• Basara, J. B., , and Crawford T. M. , 2000: Improved installation procedures for deep layer soil moisture measurements. J. Atmos. Oceanic Technol., 17 , 879884.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Betts, A. K., 1984: Boundary layer thermodynamics of a high plains severe storm. Mon. Wea. Rev., 112 , 21992211.

  • Betts, A. K., 1992: FIFE atmospheric boundary layer budget methods. J. Geophys. Res., 97 , 1852318531.

  • Betts, A. K., , and Ball J. H. , 1995: The FIFE surface diurnal cycle climate. J. Geophys. Res., 100 , 2567925693.

  • Betts, A. K., , and Ball J. H. , 1998: FIFE surface climate and site-average dataset 1987–89. J. Atmos. Sci., 55 , 10911108.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Betts, A. K., , Hong S-Y. , , and Pan H-L. , 1996: Comparison of NCEP–NCAR reanalysis with 1987 FIFE data. Mon. Wea. Rev., 124 , 14801498.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., 1993: Observations and Theory of Weather Systems. Vol. 2, Synoptic-Dynamic Meteorology in Midlatitudes. Oxford University Press, 594 pp.

    • Search Google Scholar
    • Export Citation
  • Bonner, W. D., 1968: Climatology of the low level jet. Mon. Wea. Rev., 96 , 833850.

  • Brock, F. V., , Crawford K. C. , , Elliott R. L. , , Cuperus G. W. , , Stadler S. J. , , Johnson H. L. , , and Eilts M. D. , 1995: The Oklahoma Mesonet: A technical overview. J. Atmos. Oceanic Technol., 12 , 519.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Crawford K. C. , 2000: Estimating sensible heat flux from the Oklahoma Mesonet. J. Appl. Meteor., 39 , 102116.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Duchon C. E. , 2000: A field comparison among a domeless net radiometer, two four-component net radiometers, and a domed net radiometer. J. Atmos. Oceanic Technol., 17 , 15691582.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Weber D. , 2002: Land-surface scheme validation using the Oklahoma Atmospheric Surface-layer Instrumentation System (OASIS) and Oklahoma Mesonet data: Preliminary results. Meteor. Atmos. Phys., 80 , 189206.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Crawford K. C. , 2003: Examination of the surface energy budget: A comparison of eddy correlation and Bowen ratio measurement systems. J. Hydrometeor., 4 , 160178.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Richardson S. J. , 2003: Spatial and temporal correlation among Oklahoma Mesonet and OASIS surface-layer measurements. J. Appl. Meteor., 42 , 519.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , Richardson S. J. , , Crawford K. C. , , Horst T. W. , , Brock F. V. , , Humes K. S. , , Sorbjan Z. , , and Elliott R. L. , 1999: The Oklahoma Atmospheric Surface-Layer Instrumentation System (OASIS) Project. Preprints, 13th Symp. on Boundary Layers and Turbulence, Dallas, TX, Amer. Meteor. Soc., 612–615.

    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , Richardson S. J. , , and McAloon C. M. , 2001: The OASIS Project network for monitoring the surface energy budget. Preprints, 11th Symp. on Meteorological Observations and Instrumentation, Albuquerque, NM, Amer. Meteor. Soc., 66–71.

    • Search Google Scholar
    • Export Citation
  • Haugland, M. J., , and Crawford K. C. , 2002: The diurnal cycle of dewpoint across Oklahoma's winter wheat belt. Preprints, 13th Symp. on Global Climate Change and Climate Variations, Orlando, FL, Amer. Meteor. Soc., 254–256.

    • Search Google Scholar
    • Export Citation
  • Illston, B. G., , and Basara J. B. , 2003: Climatology of soil moisture variables using the Oklahoma Mesonet. Preprints, 17th Conf. on Hydrology, Long Beach, CA, Amer. Meteor. Soc., CD-ROM, P1.13.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kanamitsu, M., , Ebisuzaki W. , , Woollen J. , , Yang S-K. , , Hnilo J. J. , , Fiorino M. , , and Potter G. L. , 2002: NCEP–DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kistler, R., and Coauthors, 2001: The NCEP–NCAR 50-year reanalysis: Monthly means CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82 , 247268.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., , and Stensrud D. J. , 1998: Mean monthly diurnal cycles observed with PRE-STORM surface data. J. Climate, 11 , 29953009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Roads, J., , and Betts A. , 2000: NCEP–NCAR and ECMWF reanalysis surface water and energy budgets for the Mississippi river basin. J. Hydrometeor., 1 , 8894.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sellers, P. J., , Hall F. G. , , Asrar G. , , Strebel D. E. , , and Murphy R. E. , 1988: The First ISLSCP Field Experiment (FIFE). Bull. Amer. Meteor. Soc., 69 , 2227.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shafer, M. A., , Fiebrich C. A. , , Arndt D. S. , , Fridrickson S. E. , , and Hughes T. W. , 2000: Quality assurance procedures in the Oklahoma Mesonetwork. J. Atmos. Oceanic Technol., 17 , 474494.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shuttleworth, W. J., 1991: Insight from large-scale observational studies of land/atmosphere interactions. Surv. Geophys., 12 , 330.

  • Stensrud, D. J., 1996: Importance of low-level jets to climate: A review. J. Climate, 9 , 16981711.

  • Stokes, G. M., , and Schwartz S. E. , 1994: The Atmospheric Radiation Measurement (ARM) Program: Programmatic background and design of the Cloud and Radiation Testbed. Bull. Amer. Meteor. Soc., 75 , 12011221.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Twine, T. E., , Kustas W. P. , , Norman J. M. , , Cook D. R. , , Houser P. R. , , Meyers T. P. , , Prueger J. H. , , and Starks P. J. , 2000: Correcting eddy covariance flux underestimates over a grassland. Agric. For. Meteor., 103 , 279300.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Uccellini, L. W., , and Johnson D. R. , 1979: The coupling of upper and lower tropospheric jet streaks and implications for the development of severe convective storms. Mon. Wea. Rev., 107 , 682703.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Willmott, C. J., 1982: Some comments on the evaluation of model performance. Bull. Amer. Meteor. Soc., 63 , 13091313.

  • View in gallery

    Oklahoma state map displaying the mesonet sites located at Boise City and Idabel

  • View in gallery

    Monthly averages during 2000 of shortwave radiation (SW; W m−2) for (a) Boise City and (b) Idabel. Monthly averages of longwave radiation (LW; W m−2) for (c) Boise City and (d) Idabel during 2000

  • View in gallery

    Daily averages of albedo (%) as estimated from OASIS and GR from 1 Jan 2000 to 31 Dec 2001 for (a) Boise City and (b) Idabel

  • View in gallery

    Daily estimates of net radiation (W m−2) as observed from the CNR1 and NR-Lite radiometers. Data were collected from Boise City (Bois) and Idabel (Idab) from 1 Jan to 31 Dec 2000

  • View in gallery

    Data collected during 1 Jan 2000–31 Dec 2001 at the Boise City mesonet site (OASIS) is compared against the NCEP–NCAR global reanalysis (GR). Monthly means (W m−2) of (a) net radiation (Rn), (b) ground heat flux (G), (c) latent heat flux (LE), and (d) sensible heat flux (H) are shown. Note that LE and H are EC measurements

  • View in gallery

    As in Fig. 5, but for the Idabel mesonet site

  • View in gallery

    Monthly averages of near-surface specific humidity (q; g kg−1) and potential temperature (θ; K). Data collected during 2000 at the (a) Boise City and (b) Idabel mesonet sites

  • View in gallery

    As in Fig. 7, but for 2001

  • View in gallery

    Monthly diurnal averages of specific humidity (q; g kg−1) and potential temperature (θ; K). Data collected during 2000 at the (a) Boise City and (b) Idabel mesonet sites

  • View in gallery

    As in Fig. 9, but for 2001

  • View in gallery

    A scatterplot of 92-day averages of specific humidity (q; g kg−1) and potential temperature (θ; K) plotted as a function of time of day (UTC). Data collected during Jun, Jul, and Aug of 2000 at the (a) Boise City and (b) Idabel mesonet sites

  • View in gallery

    As in Fig. 11, but for 2001

  • View in gallery

    The summer averages of wind speed (m s−1) for Boise City (Bois) and Idabel (Idab) plotted as a function of time of day (UTC). Data collected during Jun, Jul, and Aug of 2001

  • View in gallery

    Monthly estimates (mm) of (a) precipitation (P) and (b) evapotranspiration (ET). Data collected at the Boise City and Idabel sites during 1 Jan 2000–31 Dec 2001

  • View in gallery

    Monthly estimates (mm) of the change in the amount of water in the near-surface (0–10 cm) soil layer. Data collected at the (a) Boise City and (b) Idabel sites during 1 Jan 2000–31 Dec 2001

  • View in gallery

    Monthly estimates of the near-surface water residual (mm), defined as the monthly sum totals of precipitation (ΣP), evapotranspiration (ΣET), and the monthly change in the near-surface soil moisture (ΔSM0–10cm). Data collected at (a) Boise City and (b) Idabel during 1 Jan 2000–31 Dec 2001

All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 108 108 2
PDF Downloads 11 11 2

A Two-Year Comparison of the Surface Water and Energy Budgets between Two OASIS Sites and NCEP–NCAR Reanalysis Data

View More View Less
  • 1 Center for Analysis and Prediction of Storms, University of Oklahoma, Norman, Oklahoma
© Get Permissions
Full access

Abstract

Few surface-observation networks exist that provide comprehensive multiyear, multiseason observations of surface energy fluxes and subsurface soil moisture and temperature, combined with near-surface atmospheric data. More such networks are needed to validate, improve, and calibrate current global weather and climate models, including those used as the backbone or background models in global reanalysis. One such measurement system is the Oklahoma Mesonet–Oklahoma Atmospheric Surface-layer Instrumentation System (OASIS), which provides atmospheric, surface, and soil data in real time. This study compares 2 yr of surface energy and water budget data from two OASIS sites located in two distinct climate zones with NCEP–NCAR global reanalysis (GR) estimates. The intraseasonal hydrological and thermodynamic cycles are discussed. Results show generally good agreement between most reanalysis values and observations. Incoming and reflected shortwave radiation are largely overestimated by the GR, and incoming longwave radiation is slightly underestimated by the GR when compared to OASIS observations. The GR significantly overestimates latent heat (LE) at the Idabel, Oklahoma (IDAB) site. Furthermore, the GR likely underestimates entrainment of drier air from above the PBL and mixes the turbulent fluxes over too shallow of a layer. Both the reanalysis and observations find a positive water residual for the easternmost site (IDAB) but estimate a negative near-surface water residual for the western site [Boise City, Oklahoma (BOIS)]. Overall, surface fluxes and thermodynamic properties were well analyzed by the reanalysis at capturing the unique features associated with each OASIS site.

Corresponding author address: Dr. Jerald A. Brotzge, Center for Analysis and Prediction of Storms, University of Oklahoma, 100 E. Boyd, Suite #1110, Norman, OK 73019. Email: jbrotzge@ou.edu

Abstract

Few surface-observation networks exist that provide comprehensive multiyear, multiseason observations of surface energy fluxes and subsurface soil moisture and temperature, combined with near-surface atmospheric data. More such networks are needed to validate, improve, and calibrate current global weather and climate models, including those used as the backbone or background models in global reanalysis. One such measurement system is the Oklahoma Mesonet–Oklahoma Atmospheric Surface-layer Instrumentation System (OASIS), which provides atmospheric, surface, and soil data in real time. This study compares 2 yr of surface energy and water budget data from two OASIS sites located in two distinct climate zones with NCEP–NCAR global reanalysis (GR) estimates. The intraseasonal hydrological and thermodynamic cycles are discussed. Results show generally good agreement between most reanalysis values and observations. Incoming and reflected shortwave radiation are largely overestimated by the GR, and incoming longwave radiation is slightly underestimated by the GR when compared to OASIS observations. The GR significantly overestimates latent heat (LE) at the Idabel, Oklahoma (IDAB) site. Furthermore, the GR likely underestimates entrainment of drier air from above the PBL and mixes the turbulent fluxes over too shallow of a layer. Both the reanalysis and observations find a positive water residual for the easternmost site (IDAB) but estimate a negative near-surface water residual for the western site [Boise City, Oklahoma (BOIS)]. Overall, surface fluxes and thermodynamic properties were well analyzed by the reanalysis at capturing the unique features associated with each OASIS site.

Corresponding author address: Dr. Jerald A. Brotzge, Center for Analysis and Prediction of Storms, University of Oklahoma, 100 E. Boyd, Suite #1110, Norman, OK 73019. Email: jbrotzge@ou.edu

1. Introduction

Multiseasonal observations of the soil, surface, and atmospheric state are needed to improve climate model atmosphere and land surface coupling and parameterizations. Shuttleworth (1991) proposed that long-term, single site data be used for calibration and validation of global climate model (GCM) performance, stating that “monthly-average values of weather variables and surface exchange fluxes measured at a point are representative of the monthly-average values for a GCM grid box centered on that point.” However, until recently few measurement systems were available to provide such comprehensive data. One such network, the Oklahoma Mesonet and Oklahoma Atmospheric Surface-layer Instrumentation System (OASIS), has provided continuous observations of soil moisture, surface radiation and fluxes, and atmospheric data since 1 January 2000. Field studies such as the First International Satellite Land Surface Climatology Project (ISLSCP) Field Experiment (FIFE; Sellers et al. 1988) and the Atmospheric Radiation Measurement (ARM) Program (Stokes and Schwartz 1994) have been limited to a single climate zone, whereas the OASIS measurements cover several climate zones across Oklahoma (Brotzge et al. 1999). Thus, data from several climate regions can be tested simultaneously.

The goal of this work is to evaluate the surface energy and water budgets of the National Centers for Environmental Prediction–National Center for Atmospheric Research (NCEP–NCAR) global reanalysis (Kalnay et al. 1996; Kistler et al. 2001) by comparison with 2 yr of OASIS observations collected from two distinct climate regimes. This study provides a unique opportunity to evaluate NCEP–NCAR reanalysis values across multiple seasons to identify seasonally dependent error (e.g., seasonal changes in radiation and precipitation) and across several sites to more clearly identify the impact of site-specific surface parameters (e.g., soil and vegetation type). Furthermore, this study supports the hypothesis posed by Shuttleworth (1991) that long-term, single site data can be used to evaluate coarse-resolution model analysis. This study extends the work by Betts et al. (1996), Betts and Ball (1998), and Roads and Betts (2000).

2. Data

a. Oklahoma Mesonet/OASIS observations

The Oklahoma Mesonet (Brock et al. 1995) is an observational network of 115 meteorological stations across Oklahoma that collects, archives, and quality controls atmospheric, surface, and soil data in real time. Air temperature, atmospheric pressure, relative humidity, precipitation, and wind speed and direction are recorded every 5 min; soil temperature and moisture (soil water potential) are recorded at four depths (5, 25, 60, and 75 cm) every 30 min. During late 1999, the Oklahoma Atmospheric Surface-layer Instrumentation System (Brotzge et al. 1999) outfitted 90 mesonet sites with additional sensors to record net radiation (Rn) and surface and ground fluxes. Additionally, 10 of these 90 sites, dubbed “super sites,” were equipped with four-component net radiometers, sonic anemometers and Krypton hygrometers, and additional ground flux sensors. These stations provided 5-min observations of incoming and outgoing shortwave and longwave radiation; sensible heat (H), latent heat (LE), and ground heat (G) fluxes; and skin temperature. At least one super site was located in each of Oklahoma's nine climatic zones to ensure a diverse yet simultaneous set of observations from across the state. Data for this study were collected from 1 January 2000 to 31 December 2001 from two super sites near Boise City and Idabel (Fig. 1). Brotzge and Weber (2002) provide a complete description of all radiation and flux instrumentation and measurement techniques, and Basara and Crawford (2000) provide a description of the soil matric potential sensors, installation methods, and the conversion of the soil water potential to soil moisture estimates.

All data were calibrated and quality controlled as described by Brotzge et al. (2001). Standard mesonet data were further quality assured by a series of automated and manual checks (Shafer et al. 2000). Redundant instrumentation at the super sites allowed most missing flux data to be replaced; for example, during precipitation, data from the sonic anemometer are unavailable. Thus, sensible heat flux estimates were replaced using a gradient profile technique (Brotzge and Crawford 2000). If no redundant instrumentation were available (e.g., ground heat flux and soil moisture), missing hourly data were linearly interpolated.

Details of the two observing sites, Boise City (BOIS) and Idabel (IDAB), are provided in Table 1. BOIS is located in the westernmost county of the Oklahoma panhandle. The site is characterized by very flat terrain with excellent fetch from all directions. Vegetation is sparse and consists of short grasses and scrub brush. IDAB is in the southeasternmost county of Oklahoma. The site is characterized by gently rolling pasture with patches of trees within the fetch area. Annual rainfall at BOIS averages less than 500 mm, while IDAB averages 1030 mm yr−1. The annual mean temperature at BOIS is 13.5°C, and the annual mean temperature at IDAB is 16.6°C. (Additional information and photographs of the sites can be found online at http://www.mesonet_ou.edu/siteinfo/.)

b. NCEP–NCAR reanalysis data

The NCEP–NCAR reanalysis (Kalnay et al. 1996; Kistler et al. 2001) provides long-term model analysis ideal for the study and initialization of global- and regional-climate-scale models. Global reanalysis (GR) estimates were generated using a single, “frozen” version of an operational numerical model, which was then run for the period from 1948 to present. All reanalysis values used in this study were analyzed on a 192 × 94 point Gaussian grid (approximate 1.905°N × 1.875°E resolution) and bilinearly interpolated from the surrounding four grid points to the mesonet site for comparison.

3. Surface radiation budget

Monthly means of near-surface radiation were estimated for BOIS (Figs. 2a,c; Table 2a) and IDAB (Figs. 2b,d; Table 2b) for 2000 and 2001 from 5-min observations of incoming and outgoing shortwave and longwave radiation. Results for 2001 are identical to those from 2000 and are not shown. An examination of shortwave radiation data from the OASIS and GR shows much higher incoming shortwave radiation by the GR at both sites for every month during the 2-yr study period. Incoming shortwave radiation by GR was overestimated by 16.9% at BOIS and 27.3% at IDAB relative to the observed values. Betts et al. (1996) also discovered that GR estimates of solar radiation were too large and attributed the error to the model atmosphere being too transparent and too few clouds being produced in the model. Improved solar radiation analysis was one impetus for developing the NCEP/Atmospheric Model Intercomparison Project II (AMIP-II) Department of Energy (DOE) reanalysis (Kanamitsu et al. 2002). The GR underestimated incoming longwave radiation at both sites relative to OASIS by 5.5% at BOIS and 6.0% at IDAB during every month of the 2-yr period. The overestimation in incoming shortwave radiation and the underestimation in longwave radiation likely indicate that too few clouds were produced by the model analysis. Differences in emitted longwave radiation between GR and OASIS are almost negligible, with a 2-yr mean difference (GR − OASIS) at BOIS of −3.9 W m−2 and a 0.2 W m−2 difference at IDAB.

The significant overestimation of incoming shortwave radiation by the GR is largely offset by a model analysis albedo that is much too large. Reflected shortwave radiation is 58.3% greater at BOIS and 51.7% greater at IDAB than observations. This yields a 2-yr mean albedo at BOIS of 28.0% from the GR compared to measured OASIS values of 21.2%. Albedo estimates at IDAB were 24.4% from the GR and 20.6% from OASIS. Betts et al. (1996) found very similar results from FIFE data. A plot of the daily GR and OASIS albedo data from BOIS and IDAB shows the rather static nature of the GR estimates (Fig. 3). The daily minimum albedo from the GR is, in general, about 5% higher than those observed, and little daily to monthly variability is noted. The observed albedo from OASIS reflects daily changes in the near-surface soil moisture, vegetation state, and snow cover. Only snow cover acts to significantly alter the GR albedo estimates.

Two different sensors are installed at the super sites for measuring net radiation, and the data from these systems are compared to ensure the quality of the radiation data. The net radiation data, as summed from the four components of the CNR1 radiometer (as used in Figs. 2 and 3), are compared against the net radiation data as measured by the NR-Lite, a simpler, domeless net radiometer. Daily estimates were computed from each system from data collected during 2000 (Fig. 4). On average, the NR-Lite underestimated net radiation at BOIS by 14.8 W m−2 with a mean standard deviation of 4.8 W m−2 compared to the CNR1. Results from IDAB were similar, with the NR-Lite underestimating radiation by 11.3 W m−2 with a mean standard deviation of 7.3 W m−2. Brotzge and Duchon (2000) found a similar comparison, with the NR-Lite underestimating net radiation compared to the CNR1 because of calibration and wind- and precipitation-induced cooling errors. Nevertheless, Brotzge and Duchon (2000) and Brotzge and Crawford (2003) found the CNR1 to be a more stable and accurate sensor than the NR-Lite.

4. Surface energy budget

Net radiation estimates (Figs. 5a and 6a) were summed from 5-min observations of incoming and outgoing shortwave and longwave radiation as shown in Fig. 2. Monthly means of G, LE, and H were estimated from 5-min observations (Figs. 5b–d and 6b–d).

Because of the large model analysis albedo, the GR estimates of Rn at BOIS are 4.0% (4 W m−2) less than those measured from OASIS. Net radiation analysis at IDAB is 10.6% (10.5 W m−2) higher than those measured; the Rn bias is much greater at IDAB than at BOIS in part because the incoming shortwave radiation bias was much greater at IDAB than at BOIS.

Given the significant spatial heterogeneity often associated with soil parameters, the model analyses of ground heat flux at BOIS (Fig. 5b) and IDAB (Fig. 6b) were remarkably similar to observed values. Positive values of G represent energy absorbed by the soil, and negative values represent energy extracted from the soil. While monthly values of G during the winter season generally ranged between −5 and −10 W m−2, such values exceeded 30% of the available energy (RnG). The 2-yr mean difference (GR − OASIS) of G was 0.0 W m−2 at BOIS and −1.5 W m−2 at IDAB. The monthly variability was greater among observed values at both sites, but monthly trends were captured by the analysis. Correlation estimates between OASIS and GR values were promising, with a correlation of 0.93 at BOIS and 0.96 at IDAB.

Incoming energy not absorbed by the soil or emitted or reflected from the surface is available for conversion into sensible heat and latent heat energy. Monthly means of LE (Figs. 5c and 6c) and H (Figs. 5d and 6d) were estimated for BOIS and IDAB during 2000 and 2001. Estimates of LEsonic from GR and OASIS compared favorably at BOIS with a 2-yr mean difference of only 1.8 W m−2. Results were much less favorable at IDAB where the mean difference was 27.2 W m−2.

Sensible heat fluxes from the GR and OASIS compared more favorably than did LE (Figs. 5d and 6d). Two-year mean differences of H were only 2.2 W m−2 at BOIS and 0.3 W m−2 at IDAB. For the 2-yr, GR for BOIS underestimated H relative to the measurements during the winter months and overestimated H during the summer months. However, for IDAB, the pattern is less regular, and the GR estimates of H were less than those from OASIS during May–December 2001.

One reason for the large differences in LE at IDAB between the GR and observations is the significant overestimation in the GR values of Rn. As discussed above, GR estimates of Rn at IDAB are over 10% too high. To test the impact of this overestimation on the monthly flux totals, GR estimates of G, H, and LE were scaled as a function of the observed net radiation. The difference between the GR values and the observations decreased for LE to 19.8 W m−2; however, the differences for H and G actually increased to −2.5 and −1.7 W m−2, respectively. Rms differences remained the same. Similar adjustments made to the estimates at BOIS made all flux differences increase.

A second explanation for the large differences in LE at IDAB between the GR and observations is reanalysis model error. As noted by Kalnay et al. (1996), reanalysis surface fluxes are derived solely from the model physics and are not directly influenced by any assimilated observations. Thus, GR flux estimates are only as accurate as the model physics.

A third reason for the large differences in LE at IDAB is measurement errors of LEsonic by the OASIS. The OASIS directly estimates LE from an eddy covariance (EC) system that includes a sonic anemometer and Krypton hygrometer. A review of previous field programs found that EC systems typically underestimate surface fluxes (Twine et al. 2000; Brotzge and Crawford 2003). At EC sites, the independently measured terms of the surface energy budget, as defined by (RnH − LE − G) do not sum to zero as otherwise expected by the conservation of energy. Instead, (RnH − LE − G) > 0, and either Rn is overestimated, and/or H, LE, and/or G is underestimated. Twine et al. (2000) determined that both H and LE are underestimated, as a function of the Bowen ratio. Brotzge and Crawford (2003) found that only LE is likely underestimated. This problem is typically quantified in terms of its residual, (RnH − LE − G = Residual) or in terms of “closure,” defined by C (%) = (H + LE) / (RnG) × 100. In this study, the mean monthly residual (closure) was measured at 11 W m−2 (87.3%) at BOIS and 17.1 W m−2 (83.0%) at IDAB.

One option for determining the “true” LE is simply by estimating LE as the residual of the surface energy budget (LEresid = RnHsonicG). This solution implies that all measurement errors appear in LEresid. This option decreased (GR − OASIS) LE differences at IDAB from 27.3 to 10.1 W m−2. Furthermore, when the net radiation “adjustment” is applied, the mean (GR − OASIS) difference decreases further to 2.7 W m−2. A second option, posed by Twine et al. (2000), is to estimate a new H (HBR) and LE (LEBR) based upon the measured 30-min Bowen ratios (BRs) as defined by the original EC observations of H (Hsonic) and LE (LEsonic). This option also decreased (GR − OASIS) LE differences at IDAB to 13.9 W m−2, but increased (GR − OASIS) H differences at IDAB from 0.3 to 3.4 W m−2. The net radiation adjustment decreased LE differences to 6.5 W m−2, but increased H difference to 6.3 W m−2. Neither the residual method nor the Bowen ratio method decreased the (GR − OASIS) flux differences at BOIS.

5. Impact of surface forcing upon the boundary layer

Betts (1984, 1992) introduced the use of thermodynamic diagrams of potential temperature (θ) and specific humidity (q) to more easily illustrate the diurnal cycle of the surface mixed layer. Betts and Ball (1995) used this technique to demonstrate the impact of soil moisture on the mixed-layer thermodynamics. Markowski and Stensrud (1998) used similar plots to investigate the impact of convection on surface-layer thermodynamics.

In general, the atmospheric surface layer is influenced by mixing with the surface from below and by mixing with the free atmosphere above. During the day, surface fluxes tend to warm and moisten the mixed layer, and entrainment fluxes at the top of the PBL tend to warm and dry the mixed layer. At night, radiative cooling stabilizes the boundary layer, and possible dew and fog formation can lead to a cooling and drying of this near-surface layer. An excellent graphical demonstration of this is provided in Fig. 7 of Markowski and Stensrud (1998).

For this study, monthly means of near-surface θ versus q are plotted to further evaluate the reanalysis estimates as well as to highlight month-to-month variability between sites (Figs. 7 and 8). Data from both sites show a general increase in θ and q from January to July, followed by a general decrease in temperature and moisture through December. The annual pattern observed at IDAB shows smooth month-to-month transitions, with increasing potential temperature to specific humidity at a 2-yr mean rate of 1.73 K (g kg−1)−1. The annual cycle at BOIS, however, is much more varied, in part because of the greater sensitivity to precipitation and radiative forcing, and the ratio of potential temperature to specific humidity increases at a much faster rate of 3.54 K (g kg−1)−1. The much thicker vegetation coverage at IDAB moderates the impact of precipitation and radiative forcing.

A comparison between the observations and reanalysis shows that the GR captures the seasonal changes of θ and q at both site locations. The reanalysis estimates fall within one standard deviation of the observations. The standard deviation of θ ranges from 4.5°C in January to 1.8°C in July; q ranges from 0.5 g kg−1 in January to 2.0 g kg−1 in June. The model analysis of θq at BOIS is 1–2 g kg−1 too moist every month, but such high q estimates are likely due to the artificially much higher number of precipitation days from the GR. The model analysis of θq at IDAB is 1°–2°C too cool during the winter months and is 1°–2°C too warm during the summer of 2000.

Monthly mean diurnal cycles were plotted for BOIS and IDAB for 2000 (Fig. 9) and 2001 (Fig. 10). These diagrams reveal the interplay between heat and moisture as a function of seasonality and time of day. Most diurnal cycles progress counterclockwise with the lowest θ value observed near sunrise and the highest θ just prior to sunset. Note that the diurnal cycle at BOIS for December 2001 apparently drops below the saturated specific humidity (qs) because an annual mean qs is plotted.

For a clearer understanding of the diurnal cycle, q and θ values were averaged during the summer months of June, July, and August of 2000 (Fig. 11) and 2001 (Fig. 12). The observed diurnal cycles varied little between 2000 and 2001, but significant differences were found between sites. The diurnal cycle of θq at BOIS during 2000 and 2001 shows a minimum in θ at sunrise at 1200 UTC. By 1400 UTC, the surface layer had warmed 4°C and moistened by 0.5 g kg−1, aided by a shallow but growing surface layer and high latent heat flux. In 2000, LE at BOIS balanced dry air entrainment until about 1500 UTC. In 2001, the surface layer “mixed out” after 1400 UTC. The surface layer then warmed and dried throughout the afternoon until approximately 2200 UTC, just prior to sunset. After 2200 UTC, the surface layer again stabilized. Note how quickly the surface layer decoupled and cooled within a 2-h span between 0000 and 0200 UTC. Nocturnal radiative cooling from the surface and from the atmospheric layer itself began to cool the surface-layer air, and LE continued to moisten the surface layer even after 2200 UTC. After 0600–0800 UTC, nocturnal radiational cooling began to dominate.

The diurnal cycle of θq for IDAB showed a much different pattern from BOIS. The surface layer at IDAB moistened rapidly by nearly 1.5 g kg−1 between 1200 and 1300 UTC. The layer continued to moisten until 1500 to 1600 UTC, 1–2 h longer than observed at BOIS, because of the much lower Bowen ratio at the surface. Dry air entrainment then dominated surface fluxes as the mixed layer warmed and dried until about 2200 UTC. After sunset, the layer becomes decoupled and quickly cooled and moistened, almost as quickly as the rapid rise in moisture just after sunrise. After 0200 UTC, however, the conditions at IDAB began to dry. Contrary to observations from BOIS, the surface layer becomes much cooler and drier at the saturation rate, which is indicative of dew and fog formation. Between 0200 and 1100 UTC, the surface layer dried by approximately 1.5 g kg−1. Note that nearly all of the moisture condensed in the form of dew and fog during the night is rapidly evaporated back into the atmosphere during the early morning. Haugland and Crawford (2002) also noted that a daily double maximum in the diurnal evolution of the moisture field was commonly observed at many of the mesonet sites.

The diurnal cycles of θ and q as analyzed by the GR are statistically similar to those observed. The GR estimates fall within the standard deviation of the observations. The typical variability of the observations of θ and q at 1800 UTC range from 4.6°C and 2.1 g kg−1 at BOIS to 4.5°C and 3.1 g kg−1 at IDAB. These results are consistent with those from Betts et al. (1996), who found that model θ and q estimates compared within ±2°C and ±2 g kg−1 of the observed θ and q, respectively. The diurnal cycle at BOIS is about 1 g kg−1 too moist, but does reflect the overall daily pattern. The diurnal cycle at IDAB during 2000 is 1 g kg−1 too dry, but daily cycles during 2000 and 2001 capture some of the early morning condensation. The GR estimates from both BOIS and IDAB fail to mix the near-surface air properly by 1800 UTC, leaving the analysis much too moist compared to 1200 UTC. The overestimate of LE at IDAB likely contributed to these high afternoon estimates of q. Betts et al. (1996) found the model reanalysis mixes out over too shallow of a layer and underestimates entrainment of drier air from above the PBL.

Differences in the observed thermodynamic cycles between BOIS and IDAB are due, in part, to differences in elevation, as well as differences in their proximity to Gulf of Mexico return moisture flow. BOIS is located over 1 km higher in elevation than IDAB. In addition, BOIS is much drier and has greater sensible heating than does IDAB. Thus, BOIS requires much less time to mix out. Likewise, IDAB has greater latent heating and requires a longer period of mixing. This difference plays a crucial role in the diurnal evolution of the PBL. Furthermore, this difference in PBL depth is cited as playing a key role in the development and propagation of drylines (Bluestein 1993).

The behavior of the surface-layer thermodynamics is better understood by an examination of the near-surface wind speeds. The 2-m wind speeds were averaged during June, July, and August of 2001 at BOIS and IDAB (Fig. 13). Several features of the observed winds are to be noted. First, wind speeds observed at BOIS are nearly double those observed at IDAB. Wind speeds averaged 3.34 m s−1 at BOIS and only 1.56 m s−1 at IDAB. The windier conditions at BOIS contributed to earlier mixing (see Fig. 11) of the boundary layer (1400–1500 UTC) than is observed at IDAB (1500–1600 UTC). During the night, higher wind speeds at BOIS prevented dew formation, which kept q values relatively moist. Very light to calm winds at IDAB allowed for dew formation during much of the night. Second, wind speeds are indicative of the atmospheric-layer stability. Wind speeds at both sites slowed 1–2 m s−1 at 0100 to 0200 UTC as the boundary layer stabilized because of radiative cooling at sunset. Examination of the diurnal cycles of the surface fluxes show that while the Rn, H, and G values became negative at 2300–0000 UTC, LE remained positive until 0100–0200 UTC (yielding a negative BR during this time). This moistened the boundary layers at both IDAB and BOIS. However, the much stronger winds at BOIS kept q well mixed within a deep but stable boundary layer, whereas the much lighter winds at IDAB allowed q within the now much shallower, stable boundary layer to equal, or even exceed, the q values observed during the day. Wind speeds increased 1–2 m s−1 at 1200–1300 UTC because of rapid destabilization at sunrise.

A third feature in the wind field at BOIS (Fig. 13) is evidence of the nocturnal low-level jet (LLJ) as indicated by the local wind maximum between 0300 and 0700 UTC. Estimates of q during this same time show a coinciding increase of 0.4 g kg−1. After 0700 UTC q drops again as winds become lighter and some moisture is condensed as dew or fog. This local nocturnal maximum occurs from May to September, each time accompanied by a significant increase in q. Uccellini and Johnson (1979) identified the LLJ as a major source of heat and moisture transport. Bonner's (1968) climatology of the LLJ found that most were observed during the summer and furthermore, that the BOIS site is located near the geographical maximum. Stensrud (1996) estimated 57% of low-level jets in his study occurred ±2 h of 0600 UTC, consistent with the wind speed maximum in Fig. 13.

Failure of the GR to slow the nocturnal winds at IDAB likely inhibits proper development of the thermodynamic cycle. The high nocturnal winds of the GR prevent near-surface thermodynamic stabilization and a secondary increase in q from LE especially during 2001. Because the GR values are available only four times daily, the LLJ cannot be resolved from the GR data.

6. Surface water budget

The monthly water budget, as defined for a 1D column, is
i1525-7541-5-2-311-e1
where monthly sum totals (Σ) include precipitation (P), evapotranspiration (ET), surface and subterranean runoff (R), and drainage (Dr). The differential soil moisture (ΔSM) term represents the change in water amount in the soil through the depth of the total soil column, in this case, the soil depth is assumed to extend 2 m below the top of the soil surface. The GR uses a two-layer soil model: one layer extends from 0 to 10 cm, and a second layer extends from 10 to 200 cm below ground. Soil water potential was measured at 5, 25, 60, and 75 cm at BOIS and at 5- and 25-cm depths at IDAB. Because no in situ data were available below 75 cm, only the mesonet 5-cm derived soil moisture data were compared against the 0–10-cm estimate from the reanalysis.

Monthly totals (mm) of precipitation (Fig. 14a) and evapotranspiration (Fig. 14b) were estimated at BOIS and IDAB for 2000 and 2001. Monthly statistics of OASIS and reanalysis values for BOIS and IDAB are provided in Tables 2a and 2b, respectively. The observed 2-yr monthly mean of precipitation at BOIS is 33.6 mm, characterizing a relatively dry period on average. The reanalysis monthly mean precipitation is 36.1 mm, approximately 7.4% higher than observed. The GR also recorded a slightly greater monthly standard deviation of 43.4 mm compared to the observed 36.7 mm. An examination of monthly trends (Fig. 14a) shows that GR generally overestimated P at BOIS during February–May of both years, while GR underestimated P at BOIS during the summer months of July and August. The 2-yr observed monthly mean precipitation of 108 mm at IDAB is over 3 times the amount observed at BOIS. The reanalysis monthly mean precipitation for the corresponding years at IDAB is 99.6 mm, approximately 7.8% less than the observed. For IDAB, the GR standard deviation of P at 61.8 mm is much less than the observed 98.1 mm.

Precipitation is among the most difficult model parameters to assess. It has a much weaker spatial correlation than other atmospheric variables, especially during the spring and summer months because of its convective nature (Brotzge and Richardson 2003). Because BOIS and IDAB are near the center of the reanalysis grid boxes, bilinear interpolation was the chosen method for model evaluation. As a consequence, this approach inherently includes smoothed rainfall events “recorded” at each of the four surrounding grid points, 210 km apart. Thus, twice as many rainfall days are recorded by the GR than at the mesonet site location. This likely explains the poor correlation even for monthly precipitation between the GR estimates and mesonet observations (see Tables 2a and 2b).

Monthly sums of evapotranspiration (mm) as observed by the OASIS stations and estimated by GR are shown in Fig. 14b and summarized in Tables 2a and 2b. The sonic estimates of ET at BOIS are within an average 1.8 mm month−1 of observations. Examination of ET at IDAB reveals significant differences between OASIS and GR estimates. Global reanalysis exceeds ET at IDAB by 28.6 mm month−1 relative to OASIS measurements, despite underestimating P by 8.4 mm month−1. A review of monthly estimates in Fig. 14b shows that the GR values exceed observed ET during every month of the 2-yr period. For this study, both the residual and BR methods were used to estimate new ET statistics (see Tables 2a and 2b).

By adjusting the original ET estimates (ETsonic) at BOIS, mean monthly (GR − OASIS) differences increased from 1.8 mm for (ETGR − ETsonic) to −9.9 and −5.3 mm for (ETGR − ETresid) and (ETGR − ETBR), respectively. At IDAB, adjustments to ET lowered the (ETGR − ETsonic) difference from 28.6 to 10.6 mm for (ETGR − ETresid) and to 14.6 mm for (ETGR − ETBR). Thus, as discussed in section 4, adjustments made to the measured ET at BOIS increased (GR − OASIS) differences, while adjustments made to ET at IDAB significantly reduced (GR − OASIS) differences.

Some precipitation is stored within the soil column and is a critical control in determining evapotranspiration rates. The soil column itself becomes an important storage of available water. Soil water content (m3 m−3) is estimated indirectly from the OASIS measurements of soil water potential. For this study, estimated 5-cm soil water content is assumed to be equal to the 0–10-cm-layer average to permit direct comparisons with the GR values. The change in monthly soil water content was then calculated from both OASIS and GR estimates (Figs. 15a and 15b). Differences between OASIS and GR estimates primarily reflect differences in precipitation. However, spatial heterogeneity of soil properties, rooting depths, and vegetation characteristics also contribute to magnifying the differences between the GR values and OASIS measurements.

At a monthly time scale, precipitation minus the storage closure of the near-surface soil layer minus ET equals surface runoff (R) plus subsurface drainage (Dr) (>200 cm depth) plus storage change of the deep-layer soil column (ΔSM10–200cm). Equation (1) is rewritten in the form
i1525-7541-5-2-311-e2
where the left-hand side represents measured quantities at the OASIS sites and the right-hand side represents a residual of the water budget that includes the sum of runoff, drainage, and deep-level (below 10 cm) soil water storage. Positive residual values indicate that excess water is partitioned over runoff, drainage, and storage change in the deep-layer soil column. Negative residual values indicate that ET and near-surface soil moisture changes exceed precipitation, and that 1) a significant moisture delivery is made by the deeper soil column (below 10 cm), and/or 2) a secondary water source (i.e., irrigation) contributes to the water budget at the site.

The water budget residuals using ETsonic are plotted for BOIS (Fig. 16a) and IDAB (Fig. 16b) for 2000 and 2001 and are presented in Tables 2a and 2b as monthly averages for all three combinations of ET. The 2-yr residual sum (ΣP − ΣETsonic − ΔSM0–10cm) as measured at the BOIS site is −176.2 mm; the GR residual summation at BOIS is −156.9 mm. In other words, more water has been extracted from the site at BOIS than has been received by precipitation.

Instrument error introduces some degree of uncertainty into all observations. Rain gauge measurements from the mesonet are limited to a 1% inaccuracy reading, and Alter-style wind screens are used to reduce wind-induced precipitation loss (Brock et al. 1995). Nevertheless, mesonet rain gauges are not heated, and frozen precipitation may be a source of some underestimation.

The soil column (10–200 cm) is a significant contributor to the water balance at BOIS. Evidence of this was found from daily data collected at BOIS during 27 August–6 October 2000, a 40-day period of severe drought conditions. During the period, only 1.8 mm of rainfall was reported, and so the 5-, 25-, 60-, and 75-cm soil water potential estimates remained nearly constant at the wilting point. However, ET totaled 31.4 mm during the same 40-day period, although the amount decreased daily. Thus, a considerable amount of water must be available to the system for ET from the soil column below 75 cm. The dominant plant type at the BOIS site is blue grama, a common prairie grass with roots typically 90–180 cm deep, well below the 75-cm measurement.

Confirmation for the negative right-hand side values in Eq. (2) at BOIS is that 2000 and 2001 are part of a long-term drying trend. Illston and Basara (2003) detected a 5-yr (1997–2001) drying trend across Oklahoma from the soil moisture network. The mean statewide soil moisture over the first 75 cm decreased during each of the 5 yr and rebounded only after above-average precipitation during the latter half of 2002. An examination of 10 yr (1993–2002) of GR estimates for BOIS reflects this exact same trend for the near-surface water balance (Table 3). During the 10-yr period, only GR values during 1997 and 2002 record a surplus (positive) water balance residual. During negative water balance years, deeper level (below 10 cm) soil moisture and/or secondary water sources (below 2 m) must provide the necessary moisture for evapotranspiration. Nevertheless, even when deep-level soil moisture (ΔSM10–200cm) is included in the GR water budget analysis, the 10-yr sum total remains negative at −166.3 mm.

Global reanalysis estimates show IDAB as a water source, yielding a 2-yr water residual total of 214.2 mm (Fig. 15b; Table 2b). Because ET is overestimated by the GR, the water residual estimate is likely underestimated. Water residual estimates using EC data yield a 2-yr total of 1097.8 mm; however, because ETsonic is likely underestimated by the EC system at IDAB, a more reliable water budget residual value is 665.0 mm found using ETresid. Evidence from the BOIS and IDAB comparisons suggests that both aboveground runoff and adequate deep-reservoir water sources are critical to capturing the proper behavior of monthly to seasonal water and energy budget cycles.

7. Conclusions

This study examined the surface radiation, the surface energy, and near-surface water budget data from two OASIS sites located in two distinct climate zones with NCEP–NCAR global reanalysis estimates. The comparison identified several reanalysis deviations from the 2-yr observational data.

A comparison of the net radiation budget between OASIS observations and GR yielded several apparent problems with the reanalysis values. Several model biases were found including an overestimate in incoming shortwave radiation and an underestimate in incoming longwave radiation. As suggested by Betts et al. (1996), the model atmosphere is too transparent and the model produces too few clouds. Furthermore, the model albedo is much too high (4%–8%), and day-to-day variations are limited within the model and do not change as a function of surface wetness or vegetation cover as readily as the observations.

The surface energy budgets as computed from the reanalysis and observations compared favorably well. Mean differences (GR − OASIS) in G were less than 1.5 W m−2, and estimates of H were within 5%. Latent heat estimates from the GR compare well with observations at BOIS using LEsonic, but the GR significantly overestimates LE at IDAB. Three problems likely contribute to the overestimation in LE at IDAB: 1) a 10% overestimation of Rn by the GR, 2) the GR surface fluxes are derived solely from the model physics and are not directly influenced by any assimilated observations, and 3) the OASIS sonic observations most likely underestimate LE, as evidenced by the lack of “closure” of the surface energy budget. Despite the suspected observation problems, no corrective measures uniformly improved LE differences between the reanalysis and measurements at both sites. It is interesting to note that the largest errors in LE occurred at IDAB, which is more vegetated and has higher soil moisture than BOIS. More extensive work is needed to determine the precise cause of this overestimation in LE.

Mean monthly estimates of the thermodynamic cycles reflected the distinct seasonal climatologies associated with each site, and seasonal trends were well analyzed. However, diurnal cycles of θq by the GR failed to mix the near-surface PBL properly, which led to an overestimate in the rate of increase of q between 1200 and 1800 UTC. The GR also failed to slow the nocturnal near-surface winds at IDAB. Finally, observations indicated evidence of the nocturnal LLJ impacting the mean summertime diurnal θq cycle.

A comparison of the water budget data from the reanalysis and observations yielded several interesting results. Difficulty with observations and interpolation limits to an unknown extent accurate assessment of P and ET. Observations of frozen precipitation are likely underestimated by the mesonet because the rain gauge is unheated and because of wind-induced effects. Bilinear interpolation of the reanalysis values to a single point tends to overestimate the number of rainfall days. The high spatial variability of rainfall, particularly because of its convective nature in Oklahoma, further inhibits a reliable comparison between model and observed data.

Sum totals of P, ET, and the change in soil water from both NCEP–NCAR reanalysis and OASIS measurements yield a positive water residual (ΣP − ΣET − ΔSM0–10cm) at IDAB, but a negative water residual at BOIS. However, the negative near-surface water residual is compensated by the significant water source available in the deep soil column, particularly below 1 m. The root zone of the dominant vegetation type lies between 1 and 2 m, and significant ET is observed even after the soil column between 0 and 0.75 m has reached its wilting point. The occurrence of a negative near-surface water balance seems to be indicative of a longer-term (5+ yr) drying trend for which the annual ET exceeds (P − ΔSM0–10) and leaving vegetation to rely on soil moisture storage from deeper levels. The fact that both observations and model analysis indicate this long-term drying trend at BOIS is encouraging for the detection of other long-term hydrological changes.

Overall, the generally good agreement between the NCEP–NCAR reanalysis estimates and mesonet–OASIS observations lends support to the hypothesis posed by Shuttleworth (1991) that long-term, single site data can be used to evaluate coarse-resolution model analysis. The surface energy and water budgets as measured at the BOIS and IDAB sites represent regional radiation, surface flux, and soil moisture behavior ideal for model analysis. The next step will be to use this database as a foundation from which to begin validation of GCM and regional-scale numerical models. Future work will also include comparisons of mesonet–OASIS data with NCEP/AMIP-II DOE reanalysis and the NCEP regional reanalysis estimates to document improvements in reanalysis efforts.

Acknowledgments

The author would like to thank Dr. Kenneth Mitchell and three anonymous reviewers for their helpful comments in greatly improving this manuscript. The author appreciates the insightful discussions of Derek Arndt, Robert Hale, Matthew Haugland, and Janet Martinez, the support of CAPS, and the technical support provided by the Oklahoma Climatological Survey for their professional assistance in maintenance, ingest, and quality control of the mesonet data. Oklahoma Mesonet data are provided with courtesy of the Oklahoma Mesonet, a cooperative venture between Oklahoma State University and the University of Oklahoma. A special thanks goes to the taxpayers of Oklahoma for their continued support and funding of the Oklahoma Mesonet. NCEP–NCAR reanalysis values provided by the NOAA–CIRES Climate Diagnos-tics Center, Boulder, Colorado, from their Web site athttp://www.cdc.noaa.gov/.

REFERENCES

  • Basara, J. B., , and Crawford T. M. , 2000: Improved installation procedures for deep layer soil moisture measurements. J. Atmos. Oceanic Technol., 17 , 879884.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Betts, A. K., 1984: Boundary layer thermodynamics of a high plains severe storm. Mon. Wea. Rev., 112 , 21992211.

  • Betts, A. K., 1992: FIFE atmospheric boundary layer budget methods. J. Geophys. Res., 97 , 1852318531.

  • Betts, A. K., , and Ball J. H. , 1995: The FIFE surface diurnal cycle climate. J. Geophys. Res., 100 , 2567925693.

  • Betts, A. K., , and Ball J. H. , 1998: FIFE surface climate and site-average dataset 1987–89. J. Atmos. Sci., 55 , 10911108.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Betts, A. K., , Hong S-Y. , , and Pan H-L. , 1996: Comparison of NCEP–NCAR reanalysis with 1987 FIFE data. Mon. Wea. Rev., 124 , 14801498.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Bluestein, H. B., 1993: Observations and Theory of Weather Systems. Vol. 2, Synoptic-Dynamic Meteorology in Midlatitudes. Oxford University Press, 594 pp.

    • Search Google Scholar
    • Export Citation
  • Bonner, W. D., 1968: Climatology of the low level jet. Mon. Wea. Rev., 96 , 833850.

  • Brock, F. V., , Crawford K. C. , , Elliott R. L. , , Cuperus G. W. , , Stadler S. J. , , Johnson H. L. , , and Eilts M. D. , 1995: The Oklahoma Mesonet: A technical overview. J. Atmos. Oceanic Technol., 12 , 519.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Crawford K. C. , 2000: Estimating sensible heat flux from the Oklahoma Mesonet. J. Appl. Meteor., 39 , 102116.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Duchon C. E. , 2000: A field comparison among a domeless net radiometer, two four-component net radiometers, and a domed net radiometer. J. Atmos. Oceanic Technol., 17 , 15691582.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Weber D. , 2002: Land-surface scheme validation using the Oklahoma Atmospheric Surface-layer Instrumentation System (OASIS) and Oklahoma Mesonet data: Preliminary results. Meteor. Atmos. Phys., 80 , 189206.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Crawford K. C. , 2003: Examination of the surface energy budget: A comparison of eddy correlation and Bowen ratio measurement systems. J. Hydrometeor., 4 , 160178.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , and Richardson S. J. , 2003: Spatial and temporal correlation among Oklahoma Mesonet and OASIS surface-layer measurements. J. Appl. Meteor., 42 , 519.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , Richardson S. J. , , Crawford K. C. , , Horst T. W. , , Brock F. V. , , Humes K. S. , , Sorbjan Z. , , and Elliott R. L. , 1999: The Oklahoma Atmospheric Surface-Layer Instrumentation System (OASIS) Project. Preprints, 13th Symp. on Boundary Layers and Turbulence, Dallas, TX, Amer. Meteor. Soc., 612–615.

    • Search Google Scholar
    • Export Citation
  • Brotzge, J. A., , Richardson S. J. , , and McAloon C. M. , 2001: The OASIS Project network for monitoring the surface energy budget. Preprints, 11th Symp. on Meteorological Observations and Instrumentation, Albuquerque, NM, Amer. Meteor. Soc., 66–71.

    • Search Google Scholar
    • Export Citation
  • Haugland, M. J., , and Crawford K. C. , 2002: The diurnal cycle of dewpoint across Oklahoma's winter wheat belt. Preprints, 13th Symp. on Global Climate Change and Climate Variations, Orlando, FL, Amer. Meteor. Soc., 254–256.

    • Search Google Scholar
    • Export Citation
  • Illston, B. G., , and Basara J. B. , 2003: Climatology of soil moisture variables using the Oklahoma Mesonet. Preprints, 17th Conf. on Hydrology, Long Beach, CA, Amer. Meteor. Soc., CD-ROM, P1.13.

    • Search Google Scholar
    • Export Citation
  • Kalnay, E., and Coauthors, 1996: The NCEP/NCAR 40-Year Reanalysis Project. Bull. Amer. Meteor. Soc., 77 , 437471.

  • Kanamitsu, M., , Ebisuzaki W. , , Woollen J. , , Yang S-K. , , Hnilo J. J. , , Fiorino M. , , and Potter G. L. , 2002: NCEP–DOE AMIP-II Reanalysis (R-2). Bull. Amer. Meteor. Soc., 83 , 16311643.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Kistler, R., and Coauthors, 2001: The NCEP–NCAR 50-year reanalysis: Monthly means CD-ROM and documentation. Bull. Amer. Meteor. Soc., 82 , 247268.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Markowski, P. M., , and Stensrud D. J. , 1998: Mean monthly diurnal cycles observed with PRE-STORM surface data. J. Climate, 11 , 29953009.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Roads, J., , and Betts A. , 2000: NCEP–NCAR and ECMWF reanalysis surface water and energy budgets for the Mississippi river basin. J. Hydrometeor., 1 , 8894.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Sellers, P. J., , Hall F. G. , , Asrar G. , , Strebel D. E. , , and Murphy R. E. , 1988: The First ISLSCP Field Experiment (FIFE). Bull. Amer. Meteor. Soc., 69 , 2227.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shafer, M. A., , Fiebrich C. A. , , Arndt D. S. , , Fridrickson S. E. , , and Hughes T. W. , 2000: Quality assurance procedures in the Oklahoma Mesonetwork. J. Atmos. Oceanic Technol., 17 , 474494.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Shuttleworth, W. J., 1991: Insight from large-scale observational studies of land/atmosphere interactions. Surv. Geophys., 12 , 330.

  • Stensrud, D. J., 1996: Importance of low-level jets to climate: A review. J. Climate, 9 , 16981711.

  • Stokes, G. M., , and Schwartz S. E. , 1994: The Atmospheric Radiation Measurement (ARM) Program: Programmatic background and design of the Cloud and Radiation Testbed. Bull. Amer. Meteor. Soc., 75 , 12011221.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Twine, T. E., , Kustas W. P. , , Norman J. M. , , Cook D. R. , , Houser P. R. , , Meyers T. P. , , Prueger J. H. , , and Starks P. J. , 2000: Correcting eddy covariance flux underestimates over a grassland. Agric. For. Meteor., 103 , 279300.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Uccellini, L. W., , and Johnson D. R. , 1979: The coupling of upper and lower tropospheric jet streaks and implications for the development of severe convective storms. Mon. Wea. Rev., 107 , 682703.

    • Crossref
    • Search Google Scholar
    • Export Citation
  • Willmott, C. J., 1982: Some comments on the evaluation of model performance. Bull. Amer. Meteor. Soc., 63 , 13091313.

Fig. 1.
Fig. 1.

Oklahoma state map displaying the mesonet sites located at Boise City and Idabel

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 2.
Fig. 2.

Monthly averages during 2000 of shortwave radiation (SW; W m−2) for (a) Boise City and (b) Idabel. Monthly averages of longwave radiation (LW; W m−2) for (c) Boise City and (d) Idabel during 2000

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 3.
Fig. 3.

Daily averages of albedo (%) as estimated from OASIS and GR from 1 Jan 2000 to 31 Dec 2001 for (a) Boise City and (b) Idabel

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 4.
Fig. 4.

Daily estimates of net radiation (W m−2) as observed from the CNR1 and NR-Lite radiometers. Data were collected from Boise City (Bois) and Idabel (Idab) from 1 Jan to 31 Dec 2000

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 5.
Fig. 5.

Data collected during 1 Jan 2000–31 Dec 2001 at the Boise City mesonet site (OASIS) is compared against the NCEP–NCAR global reanalysis (GR). Monthly means (W m−2) of (a) net radiation (Rn), (b) ground heat flux (G), (c) latent heat flux (LE), and (d) sensible heat flux (H) are shown. Note that LE and H are EC measurements

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 6.
Fig. 6.

As in Fig. 5, but for the Idabel mesonet site

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 7.
Fig. 7.

Monthly averages of near-surface specific humidity (q; g kg−1) and potential temperature (θ; K). Data collected during 2000 at the (a) Boise City and (b) Idabel mesonet sites

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 8.
Fig. 8.

As in Fig. 7, but for 2001

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 9.
Fig. 9.

Monthly diurnal averages of specific humidity (q; g kg−1) and potential temperature (θ; K). Data collected during 2000 at the (a) Boise City and (b) Idabel mesonet sites

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 10.
Fig. 10.

As in Fig. 9, but for 2001

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 11.
Fig. 11.

A scatterplot of 92-day averages of specific humidity (q; g kg−1) and potential temperature (θ; K) plotted as a function of time of day (UTC). Data collected during Jun, Jul, and Aug of 2000 at the (a) Boise City and (b) Idabel mesonet sites

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 12.
Fig. 12.

As in Fig. 11, but for 2001

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 13.
Fig. 13.

The summer averages of wind speed (m s−1) for Boise City (Bois) and Idabel (Idab) plotted as a function of time of day (UTC). Data collected during Jun, Jul, and Aug of 2001

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 14.
Fig. 14.

Monthly estimates (mm) of (a) precipitation (P) and (b) evapotranspiration (ET). Data collected at the Boise City and Idabel sites during 1 Jan 2000–31 Dec 2001

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 15.
Fig. 15.

Monthly estimates (mm) of the change in the amount of water in the near-surface (0–10 cm) soil layer. Data collected at the (a) Boise City and (b) Idabel sites during 1 Jan 2000–31 Dec 2001

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Fig. 16.
Fig. 16.

Monthly estimates of the near-surface water residual (mm), defined as the monthly sum totals of precipitation (ΣP), evapotranspiration (ΣET), and the monthly change in the near-surface soil moisture (ΔSM0–10cm). Data collected at (a) Boise City and (b) Idabel during 1 Jan 2000–31 Dec 2001

Citation: Journal of Hydrometeorology 5, 2; 10.1175/1525-7541(2004)005<0311:ATCOTS>2.0.CO;2

Table 1.

Summary of site specifications

Table 1.

Table 2a. Statistics of the monthly OASIS observations and global reanalysis estimates for 2000–01 at Boise City. The observed (O) and model predicted (P) mean (X ), standard deviation (σ), mean bias error (MBE), standard deviation of the MBE (σO−P ), root-mean-square error (rmse), mean absolute error (MAE), and correlation (RO,P) are presented. Statistics are defined as given by Willmott (1982)

i1525-7541-5-2-311-t201

Table 2b. Same as Table 2a, but for Idabel

i1525-7541-5-2-311-t202
Table 3.

Mean annual, near-surface water residual, ΣP − ΣET − ΔSM 0–10cm (mm), for the NCEP–NCAR reanalysis values for 1993–2002 at Boise City

Table 3.
Save