GL020545.

Pritchard, Matthew E. and Simons, M. 2004. Surveying volcanic arcs with satellite radar interferometry: the central Andes, Kamchatka, and beyond. GSA Today 14 (8),4-11.

For more information, visit the followingweb sites:Alaska Satellite Facility DAAC

http://www.asf.alaska.edu/U.S. Geological Survey Earthquake Hazards

Programhttp://earthquake.usgs.gov/

U.S. Geological Survey Volcano Hazards Programhttp://volcanoes.usgs.gov/

Considerable damage to quality, wood-frame houses inValdivia, Chile occurred during the 1960 earthquake thatshook southern Chile. Valdivia suffered catastrophicdamage because of its proximity to the epicenter of themassive quake. (Image courtesy of NOAA, Pierre St.Armand, photographer)

Matthew Pritchard is an assistant professor of earth and atmospheric sciences at Cornell University.He earned his PhD in geophysics from California Institute of Technology (CalTech) in 2003, wherehe worked as an assistant scientist at the Caltech Seismological Laboratory. His research interestsinclude earthquakes, volcanoes, crustal deformation, subduction zones, and terrestrial planets.Pritchard completed internships at the Lunar and Planetary Institute in Houston, and at the NASAPlanetary Geology and Geophysics Undergraduate Research Program.

Supplemental Data Image Section

23

Above, left: These six images, acquired by the TOPEX/Poseidon satellite, show seasurface height anomalies for six different years. Each image represents a 365-dayaverage of that year’s data, with a "Normal Year" defined as the average from 1992to 2002 removed from the individual years. The year-to-year differences in these siximages are called anomalies, or residuals. When oceanographers and climatologistsview these anomalies, they can identify unusual patterns and estimate how heat isbeing stored in the ocean during a particular year relative to previous and future years(i.e., how the ocean climate is slowly evolving to influence the next year's planetaryclimate events). These year-to-year, and even decade-to-decade, changes in theocean indicate climate events such as the El Niño, La Niña, and Pacific DecadalOscillation. For oceanographers and climatologists, these sea-surface height imagesare one of the most modern and powerful tools for taking the "pulse" of the globaloceans. (Image courtesy of Lee-Lueng Fu)

(See Making Waves in Tsunami Research, page 2)

Above, right: This SeaWiFS image shows a spring phytoplankton (chlorophyll) bloomover the Gulf of Alaska. Red represents high chlorophyll levels and blue represents lowconcentrations. The timing and location of these blooms are crucial to fish and birdpopulations. (Image courtesy of Scott Pegau, Kachemak Bay Research Reserve)

(See Bloom or Bust: The Bond Between Fish and Phytoplankton, page 10)

24

Above, left: This composite SAR interferometry (InSAR) image shows color contours ofground deformation draped over shaded relief from three shallow thrust subductionzone earthquakes along the coast of Chile (dates and magnitudes of the earthquakesare labeled). Each contour represents 10 centimeters of deformation. Because theEuropean Resource Satellite-1 (ERS-1) satellites primarily measure vertical deforma-tion, the gross features can be interpreted as portions of the ground that were upliftedor subsided. For the 1995 earthquake, only part of the dry land was uplifted (thesouthwest corner of the Mejillones Peninsula), and the closed contours in the interfer-ogram are mostly caused by the on-land subsidence. For the 1996 earthquake, theslip was closer to land, so more uplift was recorded on-shore, but the closed contoursrepresent subsidence. Most of the fault slip from the 2001 earthquake was off-shore,so only subsidence is measured on land. The reference map in the lower left placesthe region in context on the continent of South America. (Image courtesy of MatthewPritchard)

(See Silent Signals, page 20)

Above, right: This satellite image, acquired by the Moderate Resolution ImagingSpectroradiometer (MODIS) on July 29, 2004, shows smoke from hundreds of smallfires drifting across the rainforest in Mato Grosso, Brazil. Colonists use fire to clear for-est for farming and cattle pastures. However, several years of intensive agriculturedepletes the soil of nutrients. Instead of reinvesting in the depleted land, farmers willoften just clear a new piece of land, creating a cycle of land use destructive to therainforest. (Image courtesy of MODIS Rapid Response Project at NASA/GSFC)

(See A Rainforest Divided, page 27)

25

Above, left and center images: These images, acquired by the Thermal Emission ImagingSystem (THEMIS) on the Mars Odyssey spacecraft, show two representations of thesame small crater on the surface of Mars. The crater, located in the Syrtis Major region,is slightly larger than Meteor Crater in Arizona and has been provisionally namedWinslow Crater (after the small town in Arizona near Meteor Crater). On the left side is agrayscale image taken in the visible wavelengths (18 meters per pixel) that shows eject-ed blocks, the outer crater rim, inner crater wall, and ejecta blanket. On the right side isa false-color composite made from three individual THEMIS thermal infrared (TIR) bands(100 meters per pixel). The false-color image was colorized using a technique calleddecorrelation stretch (DCS), which emphasizes the spectral differences between thebands to highlight compositional variations in the surface materials. The left image is at ascale of 31 kilometers (19.2 miles), and the right image is at a scale of 7.5 kilometers(0.75 miles) in diameter.

Above, right: This image was acquired by the Advanced Spaceborne ThermalEmission and Reflection Radiometer (ASTER) in June, 2004, and shows MeteorCrater in north-central Arizona. It is a color composite of the visible near-infrared(VNIR) bands (15 meters per pixel) that highlights the flat-lying sedimentary rockunits surrounding the crater in shades of gray and green. The ejecta, reworked bywind and blown to the northeast, appears as brighter white areas.

Both the ASTER and THEMIS instruments have very similar spatial and spectralresolutions in the VNIR and TIR. Therefore, data from ASTER can be used as anideal analog for THEMIS data in studies of certain surface features, such ascraters. (THEMIS images courtesy of NASA/JPL/Arizona State University; ASTERimage courtesy of NASA/GSFC/METI/ERSDAC/JAROS and U.S./Japan ASTERScience Team).

Right: This map, created using data from the Gravity Recovery and Climate Experiment (GRACE) mission, revealsvariations in the Earth's gravity field. Dark blue areas show areas with lower than normal gravity, such as the IndianOcean (far right of image) and the Congo river basin in Africa. Dark red areas indicate areas with higher than normalgravity. The long red bump protruding from the lower left side of the image indicates the Andes Mountains in SouthAmerica, while the red bump on the upper right side of the image indicates the Himalayan mountains in Asia. (Imageprepared by The University of Texas Center for Space Research as part of a collaborative data analysis effort withthe NASA Jet Propulsion Laboratory and the GeoForschungsZentrum in Potsdam, Germany)

(See Matter in Motion: Earth’s Changing Gravity, page 35)

(See In Search of Martian Craters, page 39)

26

Above, left: This map indicates urban areas on a global scale. The call-outs indicatethe overlay of urban areas with various ecosystems. Note that ecosystem boundariesmay often distinctly omit core parts of urban areas, for example, the cultivatedecosystem call-out. The core part of the urban area of Buenos Aires, for example, isblack, indicating that it is not part of the cultivated ecosystem, but the peri-urban andsurrounding smaller settlements (in gray) are contained within that ecosystem. In con-trast, the inland water system does not exclude urban areas.

Though drylands are the predominant ecosystem, and coastal zones the smallest,coastal areas are disproportionately urban, with about 65% of the population residingin urban areas and occupying about 10% of the coast land area (compared to 47 and3%, respectively, for global averages). Further, even in rural regions of both coastaland cultivated areas, population is disproportionately dense. Understanding how pop-ulation varies by ecosystem in both rural and urban areas is an important step in pro-viding services to populations in a sustainable manner. (Image courtesy ofSocioeconomic Data and Applications Center [SEDAC])

(See Checking Earth’s Vital Signs, page 44)

Above, right: NASA’s Short-term Prediction Research and Transition (SPoRT)Center collaborates with the National Weather Service to provide data fromNASA’s Earth Observing System satellites. This example is from the MODISinstrument on Aqua, collected on July 18, 2005 over Alabama andTennessee. The large upper left image reveals an intense thunderstorm overHuntsville, Alabama, and shows how MODIS data are displayed through theNational Weather Service’s Advanced Weather Information ProcessingSystem (AWIPS). The smaller color insert pictures show MODIS data prod-ucts that indicate regions favorable for thunderstorm development and pos-sible severe storms (indicated by the dotted oval area in the upper rightimage). In a mere 30 minutes, this storm produced more than 6.35 centime-ters (2.5 inches) of rain over a portion of Huntsville. (Image courtesy of GaryJedlovec)

(See SPoRTscast, page 48)

27

If a tree falls in the Amazon rainforest, biolo-gist William Laurance just might hear it. Andas human activity encroaches on the forests,Laurance is finding that a lot of trees arefalling.

As people penetrate the Amazon interior, theybuild roads and clear large expanses of land.Over time, this process of deforestation createsforest fragments, islands of trees surrounded bya sea of pastures, farmland, and other degradedhabitats. Laurance and his colleagues at theSmithsonian Tropical Research Institute arestudying what happens to the forests whentrees lose ground to agriculture, logging, androads. By tracking the results of forest frag-mentation, the researchers hope to assess theimpacts of land-use change on forest dynamicsand study how ecosystems respond.

To support Amazon rainforest research, theBrazilian government launched an internation-al effort in 1993, called the Large ScaleBiosphere-Atmosphere Experiment inAmazonia (LBA). LBA project investigatorsstudy a wide range of Amazon ecosystemdynamics, such as the movements of nutrientsthrough the ecosystem, the chemistry of theatmosphere among and above the canopy, andchanges in the sources and sinks of carbon. Laurance is an investigator for a com-ponent of the LBA Project, called LBA-ECO,which focuses on the effects of land usechanges in the Amazon.

A RainforestDivided

In fragmented forests, old-growth trees (shown inimage) are replaced by less dense, faster-growing trees,which tend to store less carbon. (Image fromPhotos.com)

Oak Ridge National Laboratory DAACby Laura Naranjo

In the Amazon rainforest near Manaus, Brazil,Laurance’s team monitored both fragmentedand pristine portions of a 1,000-square-kilome-ter area that was sectioned into plots. Fieldtechnicians collected data for each plot at leastfive times over a 20-year period, counting, clas-sifying, and measuring the diameters of all liv-ing and dead trees each time. Laurance’s col-leagues at the Biological Dynamics of ForestFragments Project inventoried animal, bird,and insect populations before and after frag-mentation.

Their long-term study was unusual in that theycarefully surveyed most of the plants and ani-mals in each of the fragments before clear-cut-ting isolated the fragments from the continu-ous forest. “That was the big advantage of ourstudy. We built up a very good picture of whatspecies existed before the habitat was frag-mented,” said Laurance. This picture provideda basis against which to assess changes in theforest ecosystem in its fragmented state.

Laurance admits that his study represents abest-case scenario, since a presidential decreeprotects the study area from the damagingactivities that most other fragmented rain-forests endure: logging, hunting, and fire-basedagriculture. “These four processes together canbe pretty devastating for some species, and cer-tainly for the forest ecosystem as a whole,” hesaid.

But even in the protected study area, the teamfound that forest dynamics in fragments werechanging dramatically. Trees were dying alongthe forest edges, and the biggest trees—withtrunks greater than 60 centimeters (24 inches)

in diameter—were dying the fastest. As thelarger trees died off around the edges of frag-ments, the remaining trees were more exposedto temperature changes and dry, hot winds thatleft them vulnerable to drought. These “edgeeffects” impacted not just the forest edges, butpenetrated as far as 300 meters (nearly 1,000feet) into the forest.

Aside from providing habitat for rainforestflora and fauna, the Amazon plays an impor-tant role in storing and removing carbon diox-ide from the atmosphere. Carbon dioxide is agas that contributes to the Earth’s greenhouseeffect, in which atmospheric gases trap thermalenergy and cause surface and air temperatureto rise. Adding more carbon dioxide to theatmosphere magnifies this greenhouse effect.

The Amazon rainforest covers an area morethan seven times the size of Texas, populatedwith nearly half a billion hectares (1.2 billionacres) of trees and vegetation that are critical tothe Earth’s carbon cycle. However, high ratesof deforestation in the Amazon (reaching 2.4million hectares, or nearly 6 million acres, peryear in 2002 and 2003) are diminishing therainforest’s ability to store carbon.

Each hectare of destroyed forest releases around200 tons of carbon into the atmosphere—wors-ening, rather than slowing, the greenhouseeffect, Laurance explained. The extreme mor-tality of trees in forest fragments means thatthe fragments are losing considerable amountsof carbon, perhaps as much as 150 million tonsper year in the world’s tropical regions.

In the fragments Laurance studied, less dense,faster-growing trees replaced old-growth trees,and lianas (woody vines) proliferated aroundforest edges. Both lianas and faster-growingtrees store less carbon than large, dense trees.“The denser the wood, the more carbon a treecan store. A lot of the new trees coming intend to be lighter wooded pioneer species thatstore less carbon,” said Laurance. “So fragmen-tation is reducing the amount of stored carbonin two ways. There are fewer trees and smallertrees, and the composition of the forest ischanging.”

Besides impacting tree populations, fragmenta-tion alters habitat vital to rainforest animals,particularly in the forest understory—the smalltrees and vines between the forest floor and theupper canopy. Protected from wind and rain,the shaded understory provides a dark, cool,humid environment for a variety of specializedmammals, birds, frogs, and insects. “A lot ofthe understory birds are sensitive to fragmenta-tion. They’re adapted to deep, dark forest con-

28

Birds and other animals are often specially adapted toliving in the cool, dark habitat below the thick rainforestcanopy. (Image courtesy of William Laurance)

29

ditions, and they won’t come out anywherenear a road or clearing,” said Laurance.

Because the rainforest ecosystem supports com-plex webs of interactions, when extinctionclaims one animal, plant, or insect, other crea-tures suffer as well. For instance, explainedLaurance, a number of frog species use the lit-tle wallow ponds made by peccaries (wild pigs).The peccaries need large areas for foraging, sowhen fragmentation reduces their range anddrives them out of the ecosystem, frogs in thearea start disappearing.

Fragmented forests also isolate animal popula-tions, which can be a driving force behindextinction. Animals such as jaguars and pumasrequire large areas, and fragmentation canlimit their ability to hunt or force them deeperinto the forest. In addition, fragmentationincreases human access to the forests, attract-ing hunters who kill jaguars, monkeys, deer,and agoutis (giant rodents), Laurance said.

Laurance cited road building as one of themain culprits behind forest loss and fragmenta-tion. The Brazilian government currently plansto pave about 7,500 kilometers (about 4,600miles) of new highways that will provide moreaccess to the Amazon interior, and Laurance’sresearch indicates that this will increase frag-mentation of the rainforest.

Fragmentation begins when a paved highwaypenetrates undisturbed forest. Loggers andcolonists then build numerous roads branchingoff of the highway, creating a pattern of landuse that fragments long strips of forest. “Onceyou get a major highway providing year-roundaccess, then all kinds of things happen—colo-nization, logging, and a lot more road develop-ment,” Laurance said. And once the colonistsreach their pristine destination, fire-based agri-culture then poses a problem for the Amazon’sremaining forests. Colonists hoping to tame apiece of the rainforest often clear the land foragriculture with slash-and-burn techniques.

Laurance also blames “surface fires,” annualfires that farmers light in their pastures todestroy weeds and produce fresh grass. Thesefires burn not only the pastures, but oftenspread into the surrounding forest as well.

“Because fire is foreign to the rainforest ecosys-tem, the plants there are just not adapted to it.Even a little fire that’s only 10 or 20 centime-ters (4 to 8 inches) high will slowly burnthrough the forest understory, killing 20 to 50percent of the trees and all of the vines,” saidLaurance. Dead trees and vines then accumu-late on the forest floor, becoming tinder thatfuels the following year’s fire.

“Every time there’s a new fire, it creates moreflammable material. So at the end of the cyclethere is just a scorched landscape of completelydestroyed forest,” Laurance said. “You can liter-ally see the fragments imploding over time.”The problem worsens when El Niño yearsbring drought to the region, making theforests even more vulnerable to fire.

Data from Laurance’s study, along with a vari-ety of other data sets generated by the LBAproject, will be archived and available from theLBA web site in Brazil and the Oak RidgeNational Laboratory Distributed ActiveArchive Center. As scientists gather more data,they will gain a clearer understanding of howland use change affects the Amazon and itsability to store carbon.

Laurance hopes that his research will ultimatelypromote policy change. The Brazilian govern-ment is trying to develop sustainable economicstrategies for the rainforest, but land in therainforest is so cheap that once a farmer’s soil isdepleted, he can simply purchase another patchof forest to clear and burn. “Rather than rely-ing on very destructive fire-based agriculture,”Laurance said, “we’re advocating a system inwhich there are more incentives for farmers toinvest in their land and develop more sustain-able kinds of agricultural strategies.” By mak-ing currently cleared land more productive, hesuggests, farmers could work closer to existing

“We’re advocating a system in which thereare more incentives for farmers to invest intheir land and develop more sustainablekinds of agricultural strategies.” - William Laurance

This image shows the rainforest canopy north ofManaus, Brazil. (Image courtesy of NASA LBA-ECOProject)

30

markets, and the government could then setaside larger tracts of pristine rainforest, pre-serving the rich flora and fauna of the Amazon.

References:Pimm, Stuart L. 1998. The forest fragment

classic. Nature 393: 23-24.Laurance, W.F., P. Delamônica, S.G. Laurance,

H.L. Vasconcelos, and T.E. Lovejoy. 2000. Rainforest fragmentation kills big trees. Nature 404: 836.

Laurance, W.F., S.G. Laurance, L.V. Ferreira, J.M. Rankin-de Merona, C. Gascon, and T.E. Lovejoy. 1997. Biomass collapse in Amazonian forest fragments. Science278:117-118.

Laurance, W.F., M.A. Cochrane, S. Bergen, P.M. Fearnside, P. Delômonica, C. Barber, S.D’Angelo, and T. Fernandes. 2001. The future of the Brazilian Amazon. Science291(5503): 438-439.

William F. Laurance is a staff scientist at the SmithsonianTropical Research Institute andpresident of the Association forTropical Biology and Conservation.His research interests includeassessing the impacts of land-usechanges (such as habitat fragmenta-

tion, logging, and fires) on tropical forests, and global-change phenomena. He works in the Amazon, centralAfrica, and Australasia, and holds a PhD in integrativebiology from the University of California, Berkeley.

For more information, visit the followingweb sites:Large Scale Biosphere-Atmosphere Experiment

in Amazoniahttp://lba.inpa.gov.br/lba/

Large Scale Biosphere-Atmosphere Experiment in Amazonia ECOhttp://www.lbaeco.org/lbaeco/

An Introduction to the Large Scale Biosphere-Atmosphere Experiment in Amazoniahttp://earthobservatory.nasa.gov/Study/LBA/

Smithsonian Tropical Research Institutehttp://www.stri.org/index.php

Oak Ridge National Laboratory Distributed Active Archive Centerhttp://www.daac.ornl.gov/

Stealing Rain from the Rainforesthttp://earthobservatory.nasa.gov/Study/AmazonDrought/

From Forest to Fieldhttp://earthobservatory.nasa.gov/Study/AmazonFire/

The Road to Recoveryhttp://earthobservatory.nasa.gov/Study/recovery/

Deforestation in the Amazon rainforest threatens many species of tree frogs, which are extremely sensitive to envi-ronmental changes. (Image from Photos.com)

31

War onHungerSocioeconomic Data and ApplicationsCenter (SEDAC)

by Amanda Leigh Haag

In December 2004, the eyes of the worldturned toward the tsunami-ravaged coasts ofAsia and Eastern Africa. Donations, foreignaid, and emergency health care from all nationspoured in to help deal with the loss of morethan 200,000 victims and the many homelessand destitute people that the Indian Oceantsunami left behind.

But each month, 200,000 Africans die fromwhat Pedro Sanchez calls “a silent tsunami,”the tide of chronic hunger and malnutritionthat sweeps across the African continent. Noone is immune: it reaches coastal fishing towns,rural villages, and urban centers alike.

Sanchez, co-chair of the United Nations (UN)Millennium Project’s Hunger Task Force, isworking to change the way that most peoplethink of world hunger. Once the public begins

to realize that the vast majority of hungry peo-ple around the world are not starving, but areinstead chronically hungry, the course of inter-vention will change, Sanchez believes.

Acute hunger resulting from famines, wars,and natural disasters represents only a smallfraction of the hungry but receives most of themedia attention and coverage, according toSanchez, who received the 2002 World FoodPrize. “The stereotypical image we have of anEthiopian child with flies in his eyes dying insome sort of refugee camp is not representative

of the situation faced by over 90 percent ofhungry people,” he said. “Starvation onlyaccounts for about 8 percent of the hungrypeople in the world; the other 92 percent sufferfrom hunger silently, and they die in drovesbecause of malnutrition-related diseases.”

In the summer of 2002, UN Secretary GeneralKofi Annan asked the director of the EarthInstitute at Columbia University, Jeffrey Sachs,to lead the “UN Millennium Project.” Its pur-pose was to develop a set of practical strategiesto achieve the “Millennium Development

Students line up for school meals, made with locally produced foods, at Bar Sauri Primary School in the MillenniumVillage at Sauri, Kenya. (Image courtesy of Pedro Sanchez, Earth Institute at Columbia University)

32

Goals” (MDGs), targets set by the world com-munity in the year 2000 for reducing povertyand promoting sustainable development. Theproject’s 10 task forces dealt with the roots ofpoverty and hunger, including health, waterand sanitation, gender, education, and environ-ment. The Hunger Task Force became theteam specifically assigned the goal of halvingworld hunger by 2015 and, eventually, elimi-nating it.

Through the Hunger Task Force, Sanchez andother experts identified hunger “hot spots” toget a better sense of where poor and hungrypeople are in the world. Scientists at ColumbiaUniversity’s Center for International EarthScience Information Network (CIESIN) under-took a geospatial analysis that stitched togethernational census data from maternal and childhealth surveys at the sub-national level. Thetask force found 313 provinces or districts insub-Saharan Africa where more than 20 per-cent of children under five years old are under-weight. Their analysis showed that 80 percentof hungry people in the area they studied livein the hot spot regions.

Next, the team set out to determine the differ-ent causes of hunger in various regions and topinpoint the geographical distribution of mal-nutrition. Using data from the SocioeconomicData and Applications Center (SEDAC), such

as the “Gridded Population of the World”(GPW) data set, they found that factors suchas climatic conditions, geographic remoteness,and elevation were statistically related to thedistribution of child malnutrition and infantmortality.

“You’re more likely to be poor if you’re locatedfar away from the coast or from a road,” saidMarc Levy, one of SEDAC’s project scientistsand an associate director with CIESIN. “You’realso more likely to be poor if you live in aregion that has little rainfall or a short growingseason, if you’re far from an urban center, or ifyou live at a high elevation,” said Levy. All ofthese variables assist in diagnosing the causesof poverty, he said.

The precise combination of interventions toaddress poverty depends greatly on the geo-graphical and physical factors at play in a givenarea. In sub-Saharan Africa, for example, agrowing body of evidence shows that lack ofaccess to transportation networks createshunger “traps,” according to Levy. If someoneis far from a main highway, it takes more effortto get out of a poverty trap, so investing inbetter transportation infrastructure would be akey strategy, he said.

For populations that live in higher-elevationregions, other factors such as soil or growingconditions might need to be addressed first.Yet, at the same time, these populations havethe benefit of reduced exposure to diseases likemalaria. But the net effect of all these factorscan only be understood when you have the rel-evant data at your disposal, said Levy.

Levy noted an emerging trend: countries thatcreate detailed poverty maps are doing a betterjob of charting intelligent interventions, sincethey are able to identify their own hot spotsand set priorities where the greatest attentionis needed. “This is sort of ‘brave new world’territory,” said Levy. “It’s the first time inhuman history that people have started talkingabout global poverty as seriously as they’re cur-rently talking about it.”

But another darker trend is also emerging:even when there is a plan in place to reducepoverty, many countries do not have the finan-cial resources to meet their goals. And foreignaid to fill that gap is lagging far behind; insome cases, financial commitments fromwealthy, developed countries are disproportion-ately low. In other cases, even where sizeablecommitments have been made, the money hasnot been coming in quickly enough, said GlennDenning, director of the MDG Center based inNairobi, Kenya. “So the money hasn’t been

“The stereotypical image we have of anEthiopian child with flies in his eyes dying insome sort of refugee camp is not representa-tive of the situation faced by over 90 percentof hungry people.” - Pedro Sanchez

A protected spring provides a clean water supply to res-idents in Sauri, Kenya. (Image courtesy of GlennDenning)

33

flowing at the speed it should be, and certainlynot at the speed that is needed to achieve thegoals of the Millennium Project,” saidDenning.

In 2004, the Millennium team set up theMDG Center in Nairobi to work specificallywith individual countries and to translate theMillennium Project recommendations intofinancing plans at the country level. Theirapproach is to work intensively with countryleaders, analyze the programs each nation hasin place and the means that it has to achieve itsobjectives, and assess whether those objectivesline up with the goals of the MillenniumProject. Next, they set up “MDG-based pover-ty reduction strategies,” which serve as a blue-print for how each country can address povertyand hunger.

So far, in every country the team has studied,including Kenya, Ethiopia, Tanzania, Ghana,Senegal, and others, progress is not happeningat the pace needed to reduce hunger or cut it

in half by 2015, said Denning. “Our assess-ment is that they’re lagging behind in terms ofprogress and level of investment that they’reputting into cutting hunger,” said Denning.“It’s simply not enough. The developed coun-tries need to fill that gap by increasing theiraid contributions to the countries that haveplans in place but do not have the resources toactually implement them.”

Average aid contribution from developed coun-tries is about 0.25 percent of gross nationalproduct. The United States falls far below theaverage, at 0.16 percent, although it has com-mitted itself to giving much more, saidDenning. Australia, Japan, and Italy also fallfar below average in contributions. “A numberof developed countries have expressed in princi-ple that they’re committed to delivering ontheir long-held promise of reaching 0.7 per-cent,” said Denning, “but their actual contri-butions remain appallingly small.”

One major issue with foreign aid is that somecountries do not have a sufficient plan in placeto ensure that international aid is used effec-tively. And potential donor countries “remainconcerned about corruption and inefficiency inusing aid,” said Denning. “The developedcountries might even say, ‘yes, we know there’sa need -- but show us an accountability systemwhereby most of the money really gets to thepeople who need it,’” said Denning.

Typically, it takes about 18 months from thetime that a needs assessment is done to thetime that a country has a full-blown povertyreduction strategy in place. But in the mean-time, the MDG Center has designed “quick

wins”—swift interventions that have immedi-ate or short-term payoffs. For example, onequick win strategy is to dispense anti-malarialmosquito nets. Since malaria is the number onekiller among African children, mosquito netsbegin cutting down on infirmities and mortali-ty from malaria immediately.

The MDG Center also advocates programs thatenable primary school children to have a nutri-tious noontime meal every day, using locallyproduced foods instead of food aid. This hasmultiple benefits, including increased atten-dance and improved quality of learning. Butcentral to this strategy is that the food be pro-cured locally, “not imported from Iowa,Europe, or Australia,” said Denning. Usinglocal foods creates a demand in the communityfor farmers to produce more.

Many experts in the Millennium Project con-sider agriculture to be the “engine of growth”and are promoting the central message that

Pedro Sanchez and Awash Teklehaimanot, an EarthInstitute malaria expert, learn soil science in the NyandoDistrict of Western Kenya. (Image courtesy of PedroSanchez, Earth Institute at Columbia University)

Local children sell charcoal in Kisumu, Kenya.(Imagecourtesy of Pedro Sanchez, Earth Institute at ColumbiaUniversity)

34

foreign aid organizations should facilitate localfood production, rather than encourage adependency on imported food. “It goes back tothe old Chinese proverb that if you give peoplefish, they will eat for a day, but if you teachthem how to fish, they’ll eat for a lifetime,”said Sanchez. “Imported food aid createsdependencies and sometimes really stunts localmarkets.”

Experts in the Millennium Project agree thatthe solutions to world hunger are not far out ofreach. “So much of what is needed in the poor-est parts of the world, especially in Africa, liesin information that already exists,” saidDenning. “The cost of ending poverty is verysmall compared to what people in the devel-oped world spend on their Starbucks coffees.”

Reference:Sanchez, Pedro A., and M.S. Swaminathan.

2005. Cutting world hunger in half. Science307(5708): 357-359.

For more information, visit the followingweb sites:UN Millennium Development Goals

http://www.un.org/millenniumgoals/UN Millennium Project

http://www.unmillenniumproject.org/The Earth Institute at Columbia University

http://www.earthinstitute.columbia.edu/Center for International Earth Science

Information Network (CIESIN)http://www.ciesin.columbia.edu/

Socioeconomic Data and Applications Center (SEDAC)http://sedac.ciesin.columbia.edu/index.html

In Africa, dairy cows provide income and a way out ofpoverty for many rural villagers. (Image courtesy ofGlenn Denning)

Pedro Sanchez is Director of Tropical Agriculture and Senior Research Scholar at the Earth Instituteat Columbia University. He received the World Food Prize in 2002, a MacArthur Fellowship in2004, and serves as co-chair of the Hunger Task Force of the United Nations Millennium Project.Sanchez is author of Properties and Management of Soils of the Tropics, rated among the top 10 best-sell-ing books on soil science worldwide. A native of Cuba, Sanchez received his PhD from CornellUniversity. He has dedicated his professional career to improving tropical soils management throughintegrated natural resource management approaches in an effort to achieve food security and reducerural poverty while protecting and enhancing the environment.

Marc Levy heads the Science Applications group at the Center for International Earth ScienceInformation Network (CIESIN), part of the Earth Institute at Columbia University. He also leadsCIESIN’s work on the Environmental Sustainability Index, Poverty Mapping, and the HumanFootprint; serves as a project scientist for the Socioeconomic Data and Applications Center; coordi-nates CIESIN’s involvement with the Millennium Development Project, and directs work on con-flict early warning. Levy is a Convening Lead Author for the Millennium Ecosystem Assessment.

Glenn Denning joined the Earth Institute at Columbia University in July 2004 and serves as associ-ate director of the Tropical Agriculture Program. He has played a lead role in establishing TheMillennium Development Goals (MDG) Centre in Nairobi, where he now serves as director. TheCentre works with governments, United Nations agencies, and the Millennium Project to strength-en policies, plans, and investments aimed at achieving the MDGs.

35

Matter in Motion:

Physical Oceanography DAAC

by Laura Naranjo

According to legend, Isaac Newton discovered gravity after watching anapple fall from a tree. Using the word “gravitas” (Latin for “weight”), hedescribed the fundamental force that keeps objects anchored to theEarth. Since then, scientists have used maps of the Earth’s gravity todesign drainage systems, lay out road networks, and survey land sur-faces. But Newton probably didn’t imagine that gravity could reveal newinformation about the global hydrology cycle.

Traditionally, scientists constructed gravity maps using a combination ofland measurements, ship records, and more recently, remote sensing.However, those measurements weren’t accurate enough to capture theslight changes in water movement that cause gravity to vary over time.With the help of a new satellite mission, scientists can now weigh wateras it circulates around the globe and relate these measurements tochanges in sea level, soil moisture, and ice sheets.

To better assess these gravity variations, an international team of engi-neers and scientists developed the Gravity Recovery and ClimateExperiment (GRACE) mission. Launched in March 2002 as a joint ven-ture between NASA and the Deutsches Zentrum für Luft undRaumfahrt (German Aerospace Center), the mission was implementedthrough collaboration between the University of Texas Center for SpaceResearch, the GeoforschungZentrum (Germany’s National ResearchCentre for Geosciences), and the NASA Jet Propulsion Laboratory (JPL).

GRACE relies on two identical satellites, each about the size of a smallcar. As the satellites fly approximately 220 kilometers (137 miles) apart,one following the other, a microwave ranging system monitors the dis-tance between them to within a micron—smaller than a red blood cell.Scientists can map gravity anywhere on the Earth’s surface by measuringtiny changes in distance between the two satellites as each of themspeeds up and slows down in response to gravitational force.

Archived at NASA’s Physical Oceanography Distributed Active ArchiveCenter (PO.DAAC) in Pasadena, California, and the GeoForschung-Zentrum Information System and Data Center (GFZ/ISDC), GRACEdata are changing the way scientists and modelers view gravity. GRACE

By measuring changes in the distance between the GRACE mission’s lead satelliteand trailing satellite, scientists can determine changes in the Earth’s gravity. (Imagecourtesy of NASA Jet Propulsion Laboratory)

Earth’s Changing Gravity

36

provides monthly maps that are at least 100times more accurate than previous maps atdetailing changes in the Earth’s gravity field.“The classic idea of gravity being somethingthat you measure once is no longer accepted.Gravity is an element that scientists must con-tinue to monitor,” said Byron Tapley, directorof the Center for Space Research and principalinvestigator for the GRACE mission.

Because scientists can’t see, feel, or directlyobserve gravitational forces, they map theEarth’s gravity using a mathematical modelthat describes an imaginary spherical surfacecalled the geoid. The geoid represents oceans assmooth, continuous surfaces unaffected bytides, winds, or currents. It creates a locallyhorizontal surface against which scientists canmeasure the downward pull of gravity.

Gravity is determined by how much mass agiven material has, so the more mass an objecthas, the stronger its gravitational pull. Forexample, granite is a very dense material with ahigh level of mass, so it will exert a greater pullthan the same volume of a less dense material,such as water. Earth’s mass is distributedbetween various landforms and features—suchas mountain ranges, oceans, and deep seatrenches—that all have different mass, whichcreates an uneven gravity field.

Consequently, the geoid doesn’t form a perfectsphere, and in maps based on the geoid, theEarth’s gravity field exhibits bulges and depres-sions. “Because the distribution of materialsdeep inside the Earth varies, its gravity fieldhas hills and valleys. The ocean tries to layalong that hilly surface,” said Michael Watkins,GRACE project scientist at JPL. For instance,the ocean surface off the tip of India is about200 meters (650 feet) closer to the Earth’s corethan the ocean surface near Borneo. Withouttides, currents, and wind, the ocean surfacewould follow the hills and valleys of the geoid,reflecting the variations in the strength ofEarth’s gravitational force.

“The Earth’s gravity field changes from onemonth to the next mostly due to the mass ofwater moving around on the surface,” saidWatkins. “Because water in all its forms hasmass and weight, we can actually weigh theocean moving around. We can weigh rainfall,and we can weigh changes in the polar icecaps.”

GRACE observes the Earth’s hydrologic cycleand allows scientists to track water as it evapo-

This diagram illustrates the hydrologic cycle and shows how water circulates over, under, and above the Earth’s sur-face. GRACE data may lead to the identification of new fresh water sources in arid regions on the Earth. (Imagecourtesy of NASA Goddard Space Flight Center)

37

rates into the atmosphere, falls on land in the form of rainfall or snow, orruns off into the ocean. “The biggest freshwater hydrologic events thatGRACE detects are the rainfall runoff in the larger river basins, like theAmazon, and the monsoon cycle in India,” said Tapley.

Detecting how much water is entering the oceans is key to learningabout sea level changes. Other remote sensing instruments can observesea level change, but they can’t discriminate between thermal expansion(when warmer water expands) and additional mass in the form of waterbeing added to the ocean. “GRACE is sensitive only to the portion of sealevel change that is due to water mass being added,” said DonChambers, research scientist at the Center for Space Research. “Mostmodels assume that the total mass of the ocean is constant—that there isno water being added to it or taken away. With GRACE measurements,modelers will need to account for fluctuations in mass.”

Developing a more accurate account of sea level change is important forlow-lying countries such as Tuvalu. Situated in the Pacific Oceanbetween Hawaii and Australia, the country is a combination of nineislands and atolls (ring-like coral islands that enclose a lagoon). Butbecause the islands reach a mere 5 meters (16 feet) above sea level attheir highest point, they are vulnerable to rising oceans. GRACE datacan reveal long-term climate trends that may affect sea level changes.

In addition to gauging changes in water mass on the Earth’s surface,GRACE can detect large-scale moisture changes underground. Forinstance, during record heat waves in Russia in 2002 and Europe in2003, GRACE data enabled scientists to measure the amount of mois-ture that evaporated from the soil during those very dry periods. Thisability will also alert hydrologists to changes in aquifers and under-ground water supplies. “It’s very hard to measure how much water isdeep in the ground and how much it changes from one year to the next.GRACE is one of the few tools we have to do that,” said Watkins. “It canhelp us understand local hydrology, evapotranspiration, precipitation,

and river runoff, and it can give us an idea of how much water is avail-able deep in the Earth for irrigation and agriculture,” said Watkins.

Scientists are also using GRACE data to survey frozen water in the formof ice sheets and large glaciers. Isabella Velicogna, a research scientist atthe University of Colorado, studies mass changes in the Greenland icesheet. “Some components of the seasonal cycle in Greenland are not verywell understood, like ice discharge and subglacial hydrology. GRACEsees some of these components that are difficult to measure,” she said.Other instruments, such as altimeters, can determine elevation changesin the ice sheet, but GRACE sees the total mass, alerting scientists tohow much ice and water are draining off the ice sheet. “GRACE providesinformation that you can’t get from any other satellite instrument,” saidVelicogna.

After analyzing two years of data, Velicogna reported a longer-termtrend: the ice sheet is losing mass. Although other Greenland researchsupports this finding, she added that scientists need a longer time seriesof data to understand what is happening to the ice sheet. Greenlandholds about 2,600,000 cubic kilometers (624,000 cubic miles) of ice,

Like many atolls in the Pacific Ocean, Aitutaki in the Cook Islands rises only a fewmeters above sea level. Several island nations, such as Tuvalu in the Pacific Oceanand the Maldives in the Indian Ocean, are composed entirely of low-lying islands andatolls, making them especially vulnerable to rising sea levels. (Image courtesy of LaurieJ. Schmidt)

“The classic idea of gravity being something that you measure once isno longer accepted. Gravity is an element that scientists must contin-ue to monitor.” - Byron Tapley

which, if melted, would result in a sea level rise of about 6.5 meters (22feet). Since the late nineteenth century, melting ice sheets and glaciershave increased global sea level by about 1 to 2 centimeters (0.3 to 0.7inches) per decade.

Even glaciers that melted long ago affect sea level today. For instance, alarge mass of ice covered the Hudson Bay area during the last Ice Age,which ended around 15,000 years ago. Now, without the weight of gla-ciers, the land beneath that area is slowly rebounding at a rate of about 1centimeter (0.3 inch) per year. Over time, this postglacial rebound affectsregional coastlines, complicating tide gauge readings and making itharder to monitor changes in global sea level. GRACE data will allowscientists to measure the change that can be attributed to postglacialrebound, making it easier to determine how much other factors—such asglobal warming—contribute to rising sea levels.

Investigators designed GRACE as a five-year mission, but scientists hopeto gather data for up to 10 years. Continuing the mission life will allowthem to explore new applications for GRACE data. “We’re combininggravity measurements with other data, like those from ice sheet altime-try or radar altimetry. But we’re still trying to understand what all thesedata tell us,” said Watkins. “It’s a very impressive engineering accom-plishment that allows us to make such detailed measurements. GRACEgives us high-resolution gravity mapping—it’s a pioneering remote sens-ing tool.”

References:Tapley, B.D., S. Bettadpur, M. Watkins, and C. Reigber. 2004. The

gravity recovery and climate experiment: mission overview and early results. Geophysical Research Letters, 31, L09607, doi:10.1029/2004GL019920.

Chambers, D.P., J. Wahr, and R.S. Nerem. 2004. Preliminary observations of global ocean mass variations with GRACE. Geophysical Research Letters, 31, L13310, doi:10.1029/2004GL020461.

For more information, visit the following web sites:Physical Oceanography DAAC

http://podaac.jpl.nasa.gov/grace/GRACE web site

http://www.csr.utexas.edu/grace/GRACE Fact Sheet

http://earthobservatory.nasa.gov/Library/GRACE_Revised/GRACE Space Twins Set to Team up to Track Earth’s Water and Gravity

http://earthobservatory.nasa.gov/Newsroom/NasaNews/2002/200203077838.html

38

Byron Tapley is principal investigator for the GRACE mission. Hisresearch interests focus on determining crustal motion, the Earth’sgeopotential, and ocean and atmosphere circulation. He has servedas Chairman of the Geodesy Section for the American GeophysicalUnion (AGU), is a member of the National Academy ofEngineering, and is a Fellow of AGU, the American Institute ofAeronautics and Astronautics, and the American Association for theAdvancement of Science. Tapley holds a Phd in engineering mechan-ics from the University of Texas at Austin.

Don Chambers, a research scientist at the Center for Space Research,is responsible for analyzing and verifying satellite altimetry dataused to study changes in ocean heat storage, such as those caused byEl Niño. His research includes determining accurate marine geoids.Chambers earned a PhD in aerospace engineering from theUniversity of Texas at Austin.

As project scientist and science data system manager for the GRACEmission, Michael Watkins provides an interface between spacecraftsystem engineering and science requirements, and ensures the devel-opment of all required science analysis algorithms and software. Heearned a PhD in aerospace engineering from the University of Texasat Austin.

Isabella Velicogna is a research associate at the University ofColorado in Boulder, where she examines the use of gravity observa-tions from the GRACE mission to monitor geophysical processes,including hydrology, oceanography, meteorology, glaciology, andsolid earth physics. She received a PhD in applied physics from theUniversity of Trieste, Italy.

39

Land Processes DAAC

by Laurie J. Schmidt

At first glance, Mars and Earth are two verydifferent planets. The desert-like landscape onMars is stark, without vegetation, and seem-ingly lifeless. Its surface area is a mere onequarter that of Earth, yet its largest volcano,Olympus Mons, is three times the height ofMount Everest. And the average temperatureon Mars is a frigid -63 degrees Celsius (-81degrees Fahrenheit), while Earth’s average tem-perature hovers around 15 degrees Celsius (59degrees Fahrenheit).

But when it comes to geologic processes, thetwo worlds are surprisingly alike. And it’s thissimilarity that is helping planetary scientistslearn how to interpret high-resolution remotesensing data of the Martian surface. “Mars is avery similar planet to Earth,” said MichaelRamsey, assistant professor in the Departmentof Geology and Planetary Science at theUniversity of Pittsburgh. “We see wind streaks,volcanoes, water channels—pretty much thesame geologic features that we see on Earth.”

Ramsey is part of a research team developingtechniques to accurately identify and analyzesmall features on the Martian surface in satel-lite images. “In the past, we’ve focused on the

In Search of Martian Craters

El Elegante Crater, part of the Pinacate BiosphereReserve in northwestern Sonora, Mexico, is a maarcrater about 1.6 kilometers (1 mile) in diameter and 244meters (800 feet) deep. (Image courtesy of JimGutmann)

40

large volcanoes, the giant impact basins, andthe big channels,” said Ramsey, “but we’venever really been able to look at the smallthings on Mars.”

The “small things” that Ramsey and his teamhope to shed light on are craters—bowl-shapeddepressions on a planetary surface that are typ-ically caused by one of two processes. Impactcraters form when a meteoroid, asteroid, orcomet collides with a planet. Volcanic cratersdelineate vent areas at the summit of a vol-cano. “When you get down to small sizes, it’shard to tell the difference between a volcaniccrater and an impact crater,” said DavidCrown, senior scientist at the Planetary ScienceInstitute in Tucson, Ariz., and a co-investigatoron the project.

Ramsey and Crown are particularly interestedin a type of volcanic crater known as a maarcrater, which is created by a violent explosionthat occurs as magma moves up toward thesurface and hits groundwater or a body of sur-face water. Typically, magma that containsenough gas to erupt explosively forms a cindercone, as debris accumulates around the vol-cano’s vent. But if abundant water exists in theregion of the volcano, the magma interactswith the water, causing highly explosive erup-tions that build a maar rather than a cindercone. “Maar craters are volcanic, but they’renot volcanoes as you tend to think of them,”said Ramsey. “It’s basically a steam vent, and

they look just like impact craters—a big holein the ground with a lot of rocks around theoutside.”

The researchers hope to come up with criteriathat will enable them to distinguish betweenimpact craters and maar craters solely by look-ing at satellite data. “The goal is to look atsmall craters on Earth, both impact and vol-canic, and study the differences so that as weget high-resolution images of Mars, we canmake useful interpretations,” said Crown.

On Earth, maar craters may fill with water toform a lake or pond. Because the presence ofmaar craters is a strong indication of waterbeneath the surface, the ability to identifythem on Mars has important implications forunderstanding the planet’s geologic history.“We’re interested in craters that are potentiallywater-driven features, because that means therecould be liquid water beneath the surface,which is a big deal for Mars exploration,” saidRamsey.

In the past, most scientists believed that Mars’volcanic activity was most intense in its ancientpast. “The belief has been that Mars was reallyactive in its early time (its first billion years),and that the last three billion years have beenrelatively quiet,” Ramsey said. But the abilityto now see smaller things on the Martian sur-face may change existing beliefs about theplanet’s geologic history.

One of the first things scientists do to estimatethe age of a planetary surface is count thenumber of impact craters. “An old surface willhave a lot more impact craters on it than a

young surface,” said Ramsey. “So if we find anarea on Mars that only has one impact crateron it, that means it’s a very young area.”

Although craters provide important clues intoa planet’s past, Crown said the research com-munity didn’t pay much attention to smallcraters until fairly recently, as new high-resolu-

This image, acquired by the Thermal Emission ImagingSystem (THEMIS) on the Mars Odyssey spacecraft,shows a small crater on the surface of Mars. The crater,located in the Syrtis Major region, is slightly larger thanMeteor Crater in northern Arizona and has been provi-sionally named Winslow Crater (after the small town inArizona near Meteor Crater). The grayscale image wastaking in the visible wavelengths (18 meters per pixel)and shows ejected blocks, the outer crater rim, innercrater wall, and ejecta blanket. (Image courtesy ofNASA/JPL/Arizona State University)

“Trying to understand the history of otherplanets helps us compare them to the Earthand learn about patterns and geologic evolu-tion on a planetary scale.” – David Crown

41

tion images of the Martian surface wereacquired. “The whole idea of volcanic versusimpact craters dates back to the early days oflooking at the moon,” he said. “Initially, scien-tists thought that the craters on the moon weredue to gas bubbles coming out and blastingthrough the surface—that was actually in thescientific literature.” Then in the 1960s, resultsof studies done at sites such as Meteor Craterin northern Arizona showed that the moon’scraters were due to impacts, Crown said.

Although remote sensing on Mars is still in itsinfancy stage, using satellite data to identify

surface land features on Earth is nothing new.Data from the Advanced Spaceborne ThermalEmission and Reflection Radiometer (ASTER),archived at NASA’s Land Processes DistributedActive Archive Center, have been used to studylandforms on the surface of the Earth. Becauseof its high spatial resolution, it enablesresearchers to see features that are relativelysmall. Ramsey, a member of the ASTERScience Team, has used ASTER data to studyvolcanic domes, fire scars, and urban growth.But on Earth, scientists can compare satelliteimagery with ground data to verify their inter-pretations, in a process known as validation—something they can’t do on Mars.

The Thermal Emission Imaging System(THEMIS), one of three instruments launchedonboard the Mars Odyssey in April 2001, gen-erates high-resolution data very similar to thatof ASTER. Images from THEMIS helpedNASA mission scientists choose landing sitesfor the Mars Exploration Rovers in 2003. Oneof the instrument’s major mission goals is tostudy small-scale geologic processes on theMartian surface. But before scientists can beginto interpret the data from THEMIS, they haveto be certain that what they think they’re see-ing is what’s actually there.

So Ramsey, Crown, and their colleagues cameup with a plan: If they could develop sometechniques to accurately identify and examineimpact and maar craters on Earth usingASTER data, then these same techniques couldhelp them distinguish between the two cratertypes in satellite imagery of Mars and assesstheir geologic implications.

Because ASTER and THEMIS generate datathat are so similar in wavelength and pixel size,they offer an ideal comparison opportunity.“It’s almost a perfect analog—like launchingan instrument on Earth and an identical one onMars, and then studying the results,” saidRamsey.

The researchers chose two different crater siteson Earth—an impact crater and a maar crater—and set out to look for differences betweenthe two, using both ASTER data and groundfieldwork.

Meteor Crater (also known as BarringerCrater), located in north-central Arizona,served as the impact crater study site. Knownas one of the best-preserved impact craters onEarth, it is 180 meters (590 feet) deep and 1.2kilometers (0.75 miles) in diameter. Arid cli-mate and a lack of vegetation in the vicinity ofMeteor Crater make it an excellent comparisonsite for similar-sized impact craters on Mars,Ramsey said.

Relatively close to Meteor Crater, just acrossthe Mexican border from southern Arizona, liesthe Pinacate Biosphere Reserve—home to avolcanic field and rare collection of maarcraters. Measuring 1,400 meters (4,593 feet)wide, El Elegante Crater is the largest of theseand was chosen as the maar crater study site.“The maar craters in Pinacate are almost thesame age as Meteor Crater,” said Ramsey, “soyou basically have the same size craters in thesame weathering environment.”

The idea was to make comparisons betweenthe ejecta (material thrown out of a crater dur-

The surface of Earth's moon is covered with impactcraters. They represent an early period in the Moon'shistory when intense meteorite bombardment occurred.(Image from Photos.com)

42

ing an explosion) they found at the rims of thetwo craters. “Since the walls of a crater arepretty vertical and don’t show up in a satelliteimagery, you have to concentrate on looking atthe stuff that was thrown out,” said Ramsey.“Then you can make inferences based on map-ping the ejecta and rocks around the edge andin the general vicinity of the crater.”

During two field campaigns in 2004 at bothcrater sites, the team, led by Ramsey and hisgraduate student Veronica Peet, created adataset that included details about crater rimtopography, block sizes and composition, andvegetation types and percentages—informationthat may reveal some definitive differencesbetween the two crater types.

Although Ramsey is quick to point out thatthe study is still a work in progress, he saidthey already have some initial results. “Rightnow, we can’t tell you that X, Y, and Z tells animpact and maar crater apart,” he said. “Wecould end up with a null result that saysthey’re just too similar to tell them apart usingremote sensing data. But I don’t think that’sthe case—we’ve already found some tantalizingdifferences.”

One difference the team found relates to blocksize and location. “It appears that blocks at ElElegante Crater are bigger, on average, and arelocated closer to the crater rim than the blocksat Meteor Crater,” said Ramsey. If the field-work results show this to be a consistent find-

ing, it could serve as an important yardstick intelling the two crater types apart—on Earthand on Mars, the researchers said.

But the results are preliminary, and furthercomparisons need to be done. “You need toconstantly calibrate what you see on theground with what you see from space,” saidCrown. “It’s a learning process—some thingswe see on the surface are expected, and othersare a big surprise.” Crown cited the example ofthe Martian “blueberries” photographed by theMars rover Opportunity in December 2004.Blueberries are marble-sized pebbles that con-tain hematite, a mineral that supports the ideathat water existed in Mars’ past. “Nobodyknew what that was going to look like,” hesaid, “and every bit of new information canchange things dramatically.”

Understanding how to use satellite data tomap subtle differences around a crater on Earthwill give scientists a better handle on how todo it on Mars. But finding impact craters tostudy on Earth is like a complicated treasurehunt. Erosional processes and plate tectonicseffectively erase impact craters from view. “Ifyou look at the Earth’s surface, your firstimpression would be that there are no impactcraters on Earth. In reality, there are lots ofolder impact craters here, but they’ve beenweathered away,” said Ramsey.

While only about 120 impact craters have beenidentified on the Earth, scientists estimate thaton the surface of Mars, there are more than43,000 impact craters with diameters greaterthan 5 kilometers (3 miles), and probably over

Meteor Crater, located in north-central Arizona, is one of the most recent and well-preserved impact crater sites onEarth. The region's arid climate and lack of vegetation make Meteor Crater an excellent analog for similar-sizedimpact sites on Mars. (Image courtesy of Michael Ramsey)

43

a quarter of a million impact craters that aresimilar in size to Meteor Crater. Scientistsbelieve that most craters on Mars were formedby meteorite impact early in Mars’ history, butsome may be from more recent impacts.

“Things on Mars stay around a lot longer thanon Earth. You don’t have plate tectonics eras-ing things as they go down a subduction zone,”said Ramsey. “So our intent is to do the best wecan making comparisons with these twocraters, come up with some classifications, andthen go to work on the THEMIS data fromMars.”

Learning about small-scale processes on Marscan also provide valuable clues into the planet’sclimate history. “I don’t think there’s any ques-tion that there’s water and ice near the Martiansurface,” said Crown. “The question is, ‘whatdoes this say about the climate history of theplanet?’ Does the existence of water on Mars inits earlier history mean that there was anatmosphere and a warmer planet? Or does itmean that there is just water locked up in thesurface that sometimes gets released from theinterior?” Answers to questions such as thesecould reveal information about Earth’s futureclimate.

“It’s all part of the big question: Why studythe planets?” said Crown. “Trying to under-stand the history of other planets helps us com-pare them to the Earth and learn about pat-terns and geologic evolution on a planetaryscale.”

For now, the researchers continue to focus onone very small piece of the planetary geology

puzzle: learning how to recognize differenttypes of craters using satellite data. “We can’twalk around to every crater on Mars and exam-ine the material around the rims,” said Ramsey.“Right now, we have to look at these featuresfrom space.”

References:McGeary, D. and C.C. Plummer. 1992. Physical

Geology, Earth Revealed. Wm. C. Brown Publishers.

Peet, V.M., M.S. Ramsey, and D.A. Crown. 2005. Comparison of terrestrial morphology,ejecta, and sediment transport of small craters: volcanic and impact analogs to Mars. Lunar Planet. Sci. Conf. #XXXVI, abs.#2080 (CD-ROM), 2005.

For more information, visit the followingweb sites:Land Processes (LP) Distributed Active Archive

Centerhttp://lpdaac.usgs.gov/main.asp

Image Visualization and Infrared Spectroscopy (IVIS) Laboratoryhttp://ivis.eps.pitt.edu/

NASA’s Mars Exploration Programhttp://mars.jpl.nasa.gov/

Advanced Spaceborne Thermal Emission and Reflection Radiometer (ASTER)http://asterweb.jpl.nasa.gov/

Thermal Emission Imaging System (THEMIS)http://themis.asu.edu/

Michael Ramsey is an assistant pro-fessor in the Department of Geologyand Planetary Science at theUniversity of Pittsburgh, where heformed and supervises the ImageVisualization and InfraredSpectroscopy (IVIS) Laboratory. Heis also a science team member for

NASA’s Advanced Spaceborne Thermal Emission andReflectance Radiometer (ASTER) instrument, with a focuson using ASTER's unique thermal infrared capability foremergency hazard response. His varied research interestsinclude remote sensing of active volcanoes, planetary sur-face processes using Earth analogs, and the science of nat-ural hazard reduction. He holds a PhD in geology fromArizona State University.

David A. Crown is a senior scientistat the Planetary Science Institute(PSI) in Tucson, Arizona, withresearch interests in planetary geolo-gy, physical volcanology, and remotesensing. His current research focuseson the geologic history of Mars andintegrating analyses of spacecraft

imaging with field and remote sensing studies of terrestri-al analogs. Prior to joining PSI in 2001, Crown was anassistant professor in the Department of Geology andPlanetary Science at the University of Pittsburgh and aNational Research Council Research Associate at NASA’sJet Propulsion Laboratory. He received a PhD in geologyfrom Arizona State University.

44

Socioeconomic Data and ApplicationsCenter (SEDAC)

by Amanda Leigh Haag

Mention to someone that a little known speciessuch as the “Alerce”—a large tree native toChile and Argentina—is endangered, andthey’re likely to not bat an eyelash. But thendrop the name of a more relatable endangeredspecies such as, say, the Slender Loris from SriLanka—a fuzzy little primate with brown eyesthat would give any teddy bear a run for itsmoney—and it’s likely to elicit more compas-sion.

Both are endangered, due in large part to habi-tat fragmentation from logging and otherhuman activities. And both are evidence of theEarth’s dwindling natural resources, accordingto the Millennium Ecosystem Assessment(MA), the first-ever global effort to take stockof the planet’s ecosystem health. And the vital

Checking Earth’s Vital Signs

The Madagascan Tomato Frog, found only on the north-eastern coast of Madagascar, has suffered severe habi-tat loss due to encroachment by humans and deforesta-tion. (Image from Photos.com)

45

signs aren’t looking good. Of the 24 categoriesof ecosystem health that were evaluated, 15 arebeing seriously degraded at a rate that cannotbe sustained, said Walt Reid, director of theMA, an international, multimillion-dollarundertaking. “If we think of the planet’secosystem services as a bank account that couldlast indefinitely if managed wisely, we areinstead spending the principal. That does pro-vide short-term benefits, but the long-termcosts will be significant,” said Reid. By alteringthe planet, be it through deforestation, over-fishing, or degradation of land and climatechange, “we’re depleting a capital asset,” hesaid.

The assessment represents scientists’ first glob-al-scale attempt at putting a price tag on thebenefits that humans get from natural systems,a concept dubbed ‘ecosystem services.’ Some ofthe services studied include things that can betraded in markets, such as food and fibers.Others have less tangible, yet equally impor-tant, values, such as the flood control and pre-vention of erosion that forests provide. Thesummary of the MA findings, released onMarch 30, 2005, was the result of four years ofwork by 1,300 scientists from 95 countries.

“We looked at ecosystems through this lens ofthe services they provide to people,” said Reid.“For example, we weren’t looking at tropicalforests as just a nice thing society ought to pro-tect. Instead, the researchers examined the eco-nomic and health benefits that people obtainfrom the forest.” And what they found is causefor grave concern. “In these utilitarian terms,the degradation of ecosystem services repre-sents a substantial cost,” said Reid. “We’re los-

ing in economic terms, and we’re losing interms of human health.” The study showedthat not only have ecosystems been dramatical-ly altered, but they’ve been changed more inthe last 50 years than in any comparable timeperiod in human history.

By analyzing data from the NASA Socioeco-nomic Data and Applications Center (SEDAC)at Columbia University, scientists determinedthat in dryland ecosystems, such as sub-Saharan Africa and Central Asia, land degrada-tion through agricultural use and climatechange is closely linked to growing poverty.Drylands cover 41 percent of the Earth’s landsurface and are home to more than 2 billionpeople—a third of the human population.Desertification, or the expansion of dry landsunsuitable for crop production, is one of thegreatest threats to human well-being, accord-ing to the report. “Drylands, which are amongthe most fragile ecosystems, are now the areaswith the highest rate of population growth,”said Reid. “Many dryland regions have becomepoverty traps, where a spiral of growing pover-ty and environmental degradation feed on oneanother.”

People living in drylands depend primarily onthe productivity of their land for their liveli-hoods. Crop and dairy production, livestock,and growth of fuel and construction materialsall depend upon plant productivity, which isinextricably linked to water availability.Overgrazing and intensive cultivation in areasthat do not have adequate levels of nutrientsand water supply can lead to greater soil lossthrough erosion, reduced water quality, andultimately, less vegetation to sustain life.

In addition, climate change has led to moreextreme drought cycles, according to the MAreport. At least 90 percent of dryland popula-tions live in the developing world, makingthem far less technologically able to adapt tothese vulnerabilities. As a result of the MAfindings, the link between environmentaldegradation and poverty is becoming the focusof considerable attention in high-level interna-tional circles, said Reid.

Other “costs” to human well-being, accordingto the MA study, include soil degradation andloss of pollinating insects, loss of water purifi-cation services, and decreased flood control, toname a few. In addition, scientists were able toput a specific number—2 degrees Celsius (3.6degrees Fahrenheit)—on the amount of warm-ing that the planet can sustain before ecosys-tems begin to deteriorate. While some north-ern latitude countries might actually benefitfrom a warmer climate, due to longer growing

People living in dryland regions depend heavily on cropproduction and livestock for their livelihoods.Overgrazing and intensive cultivation in areas that donot have adequate levels of nutrients and water supplycan lead to greater soil loss through erosion, reducedwater quality, and less vegetation to sustain life. (Imagefrom Photos.com)

46

seasons, “once you get beyond 2 degreesCelsius, there’s really no region of the worldwhere the benefits outweigh the costs,” saidReid.

The study also determined that degradation offisheries resources warrants immediate action.In most cases, both marine and freshwater fish-eries have either peaked or are being seriouslyover-harvested, according to the report. Theconcept of over-fishing is certainly not a newone. But by studying population data fromSEDAC, scientists participating in the MAwere able to determine where fisheries-depend-ent populations live and estimate the potentialcost to their livelihoods.

“A handful of earlier reports awakened peopleto the threats of biodiversity loss and habitatloss, but for the most part, they have not gen-erated a groundswell of concern or action,” saidMarc Levy, who co-authored one of the chap-ters of the assessment and coordinated the useof SEDAC data in the MA study. “But directlyrelating the ecosystem well-being to humanwell-being can make a difference, because itgives people a better sense of what’s at stake.”

Before the MA study, scientists had to “engagein fairly idle guesswork” in order to determinewhere people live with respect to vulnerableecosystems, said Levy. “You can look throughthe literature and find these numbers that peo-

ple just pulled out of thin air.” So researchersapplied the highest-quality population data setavailable—known as the “Gridded Populationof the World” (GPW)—and superimposed iton the countries that the MA was studying,yielding much more precise estimates.

One challenge faced by the MA in trying toreconcile ecosystem services with their value tohuman welfare was the potential for conflict ofinterest, according to Levy. “The conservation

folks suspect that the poverty reduction peoplewant to go in and build roads to connect mar-kets and subsidize fertilizers to help farmersgrow higher yields,” said Levy. “At the sametime, the poverty reduction people think theconservationists just don’t care about people.But the MA brought together a lot of peoplewho overcame those tensions in order to worktogether in a framework that was equally fairto both sides.”

“Directly relating ecosystem well-being tohuman well-being can make a difference,because it gives people a better sense ofwhat’s at stake.” - Marc Levy

Loss of pollinating insects is among the many "costs" associated with deteriorating ecosystem health, according tothe Millennium Ecosystem Assessment. (Image from Photos.com)

47

Another challenge that scientists faced over thecourse of the study is that until now, almost allenvironmental monitoring has taken place atthe national level. There is simply no methodin place to survey environmental variables on aglobal scale, said Anthony Janetos, a coordinat-ing lead author of the study from The H. JohnHeinz III Center for Science, Economics andthe Environment. In some cases, countriesdon’t have the financial resources to do large-scale surveys; in other cases, it is difficult tomeasure resources that no one country owns,such as fish in the ocean, said Janetos. “It’s notlike weather data, where every nation under-stands that in order to do a good job under-standing weather and climate, you really needa global observation system,” said Janetos.“That’s still a new concept for looking atecosystems.”

But the scientists stress that the upside of theMA findings is that many of the trends thatare occurring can be reversed. The findingsweren’t the “traditional doom and gloom,‘we’re destroying the world and there’s nothingwe can do about it,’” said Reid. “Three of thefour scenarios that we developed in this assess-ment actually showed that in the next 50years, it is possible to turn this situationaround and protect many of the ecosystemservices.”

Reid noted that one exception to this is the lossof biodiversity. The MA concluded that due tohabitat loss from human activities, a 10-15percent extinction rate of species is expected bythe year 2050. “We’ve got to feed another 3billion people on the planet, so there’s going tobe more loss of habitat,” said Reid. “The endresult is you still end up with a substantialamount of species committed to extinction. Itwould be very hard to end the crisis of speciesextinction over the next 50 years.”

For more information, visit the followingweb sites:Socioeconomic Data and Applications Center

(SEDAC)http://sedac.ciesin.columbia.edu/

Center for International Earth Science Information Networkhttp://www.ciesin.columbia.edu/

Millennium Ecosystem Assessmenthttp://www.millenniumassessment.org/en/index.aspx

Anthony C. Janetos is vice presidentof The Heinz Center, where hedirects the Center’s Global ChangeProgram. He has also served as vicepresident for Science and Researchat the World Resources Institute,senior scientist for the Land-Coverand Land-Use Change Program in

NASA’s Office of Earth Science, and program scientistfor NASA’s Landsat 7 mission. Janetos has written andspoken widely to policy, business, and scientific audienceson the need to understand the scientific, environmental,economic, and policy linkages among the major globalenvironmental issues, and on the importance of consider-ing basic human needs in environmental policymaking.He holds a PhD in biology from Princeton University.

Marc Levy heads the ScienceApplications group at the Center forInternational Earth ScienceInformation Network (CIESIN),part of the Earth Institute atColumbia University. He also leadsCIESIN’s work on theEnvironmental Sustainability Index,

Poverty Mapping, and the Human Footprint; serves as aproject scientist for the Socioeconomic Data andApplications Center; coordinates CIESIN’s involvementwith the Millennium Development Project, and directswork on conflict early warning. Levy is a convening leadauthor for the Millennium Ecosystem Assessment.

Walter Reid directed theMillennium Ecosystem Assessment(MA) from 1998 until the release ofthe project’s findings in March2005. He continues to work withthe MA secretariat to disseminatethe findings and support ongoingactivities linked to the Assessment.

Reid has also served as coordinator of the Puget SoundSalmon Collaboration, and as vice president of the WorldResources Institute (WRI) in Washington D.C., where hewas responsible for the oversight of WRI’s strategic focusand research programs. He earned his PhD in zoology(ecology and evolutionary biology) from the University ofWashington.

The Slender Loris (Loris tardigradus) from Sri Lanka isassessed as endangered and is on the WorldConservation Union's 2004 Red List of ThreatenedSpecies. Habitat fragmentation over the years has seri-ously reduced the area available for this species.Between 1956 and 1993, Sri Lanka lost more than 50%of its forest cover to human activities, followed by asimilar rate of decline in the remaining forest coverbetween 1994 and 2003. (Image copyright K.A.I.Nekaris)

48

To help forecasters better incorporate satellitedata into their decision-making process, NASApartnered with the National Weather Service(NWS) to form the Short-term PredictionResearch and Transition Center (SPoRT), basedat the Global Hydrology and Climate Centerin Huntsville, Ala. SPoRT provides a centrallocation where NASA and scientists can inter-act with weather service meteorologists andsupply them with data products specificallysuited to their needs.

SPoRT atmospheric scientist Gary Jedlovec andhis colleagues rely on feedback from forecastersto create specialized data products. “We askedthem, ‘what are your forecast issues and prob-lems?’ We didn’t just send data to them to seewhat they thought,” said Jedlovec. “We want-ed to match up a product or a data set with aparticular problem.”

What forecasters wanted was data in near-realtime, meaning they need access to the datavery soon after it is received from the satellite.“If it takes six hours for forecasters to receivethe data, particularly in the case of clouds orthunderstorms that change on a short timescale, the information is not going to be ofmuch value to them,” said Jedlovec. Forecastersalso requested higher-resolution imagery thatwould help them estimate cloud cover andheight, fog, sea and land surface temperatures,and precipitation.

Although the weather service had already beenreceiving imagery from the GeostationaryOperational Environmental Satellite (GOES)mission, some of the products had low resolu-tion. In particular, resolution for the atmos-

Global Hydrology Resource Center

by Laura Naranjo

Weather has always been difficult to predict.For centuries, people relied on myths and folk-lore to reckon the weather. Some sayings, likethe rhyme “Ring around the moon, rain bynoon. Ring around the sun, rain before night isdone,” actually contain some truth. Because icecrystals in clouds can cause haloes around thesun or moon, these rings are a genuine indica-

tion that wet weather is on the way. But otherfolklore legends, such as interpreting a ground-hog’s shadow to mean six more weeks of win-ter, are pure fiction.

Weather forecasters can now forego the folk-loric guesswork and instead rely on a variety ofground and space-based instruments to makepredictions. But accurate forecasts require asteady stream of satellite data, and modernmeteorologists must combine traditional fore-casting methods with the latest remote sensing techniques.

SPoRTscast

Fog obscures the landscape around Interstate 40 in the Smoky Mountains west of Asheville, North Carolina. NASA’sShort-term Prediction Research and Transition (SPoRT) Center developed a fog product using MODIS data thathelps forecasters identify regions of developing fog at night and early in the morning. (Image courtesy of NOAAPhoto Library; Ralph F. Kresge, Photographer)

49

pheric water vapor product was about 50 kilo-meters (31 miles), which is slightly larger thanthe north-to-south width of Oklahoma’s pan-handle. Jedlovec was familiar with NASA’sEarth Observing System (EOS) satellites andknew they could be used to provide productswith the improved resolution that forecastersneeded.

NASA’s Moderate Resolution ImagingSpectroradiometer (MODIS) instrument, forinstance, provides measurements with 250 to1,000 meter resolution, resulting in imagerywith much finer detail. By detecting a broaderrange of energy reflected or emitted by theEarth, MODIS can observe weather variablesand physical characteristics of land and oceansurfaces that can’t be detected by GOES.

MODIS gives forecasters improved data forcloud thickness and land and sea surface tem-peratures—all of which contribute to evolvingweather. For instance, harmless, fluffy cumulusclouds can rapidly evolve into towering cumu-lonimbus clouds that often result in powerfulstorms. And land and ocean surface tempera-tures affect wind direction along the coasts. “Ifforecasters have timely, accurate, high-resolu-tion sea and land surface temperature informa-tion, they can better understand and predictland and sea breeze circulations, which affect

cloud cover and also influence where thunder-storms will occur,” said Jedlovec.

For instance, on July 18, 2005, SPoRT provid-ed forecasters with MODIS imagery thatshowed severe weather conditions developingover Huntsville. The imagery indicated atmos-pheric conditions that were likely to producestorms, and indeed, a severe storm formed overthe city, dropping more than 2.5 inches of rainin 30 minutes. During stormy weather, high-resolution MODIS data help forecasters moreaccurately pinpoint and follow storm develop-ment.

The SPoRT team also produced a nighttime fogproduct. Fog develops when air near the sur-face cools, condensing atmospheric moistureinto small water droplets suspended near theground. Due to reduced visibility, fog can cre-ate hazardous travel conditions. BecauseMODIS can observe subtle changes in temper-ature and cloud cover, the SPoRT fog producthelps forecasters identify regions of developingfog at night and early in the morning.

Meteorologists also need accurate informationabout rainfall, so in May 2005, SPoRT begansupplying several NWS Forecast Offices datafrom NASA’s Advanced Microwave ScanningRadiometer-EOS (AMSR-E), which measureswater vapor, precipitation, and other hydrolog-ic factors in the Earth’s environment. By moni-toring marine weather phenomena before theyreach land, forecasters can better anticipate theimpact of storm systems too far off shore to bedetected by land-based radars. “Forecasters arenow using real-time precipitation data fromAMSR-E to monitor offshore rain systems ap-

proaching coastal Florida,” said Jedlovec.AMSR-E data are archived at NASA’s GlobalHydrology Resource Center (GHRC) inHuntsville, which handles data managementand distribution for the SPoRT project.

In addition to MODIS and AMSR-E imagery,the SPoRT team also plans to include datafrom NASA’s Atmospheric Infrared Sounder(AIRS) in its products by the fall of 2005.

“If it takes six hours for forecasters to receivethe data, particularly in the case of clouds orthunderstorms that change on a short timescale, the information is not going to be ofmuch value.” - Gary Jedlovec

Because fluffy, fair weather cumulus clouds (top) growvertically, they can quickly develop into the powerful,threatening-looking cumulonimbus clouds (bottom) thatoften precede severe thunderstorms. The tops of fairweather cumulus clouds usually range from about 5,000to 8,000 meters (3,000 to 5,000 feet) high, while thetops of towering cumulonimbus clouds can extend over12,000 meters (39,000 feet) into the air. (Image courtesyof NOAA Photo Library; Ralph F. Kresge, Photographer)

50

Unlike the MODIS and AMSR-E products,which are provided directly to forecasters,AIRS data will be incorporated into weatherservice models used to predict regional weath-er. “We believe that AIRS data will improvemodel predictions of temperature, moisture,clouds, and precipitation in the 0 to 24-hourtime range,” said Jedlovec. In addition, usingMODIS sea surface temperatures in models isexpected to improve the prediction of wind andtemperature along major U.S. coastlines.

But simply sending data to forecasters isn’tenough. So a key part of SPoRT’s mission istraining forecasters how to use the data. “Weshow them how to access and interpret thedata,” said Jedlovec. SPoRT also holds bi-week-ly meetings with the Huntsville NWS ForecastOffice, providing a forum for NASA scientists

and NWS forecasters to share ideas for addi-tional uses of the data.

When severe weather strikes, forecasters don’thave time to visit satellite web sites, downloaddata, and hope that it’s relevant to the weathersituation at hand. By working closely withforecasters, SPoRT designed data imagery thatworks specifically within existing weather serv-ice computer and display systems. “Forecasterscan overlay all their other data right on top ofthe imagery, which enables them to makedirect comparisons and detect changes,” saidJedlovec.

Although SPoRT interacts primarily withsoutheastern weather service offices, where pre-cipitation and coastal circulations are importantweather factors, it also provides unique prod-

ucts to the NWS office in Great Falls,Montana. “Once word got out that we wereproviding real-time MODIS data products tosome NWS Forecast Offices, the science opera-tions officer in Great Falls contacted us to seehow they could get the data,” Jedlovec said. Inresponse, SPoRT generated a MODIS snowproduct that helped forecasters in Montanaunderstand the distribution of snow cover, howit recedes as it melts, and how it contributes toregional flooding. “They can now view dailysnow cover changes, as well as detect possibleice and log jams that could exacerbate floodconditions caused by snowmelt,” said Jedlovec.

Incorporating MODIS and AMSR-E data intoforecasts is also preparing meteorologists towork with products generated by future satel-lites. The Geostationary OperationalEnvironmental Satellite R series (GOES-R),scheduled to launch in 2012 as a joint develop-ment effort between NASA and the NationalOceanic and Atmospheric Administration(NOAA), will provide more accurate data andimprove forecasters’ detection of meteorologicalphenomena. Another collaborative venturebetween NASA, NOAA, and the Departmentof Defense will result in a suite of NationalPolar Operational Environmental SatelliteSystem (NPOESS) satellites, which will collectatmospheric, land, and ocean weather data.

“As soon as the data, products, and imagery aremade available from NPOESS and GOES-R,offices with experience using MODIS data willinstantly know how to use data from these newsatellites,” Jedlovec said. “And the weatherservice will be able to get the data every 15

The MODIS image on the left shows greater cloud and surface detail than the GOES image on the right. MODIS res-olution is fine enough to show developing cloud lines (small white spots near the center of the left image) and pro-vides crisper observations of the cloud cover. (Image courtesy of SPoRT)

51

minutes, instead of only four times a day like they do with MODIS andAMSR-E data.”

While supplying MODIS and AMSR-E data to NWS Forecast Officesnationwide is outside the SPoRT mandate, Jedlovec and his colleaguesare working with the weather service and NOAA to begin sending EOSsatellite data to additional offices. “We’ve established a model and docu-mented a procedure so that this approach can be more easily implement-ed in other locations,” he said.

SPoRT not only generated specialized products and offered training, butstreamlined the data delivery process. Rather than sending data to indi-vidual offices, SPoRT delivers data to a regional weather service hub,where it can be downloaded by many offices. “We only have to send datato one place to serve 25 different weather service offices in the southeast-ern U.S.,” said Jedlovec.

Interacting directly with forecasters has allowed the SPoRT team todevelop high-resolution satellite data products that smoothly integrateinto existing weather service forecasting processes in an efficient andtimely way. “By giving the data directly to forecasters within the NWSdecision support system, it really saves a lot of time. It’s right thereinstantly for them in a format they are familiar with,” Jedlovec said.

References: Jedlovec, G., S. Haines, R. Suggs, T. Bradshaw, C. Darden, and J. Burks.

2004. Use of EOS Data in AWIPS for weather forecasting. Presented at the 20th Conference on Weather Analysis and Forecasting. http://weather.msfc.nasa.gov/sport/publications/jedlovec1.pdf

Goodman, S.J., W.M. Lapenta, G.J. Jedlovec, J.C. Dodge, and J.T. Bradshaw. 2004. The Short-term Prediction Research and Transition (SPoRT) center: a collaborative model for accelerating research into operations. Presented at the 20th Conference on Interactive Information Processing Systems. http://weather.msfc.nasa.gov/sport/publications/ goodman1.pdf

Goodman, S.J., W.M. Lapenta, and G.J. Jedlovec. 2004. Improving the transition of Earth satellite observations from research to operations. Presented at the American Institute of Aeronautics and Astronautics Space 2004 Conference. http://weather.msfc.nasa.gov/sport/publications/aiaa22800sgoodman.pdf

For more information, visit the following web sites:Short-term Prediction Research and Transition Center

http://weather.msfc.nasa.gov/sport/Global Hydrology and Climate Center

http://weather.msfc.nasa.gov/Global Hydrology Resource Center

http://ghrc.msfc.nasa.gov/

Gary Jedlovec is an atmospheric scientist at NASA’s GlobalHydrology and Climate Center. He is involved with NASA’s Short-term Prediction Research and Transition Center (SPoRT), a projectdesigned to transition the use of unique NASA EOS satellite datainto the Huntsville National Weather Service Forecast Office toimprove aviation and public weather forecasts. Jedlovec received hisPhD in meteorology from the University of Wisconsin-Madison.

52

SAR, Sea Ice, Polar Processes, and GeophysicsAlaska Satellite Facility DAAC (ASF DAAC)ASF DAAC User ServicesGeophysical InstituteUniversity of Alaska FairbanksFairbanks, AlaskaVoice: +1 907-474-6166 Fax: +1 907-474-2665Email: asf@eos.nasa.gov Web: http://www.asf.alaska.edu

Upper Atmosphere, Atmospheric Dynamics,Global Land Biosphere, Global Precipitation,and Ocean ColorGSFC Earth Sciences Data and InformationServices Center DAAC (GES DAAC)GSFC DAAC User ServicesNASA Goddard Space Flight CenterGreenbelt, MarylandVoice: +1 301-614-5224, +1 877-422-1222Fax: +1 301-614-5268Email: gsfc@eos.nasa.gov Web: http://daac.gsfc.nasa.gov

Hydrologic Cycle, Severe Weather Interactions,Lightning, and ConvectionGlobal Hydrology Resource Center (GHRC)GHRC User ServicesNASA Marshall Space Flight CenterHuntsville, AlabamaVoice: +1 256-961-7932 Fax: +1 256-961-7123Email: ghrc@eos.nasa.gov,

user-services@ghrc.nsstc.nasa.govWeb: http://ghrc.msfc.nasa.gov

Radiation Budget, Clouds, Aerosols, andTropospheric ChemistryLangley Research Center DAAC (LaRC DAAC)LaRC DAAC User ServicesNASA Langley Research CenterHampton, VirginiaVoice: +1 757-864-8656 Fax: +1 757-864-8807Email: larc@eos.nasa.gov Web: http://eosweb.larc.nasa.gov

Land Processes Land Processes DAAC (LP DAAC)LP DAAC User ServicesU.S. Geological SurveyCenter for Earth Resources Observation andScienceSioux Falls, South DakotaVoice: +1 605-594-6116 Fax: +1 605-594-6963Email: LPDAAC@eos.nasa.gov Web: http://LPDAAC.usgs.gov

Snow and Ice, Cryosphere and ClimateNational Snow and Ice Data Center DAAC(NSIDC DAAC)NSIDC DAAC User ServicesUniversity of ColoradoBoulder, ColoradoVoice: +1 303-492-6199 Fax: +1 303-492-2468Email: nsidc@eos.nasa.gov, nsidc@nsidc.orgWeb: http://nsidc.org

Biogeochemical Dynamics, Ecological Data,and Environmental ProcessesOak Ridge National Laboratory DAAC (ORNLDAAC)ORNL DAAC User ServicesOak Ridge National LaboratoryOak Ridge, TennesseeVoice: +1 865-241-3952 Fax: +1 865-574-4665Email: ornl@eos.nasa.gov Web: http://www.daac.ornl.gov

Oceanic Processes and Air-Sea InteractionsPhysical Oceanography DAAC (PO.DAAC)PO.DAAC User ServicesNASA Jet Propulsion LaboratoryPasadena, CaliforniaVoice: +1 626-744-5508 Fax: +1 626-744-5506Email: podaac@eos.nasa.gov,

podaac@podaac.jpl.nasa.govWeb: http://podaac.jpl.nasa.gov

Population, Sustainability, Geospatial Data,and Multilateral Environmental AgreementsSocioeconomic Data and Applications Center(SEDAC)SEDAC User ServicesCIESIN, Earth Institute at ColumbiaUniversityPalisades, New YorkVoice: +1 845-365-8920 Fax: +1 845-365-8922Email: sedac@eos.nasa.gov Web: http://sedac.ciesin.columbia.edu

Distributed Active Archive Centers

Applications of Earth System Science data from the NASA Science Mission Directorate

NASA Distributed Active Archive Centershttp://nasadaacs.eos.nasa.gov