Development of a daily long term record of NOAA-14 AVHRR land surface temperature over AfricaIntroductionDataNOAA-14 AVHRRMethodsGeneral processing segmentsData ingest, calibration and navigationCloud maskImage geolocation and projectionEstimating LST From AVHRRUlivieri's split window LST algorithmSurface emissivityDevelopment of a moderate resolution emissivity mapsGeneration of the continental AVHRR data setProduct evaluationApproachResultsDiscussionConclusionsAcknowledgementsReferencesDevelopment of a daily long term record of NOAA-14 AVHRR land surfacetemperature over AfricaA.C.T. Pinheiroa,⁎, R. Mahoneyb, J.L. Privettea, C.J. TuckeraaNASA GSFC, Greenbelt, MD, USAbGlobal Science and Technology Inc., Greenbelt, MD, USAReceived 29 July 2005; received in revised form 15 March 2006; accepted 19 March 2006AbstractWe developed a new 6-year daily, daytime and nighttime, NOAA-14 AVHRR based land surface temperature (LST) dataset over continentalAfrica for the period 1995 through 2000. The processing chain was developed within the Global Inventory Modeling and Mapping System(GIMMS) at NASA's Goddard Space Flight Center. This paper describes the processing methodology used to convert the Global Area CoverageLevel-1b data into LST and collateral data layers, such as sun and view geometries, cloud mask, local time of observation, and latitude andlongitude. We used the Ulivieri et al. [Ulivieri, C., M.M. Castronuovo, R. Francioni, and A. Cardillo (1994), A split window algorithm forestimating land surface temperature from satellites, Adv. Space Research, 14(3):59–65.] split window algorithm to determine LST values. Thisalgorithm requires as input values of surface emissivity in AVHRR channels 4 and 5. Thus, we developed continental maps of emissivity using anensemble approach that combines laboratory emissivity spectra, MODIS-derived maps of herbaceous and woody fractional cover, and theUNESCO FAO soil map. A preliminary evaluation of the resulting LST product over a savanna woodland in South Africa showed a bias of<0.3 K and an uncertainty of < 1.3 K for daytime retrievals (< 2.5 K for night). More extensive validation is required before statistically significantuncertainties can be determined. The LST production chain described here could be adapted for any wide field of view sensor (e.g., MODIS,VIIRS), and the LST product may be suitable for monitoring spatial and temporal temperature trends, or as input to many process models (e.g.,hydrological, ecosystem).© 2006 Elsevier Inc. All rights reserved.Keywords: Land surface temperature; Advanced Very High Resolution Radiometer; Global Area Coverage; Split window algorithm; Africa1. IntroductionSatellite land surface temperature (LST) products provide anestimate of the kinetic temperature of the earth's surface skin(Norman & Becker, 1995), i.e., the aggregate surface mediumviewed by the sensor to a depth of about 12 μm. LST is a keyparameter for the understanding of the physics of land surfaceprocesses (Sellers et al., 1988). It can be u sed for monitoringvegetati on water stress, assessing surface energy balance,detecting land surface disturbance, and monitoring conditionsuitability for insect–vector disease proliferation, among otheruses. Many of these applications are particularly important overthe African continent given its large expanse and uniqueenvironmental conditions (e.g., strong wet/dry seasonality).Satellites offer the opportunity for the synoptic and continentalor global monitoring of LST.The National Oceanographic and Atmospheric Adminis-tration's (NOAA) Advanced Very High Resolution Radiometer(AVHRR) sensor has recorded top-of-atmosphere brigh tnesstemperature in two thermal infrared channels (TIR; channels 4and 5) from 1981 to the present. These brightness temperaturesdepend on surface emittance, and atmospheric absorption andemission along the path of observation, among other factors.The surface emittance depends on the skin's kinetic temperatureand emissivity.At its coarsest resolution, AVHRR samples each 4 km (perside) at nadir, although pixel size increases with scan angle.Each Earth location is sampled at least once each day and onceRemote Sensing of Environment 103 (2006) 153 – 164www.elsevier.com/locate/rse⁎Corresponding author.E-mail address: [email protected] (A.C.T. Pinheiro).0034-4257/$ - see front matter © 2006 Elsevier Inc. All rights reserved.doi:10.1016/j.rse.2006.03.009each night. Thus, the AVHRR record is valuable for monitoringand trending of many global change phenomena.Several AVHRR-derived temperature products have beencreated to capitalize on the long term record of AVHRR thermalinfrared measurements. For example, the second-generation ofthe Global Vegetation Index (GVI) dataset includes top-of-the-atmosphere brightness temperature values in channels 4 and 5for the daytime overpasses of NOAA-9 and NOAA-11, from1985 to 1994. The global products are provided at 0.15°resolution and at weekly and monthly time scales. An additional5-year monthly climatology is also available (Gutman et al.,1995).The World Land Surface Temperature Atlas dataset,produced by the European Space Agency, also provides anAVHRR temperature product. Thi s dataset relies on brightnesstemperature data collected for years 1992 and 1993. The dataconsists of 1 km resolution LST fields over Europe, and aclimatic monthly LST at half-degree resolution (Kerr et al.,1998). A more recent effort by Jin (2004) provides monthlydiurnal-averages of 10-day composites of skin temperaturefields from 1981 to 1998. The dataset is based on typicalpatterns of the LST diurnal cycle as determined with a GeneralCirculation Model (GCM), and was scaled to 0.5° and 5° spatialresolutions.Despite these efforts, no standard global daily AVHRR LSTproduct exists (Wan et al., 2002). For more than two decades,the Global Inventory Modeling and Mapping System(GIMMS), at NASA's Goddard Space Flight Center, hasproduced global, multimission AVHRR vegetation indexproducts. With its Global Area Processing System (GAPS;Tucker et al., 1994), GIMMS has significantly reduced productcontamination from cloud, atmosphere, bidirectional reflec-tance and sensor artifacts. The scientific value of theseproducts is evident in the major discoveries based on them(e.g., Myneni et al., 1997; Tucker et al., 2001; Nemani et al.,2003).Although GIMMS has packaged AVHRR bands 4 and 5 withsome products (e.g., the NASA AVHRR Pathfinder), it had noLST processing segment.