2SWLPL]DWLRQRI7URSRVSKHULF'HOD\0DSSLQJ )XQFWLRQ3HUIRUPDQFHIRU+LJK3UHFLVLRQ *HRGHWLF$SSOLFDWLRQV V. B. Mendes (1) and R. B. Langley (2) (1)Faculty of Sciences of the University of Lisbon e-mail: vmendes@lmc.fc.ul.pt (2)Department of Geodesy and Geomatics Engineering University of New Brunswick e-mail: lang@unb.ca '25,6 '$<6 $SULO 7RXORXVH
¾ Neutral Atmosphere 0RWLYDWLRQ ¾ Non-dispersive medium at radio frequencies ¾ Effects: ÖPropagation delay ÖRay bending ¾ Troposphere accounts for most of the delay (hence the denomination tropospheric delay ). ¾ Major modeling error for radiometric techniques, which affects the height component of position. ¾ Sea-level rise monitoring, postglacial rebound, earthquake hazard mitigation and other geodetic applications that require mm-level accuracy.
7URSRVSKHULF'HOD\ d trop = d z h m h ( ε) + d z w m w ( ε) +\GURVWDWLF 0DSSLQJ)XQFWLRQ :HW 0DSSLQJ)XQFWLRQ +\GURVWDWLF =HQLWK'HOD\ :HW =HQLWK'HOD\
7HPSHUDWXUH3URILOH3DUDPHWHUV ¾ Nominal global values (based on a standard atmosphere) / No latitudinal or seasonal variation accounted for. ¾ Tropopause height: 11,231 m. ¾ Temperature Lapse Rate: 6.5 C/km ¾ Predicted from tables / In general, no seasonal variation accounted for. ¾ Predicted from databases relative to a given station / In general, available only for a limited number of sites. ¾ Predicted from models
0DSSLQJ)XQFWLRQV ¾ CfA-2.2 (Harvard-Smithsonian Center for Astrophysics) ¾ Davis et al. (1985). Radio Science, Vol. 20, No. 6, pp. 1593-1607. ¾ Lanyi (Jet Propulsion Laboratory) ¾ Lanyi (1984). TAD Progress Rep. 42-78, JPL, pp. 152-159. ¾ Sovers and Jacobs (1996). JPL Publication 83-39. ¾ UNSW931 (Shanghai Astronomical Observatory) ¾ Yan and Ping (1995). The Astronomical Journal, Vol. 110, pp. 934-939.
North America Tropopause Height Variability
North America Lapse Rate Variability
7URSRSDXVH+HLJKWDQG/DSVH5DWH 3UHGLFWLRQ81%PRGHOV UNB98TH1 H t (km) = 7.508 + 2.421exp ts 22.90 UNB98LR1 α( C/ km) = 5.930 + 0.0359 t s t s = surface temperature ( C)
3DUDPHWHU6HWWLQJV &I$DQG816: Version Code α (K/km) H t (km) CfA1 Mean Mean CfA2 6.5 11.231 CfA3 UNB98LR1 UNB98TH1 UNSW1 Mean Mean UNSW2 6.5 11.231 UNSW3 UNB98LR1 UNB98TH1 monthly mean values based on radiosonde observations.
3DUDPHWHU6HWWLQJV /DQ\L Version S P (hpa) T (K) α (K/km) H i (km) RAOB radiosonde observed values. SJ96 interpolation scheme by Sovers and Jacobs [1996]. Mean monthly mean values based on radiosonde observations. H t (km) LA1 { 1013.25 292 6.8165 1.25 12.2 LA2 RAOB SJ96 SJ96 0 SJ96 LA3 RAOB RAOB Mean Mean Mean LA4 RAOB Mean Mean Mean Mean LA5 RAOB Mean 6.5 0 11.231 LA6 RAOB Mean UNB98LR1 0 UNB98TH1
$VVHVVPHQW5HVXOWVε ƒ $VVHVVPHQW5HVXOWVε ƒ 0.06 0.03 0.00-0.03-0.06 Model minus Trace (m) CfA1 CfA2 CfA3 UNSW1 UNSW2 UNSW3
Locations of the Radiosonde Stations
$VVHVVPHQW5HVXOWVε ƒ $VVHVVPHQW5HVXOWVε ƒ 0.16 0.08 0.00-0.08-0.16 Model minus Trace (m) CfA1 CfA2 CfA3 UNSW1 UNSW2 UNSW3
5HVXOWVIRU$OEDQ\ε ƒ &I$DQG816: Differences (cm) Differences (cm) Differences (cm) 6 CfA1 4 UNSW1 2 0-2 -4-6 0 50 100 150 200 250 300 350 6 CfA2 4 UNSW2 2 0-2 -4-6 0 50 100 150 200 250 300 350 6 CfA3 4 UNSW3 2 0-2 -4-6 0 50 100 150 200 250 300 350 Day of Year 1992 Note: Bias of -1 cm added to UNSW
5HVXOWVIRU$OEDQ\ε ƒ /DQ\L 4 2 0-2 4 2 0-2 4 2 0-2 Differences (cm) Differences (cm) -4 4 2 0-2 Differences (cm) Differences (cm) -4 4 2 0-2 Differences (cm) -4-4 -4 4 2 0-2 Differences (cm) -4 LA1 LA2 LA3 LA4 LA5 LA6 0 50 100 150 200 250 300 350 Day of Year 1992
6WDWLVWLFVIRU&I$ε ƒ { &I$ &I$ &I$ Mean Differences (m) BEL HOB QUI BLO FOR BLB KPP GUA JSJ MEX LIH MZT EYW PBI CRP MAF TUS JAN NKX ABQ TAT GSO BNA OAK WAL DEN MAD BRI SLC OVN CHH MFR ALB YSA YYT GGW MUN INL YZT WSE YQD LAN YSM YXY SUN YVN FAI LUL OTZ YLT BEL HOB QUI BLO FOR BLB KPP GUA JSJ MEX LIH MZT EYW PBI CRP MAF TUS JAN NKX ABQ TAT GSO BNA OAK WAL DEN MAD BRI SLC OVN CHH MFR ALB YSA YYT GGW MUN INL YZT WSE YQD LAN YSM YXY SUN YVN FAI LUL OTZ YLT 0.04 0.02 0.00-0.02-0.04 0.02 0.01 0.00 Rms Scatter (m)
6WDWLVWLFVIRU816:ε ƒ { 816: 816: 816: Mean Differences (m) BEL HOB QUI BLO FOR BLB KPP GUA JSJ MEX LIH MZT EYW PBI CRP MAF TUS JAN NKX ABQ TAT GSO BNA OAK WAL DEN MAD BRI SLC OVN CHH MFR ALB YSA YYT GGW MUN INL YZT WSE YQD LAN YSM YXY SUN YVN FAI LUL OTZ YLT BEL HOB QUI BLO FOR BLB KPP GUA JSJ MEX LIH MZT EYW PBI CRP MAF TUS JAN NKX ABQ TAT GSO BNA OAK WAL DEN MAD BRI SLC OVN CHH MFR ALB YSA YYT GGW MUN INL YZT WSE YQD LAN YSM YXY SUN YVN FAI LUL OTZ YLT 0.06 0.03 0.00-0.03-0.06 0.02 0.01 0.00 Rms Scatter (m)
6WDWLVWLFVIRU/DQ\Lε ƒ 6WDWLVWLFVIRU/DQ\Lε ƒ BEL HOB QUI BLO FOR BLB KPP GUA JSJ MEX LIH MZT EYW PBI CRP MAF TUS JAN NKX ABQ TAT GSO BNA OAK WAL DEN MAD BRI SLC OVN CHH MFR ALB YSA YYT GGW MUN INL YZT WSE YQD LAN YSM YXY SUN YVN FAI LUL OTZ YLT Mean Differences (m) -0.03-0.02-0.01 0.00 0.01 0.02 0.03 BEL HOB QUI BLO FOR BLB KPP GUA JSJ MEX LIH MZT EYW PBI CRP MAF TUS JAN NKX ABQ TAT GSO BNA OAK WAL DEN MAD BRI SLC OVN CHH MFR ALB YSA YYT GGW MUN INL YZT WSE YQD LAN YSM YXY SUN YVN FAI LUL OTZ YLT Rms Scatter (m) 0.000 0.005 0.010 0.015
&RQFOXVLRQV Use of nominal values for temperature profile characterization results in degradation in mapping function performance. Use of UNB models for tropopause and lapse rate prediction improves significantly the performance of CfA and UNSW. Lanyi behaves statistically better when driven by mean temperature profiles. The best solution for Lanyi corresponds to the use of UNB models driven by mean surface temperatures, as it improves the r.m.s. scatter. Inversion height information is essential to reduce the bias in Lanyi.