Leo J. Fritschen

Bowen ratio, eddy correlation, and portable chamber measurements of sensible and latent heat flux over irrigated spring wheat*

Measurements of the latent (LE) and sensible (H) heat flux density in the atmospheric boundary layer of irrigated crops have applications for understanding processes in agriculture and meteorology and for water management. The objective of this research was to compare measured Bowen ratios and calculated LE and H from four Bowen ratio systems (BR1–BR4) of different design with each other and with fluxes measured by three sets of eddy correlation instrumentation (H and LE) and a portable chamber (LE). Measurements were made on 9 and 10 April 1989 in an irrigated wheat field at the Maricopa Agricultural Center near Maricopa, Arizona. The Bowen ratio system designs varied in terms of temperature and humidity sensors and measurement arm movement. Bowen ratios were lower (more negative) on 9 April for all of the systems. The range of the four Bowen ratios was greatest in the early morning and late afternoon (±0.1) and least around noon (+0.02). Measured net radiation and soil heat flux density were constant in the Bowen ratio LE calculations. The range of daytime LE from the four systems on 9 April and from three on 10 April was 11% and 1% of the mean LE, respectively. The three eddy correlation H measurements were essentially equal to each other. The average eddy correlation H was 82% and 69% of Bowen ratio H on 9 and 10 April, respectively whilst the eddy correlation LE was 77% and 67% of Bowen ratio LE on the two days. On 9 and 10 April, portable chamber LE was greater than Bowen ratio LE during periods of southerly winds owing to the effect of advected energy to the southern field edge where chamber measurements were made. On 10 April, portable chamber LE was 125% of Bowen ratio LE. This study has shown that: (1) Bowen ratios from instrumentation of different designs were similar; (2) eddy correlation H from three systems were similar to each other and were slightly less than Bowen ratio H; (3) eddy correlation LE was consistently and significantly less than Bowen ratio LE; (4) measurements of portable chamber LE on the edge of a field were affected by surrounding conditions.

Net and solar radiation relations over irrigated field crops

Abstract Many of the meteorological methods used to estimate evapotranspiration from crop surfaces require net radiation information. Since net radiation data are not generally available and solar radiation is measured at several locations throughout the world, it would be desirable to estimate net radiation from solar radiation. Consequently, a relation was sought between solar radiation and net radiation measured over irrigated field crops of alfalfa, barley, wheat, oats, cotton, and sorghum. Data collected under field conditions were analyzed by linear regression techniques. Standard deviation from regression was 0.02 ly/min for individual days and for individual crops. The regression equations changed from day to day; therefore, were of little value for estimation purposes. Seasonal data for individual crops were pooled and analyzed by linear regression. The resulting standard errors were about twice as large as those for individual days, ranging from 0.02 to 0.05 ly/min. When the data for all crops and all days were pooled and analyzed by linear regression, the resulting standard error was 0.06 ly/min. Thus, estimating net radiation for any of the crops from solar radiation would result in a standard error of 38 ly, or approximately 10%, for a 12-hour day having 649 ly incoming radiation. The regression equation may be solved using either hourly or daily solar radiation data. In the absence of net radiation data either the pooled regression or the individual crop regressions may provide sufficiently accurate estimates for some applications—for example, estimating evapotranspiration to be used in the design of irrigation projects. However, the error appears to be too large where daily evapotranspiration results are required. Therefore, measurements of net radiation are still desirable. Since reflected solar radiation is one of the components of net radiation, a better relation might be expected between net radiation and net solar radiation incoming minus reflected solar radiation). The inclusion of the reflected solar radiation data does not reduce the standard errors. Calculation of albedos of the various surfaces from daily totals of incoming and reflected solar radiation indicated a range of 0.14 over wet soil to 0.24 over dry soil, and up to 0.27 over crop surfaces. The average albedo of crop surfaces was 0.24. Row crops tended to have lower albedos until the maximum canopy was developed; then the albedos were similar to those of continuous crops such as alfalfa. Broadleaf plants tended to have larger reflections than grasses. The crops studied rank in order of increasing albedos as sorghum, wheat, barley, oats, cotton, and alfalfa. Increasing the surface albedo may result in water conservation by reducing the amount of energy absorbed which could be used in evaporation. It appears that the greatest effect could be achieved by increasing the albedo of wet bare soil.

Surface Energy and Radiation Balance Systems: General Description and Improvements

Abstract Surface evaluation of sensible and latent heat flux densities and the components of the radiation balance were desired for various vegetative surfaces during the ASCOT84 experiment to compare with modeled results and to relate these values to drainage winds. Five battery operated data systems equipped with sensors to determine the above values were operated for 105 station days during the ASCOT84 experiment. The Bowen ratio energy balance technique was used to partition the available energy into the sensible and latent heat flux densities. A description of the sensors and battery operated equipment used to collect and process the data is presented. In addition, improvements and modifications made since the 1984 experiments are given. Details of calculations of soil heat flow at the surface and an alternate method to calculate sensible and latent heat flux densities are provided.

Construction and Evaluation of a Miniature Net Radiometer

Abstract The miniature net radiometer was constructed by enclosing a blackened thermal transducer within four polystyrene radiation windows. The outer two radiation windows were hemispherically shaped (one-inch radius) so that the response to varying angles of incidence would follow the cosine law and precipitation runoff would be facilitated. The miniature net radiometer was calibrated in the laboratory in a chamber capable of simulating both solar and terrestrial radiation, and was found to be equally sensitive to both. The response to varying angles of incidence follows the cosine law. The output is not affected by wind or ambient temperature, and the size of the instrument enables its use to measure net radiation close to a surface or within a, plant canopy without appreciably altering the net radiation by shading.

Miniature Net Radiometer Improvements

Abstract An improved net radiometer was designed with increased sensitivity to match a ±5-mv potentiometer used in a multipoint recording system. Its sensitivity is sufficient to give a full-scale range of ±1.4 ly min−1, with a recording sensitivity and accuracy of 0.003 ly min−1. The radiometer sensor consists of a 22-junction manganin-constantan thermopile with compensating thermistor embedded in epoxy resin. Effects of ambient temperature from 11 to 54C are virtually eliminated by means of a circuit comprised of a shunt thermistor and micronminiature rheostat. The blackened surfaces of the transducer are shielded against the effects of wind by thin polyethylene hemispheres. The improvement also include uniform sensitivity among sensors, better spectral response, and relative ~ of construction.

Deep Valley Radiation and Surface Energy Budget Microclimates. Part II: Energy Budget

Abstract Surface energy budget measurements were made concurrently at five sites located on the valley floor, sidewalls and ridgetop of Colorados 650-m deep Brush Creek Valley (39°32′N, 108°24′W) on the nearly clear day of 25 September 1984 using the Bowen ratio energy budget technique. Daily average surface heat flux values for a natural sagebrush ecosystem on the floor of the semiarid valley included an input of 109 W m−2 net all-wave radiation and 15 W m−2 ground heat flux, and a loss of 48 W m−2 latent heat flux and 76 W m−2 sensible heat flux. Significant differences in instantaneous, daily, and daytime fluxes occurred from site to site as a function of slope aspect and inclination angles and surface properties, including vegetation cover and soil moisture. Strong contrasts in instantaneous latent and sensible heat fluxes occurred between the opposing northeast-and southwest-facing sidewalls of the valley as solar insolation varied through the course of the day and as shadows propagated across the v...

Modeling a coniferous forest canopy

Abstract A naturally regenerated Douglas fir stand was modeled by assuming the canopy to be a composite of individual model crowns. The model crown was developed by normalizing total needle surface area per branch position with respect to the maximum foliage surface area and position in the crown. Because of the similarity of foliage distribution between the trees which were examined, it was assumed that all trees in the forest could be represented by this double normalized model crown. The model forest canopy was compiled by scaling model crowns to length of live crown and surface area representative of trees of the respective size classes. The composite forest canopy was obtained by multiplying the number of trees in each size class by its respective surface area (the scaled model crown) and then summing these products. The cumulative plant surface area obtained from the composite canopy was compared with wind profiles obtained within the forest. The results indicate that the plant surface area distributions, as obtained by this method, are valid in that they satisfactorily account for the shape of the wind profiles within the forest. It is suggested that this method of plant surface area determination may be satisfactory for many other types of vegetation.

Characterization of Boundary Conditions Affecting Forest Environmental Phenomena

Analyses of environmental factors at forested sites generally are not applicable to other sites because the sites have not been described adequately. To be more useful to other researchers, a description of the forest site should convey a physical image of the forest. Data on wind, temperature, humidity, radiation and vegetation have been presented to illustrate the variability into and within various forests. Methods of inverting environmental data to obtain information on the vertical distribution of the biomass have been cited. A strategy for describing coniferous forests has been presented.

Net radiation, sensible and latent heat flux densities on slopes computed by the energy balance method

Energy balance components obtained over five grass-covered sloping surfaces near Manhattan, KS, using the Bowen ratio energy balance technique with the instruments mounted horizontally were compared with calculated values when the instruments were mounted parallel to the surfaces. Hourly values of the components changed when the instruments were parallel to the surfaces. The changes were larger at low solar angles (spring and fall) and on steeper slopes. An area average of daylight totals, assuming that all aspects were equally represented, changed only 0.1% on June 6 and 2.3% on October 11. The calculations, extended to steeper slopes, indicated small changes in the daylight totals for slopes of less than 10 deg.

A Sensitive Cup-Type Anemometer

Abstract A sensitive cup-type anemometer was developed which is mechanically and electrically simple. The characteristics include a starting speed of 6 cm sec−1, distance constant (63%) of 90 cm, and a linear response.