Eddy Flux Measurements Research Articles

GOES surface insolation to estimate wetlands evapotranspiration

Incoming solar radiation derived from GOES-8 satellite observations, in combination with local meteorological measurements, were used to model evapotranspiration from a wetland. The wetland experiment was conducted in the Paynes Prairie Preserve, North Central Florida during a growing season characterized by significant convective activity. The satellite solar radiation measurements generally agreed with pyranometer data gathered at the site. The satellite net radiation estimates were in good agreement with the 30-min averages of measured net radiometer data. Satellite derived net radiation estimates were used in the Penman–Monteith and Priestley–Taylor models to calculate evapotranspiration. The calculated instantaneous evaporative fluxes were in good agreement with 30-min average ground-based eddy correlation system measurements. The daily averages of modeled evapotranspiration were in very good agreement ( r 2=0.90) with reference eddy flux measurements.

Two years of continuous CO 2 eddy-flux measurements over a Danish beech forest

As part of the EUROFLUX network a long-term monitoring station for fluxes of CO 2 and water vapour has been established in an 80-year old beech forest in Denmark. The station has been in continuous operation since June 1996 and will be so at least to the end of 2002. A primary goal of EUROFLUX is to combine flux measurements on a continuous multi-year time basis with ecological processes interpretation and modeling. The station consists of a 57 m high mast with conventional meteorological profile instrumentation and one level of eddy-flux measurements. Ancillary measurements such as soil respiration, soil moisture, soil temperature, leaf surface temperature and leaf area index are also made. Results from the first 2 years of measurements are described. The observed diurnal and seasonal variation in the fluxes are discussed and the monthly and annual sums of ecosystem exchange are contrasted between the 2 years. The first year (1 June 1996–31 May 1997) had a net uptake of 223 g C m −2 whereas the net uptake during the second year (1 June 1997–31 May 1998) was only 144 g C m −2. It was found that the difference was mainly the result of higher respiration in the second year caused by a very warm summer of 1997 and an average soil and air temperature, respectively, 1.2 and 1.6°C higher than the first year. The ecosystem photosynthetic assimilation was slightly higher during the second year, mainly caused by increased incoming radiation.

Validation of eddy flux measurements above the understorey of a pine forest

Measurements of turbulent exchange in the understorey of a forest canopy are necessary to understand and model both the functioning of the lower stratum and its contribution to turbulent exchange at the whole canopy scale. Eddy covariance measurements of eddy fluxes just above the floor of a forest canopy must be thoroughly validated, given the particular conditions prevailing there. Our objective is two-fold: (i) check the overall quality of such eddy flux measurements through the analysis of the understorey energy balance closure and (ii) define quality criteria for each half-hourly sample, based on the residual term of the energy balance. A subset of the EUROFLUX data base, collected within a pine forest canopy in south-west France, was used for this purpose. During this experiment, all heat storage terms were carefully measured, which allowed the closure of the understorey energy balance to be rigorously tested. As in most experiments storage term measurements are not available, we also developed a method to estimate them, in order to apply the above-mentioned data selection method. The energy balance closure was found to be quite satisfactory (the slope of the sum of eddy fluxes and storage terms versus transmitted net radiation is 0.99, the intercept is less than 1 W m −2, r 2 is 0.94, there is no deviation from a linear trend). The data selection procedure allows a fair description of the daily and day-to-day variation of turbulent fluxes while rejecting the most dubious data, whether experimental or estimated storage terms are used. This analysis proves the validity of eddy flux measurements in the lower part of the forest and offers tools for flux data selection, depending on the type of studies such data are intended for.

The carbon balance of a young Beech forest

Abstract1. We present measurements of CO2 fluxes over 2 years above and within a young Beech stand in the east of France. This site is part of the Euroflux network set up to monitor fluxes over representative European forests.2. The net ecosystem carbon (C) exchange was derived from continuous eddy flux measurements. Major components of the total flux (i.e. soil and above‐ground biomass respiration and assimilation of leafy branches) were measured independently using chambers. The main C stocks (i.e. root, stem and branch biomass) were also quantified.3. Daily minima of CO2 flux were typically around −20 µmol CO2 m−2 s−1 during the period of full leaf expansion, while night‐time ecosystem respiration varied between 5 and 15 µmol CO2 m−2 s−1. The seasonal pattern of net ecosystem assimilation was very close to that of net assimilation at the single branch scale. The seasonal variation of net ecosystem exchange was closely related to leaf expansion and soil water content during the dry year of 1996.4. Measurements of ecosystem respiration (eddy flux) were corrected for CO2 storage within the stand. This C flux showed a seasonal pattern, the maximum rates (4–7 g C m−2 day−1) occurring in spring and summer, and appeared to be correlated with soil temperature. Temporal variation of soil respiration showed the same pattern, and effects of both temperature and soil drying were found. Annual soil respiration was ≈ 70% of ecosystem respiration. Root respiration was 60% of the total below‐ground respiration.5. Annual net C exchange was −218 and −257 g C m−2 in 1996 and 1997, respectively, corresponding to net C uptake by the forest. These values are much lower than the annual biomass increment (stems and large roots) of the stand: 427 and 471 g C m−2 year−1, respectively. The difference may be explained by a release of CO2 from the decomposition of woody debris.6. Ecosystem C loss by respiration was 800–1000 g C m−2 year−1. Gross C gain was 1000–1300 g C m−2 year−1. Ecosystem respiration therefore played a major role in the annual C balance of this forest.

Short-term environmental controls on carbon dioxide flux in a boreal coniferous forest: model computation compared with measurements by eddy covariance

CO 2 eddy flux measurements were coupled with detailed microclimate and soil measurements made in an undisturbed boreal pine forest in the late growing season of 1998 in order to understand the effects of short-term variation in weather on daily canopy CO 2 flux. To test whether our present knowledge of leaf-level processes is sufficient to understand stand-level CO 2 exchange, a parameterised forest carbon flux model was presented in which the scaling of CO 2 exchange from leaf to canopy is achieved by integrating mechanistic models for physiological processes (photosynthesis, stomatal conductance and respirations of the ecosystem components) and micrometeorological ones (radiative transfer, turbulent transfer and surface energy exchanges). Testing of the model against 25 day flux measurements showed that the predicted day-time forest ecosystem CO 2 fluxes matched well with those measured by the eddy covariance method but gave a poor estimate for the night-time CO 2 flux. Based on 30 min average eddy flux data during the 25 day period, the mean net forest system CO 2 uptake was 5.76 μmol CO 2 m −2 ground s −1. The maximum rate was 15 μmol m −2 s −1, and the mean and maximum night-time rates were 2.2 and 4.8 μmol m −2 s −1, respectively. Maximum CO 2 storage in the air column was 7.3 μmol m −2 s −1 in the early morning and 3.6 μmol m −2 s −1 during the night. The daily total net CO 2 exchange for the forest ecosystem varied from 0.06 to 0.47 mol m −2 s −1 in the day-time and from 0.04 to 0.13 mol m −2 s −1 in the night-time. Day-time CO 2 uptake in the forest ecosystem was significantly coupled with changes in photosynthetic photon flux density and modified significantly by air temperature and the canopy-air vapour pressure deficit, but it was less dependent on soil water content on most days during the growing season. These results are in broad agreement with expectations based on the biochemistry of leaf gas exchange and the transfer of radiation through a canopy. The variance of about 60% in the rate of CO 2 exchange at night could be explained by changes in air temperature and other factors related to the dynamics of turbulence mixing.

Simulated evapotranspiration from the Norunda forest stand during the growing season of a dry year

A SVAT model was used to simulate evaporation from the Norunda forest stand during the dry growing season of 1994. Daily mean values of meteorological data were used as input and compared with actual data on soil moisture, transpiration and total evaporation. The soil moisture data of TDR were used as a basis for defining combinations of parameters values that resulted in acceptable agreement. Different threshold for reduction of transpiration during the dry spell in July resulted in small changes in soil moisture dynamics. Using also sap-flow measurements the final estimated value of −150hPa as the threshold value below which water uptake is reduced was found in combination with a flexibility coefficient of 0.7 for compensatory water uptake between soil horizons. The agreement with daily rates of evaporation was good, and coefficients of determination of 0.44, 0.78 and 0.89 were obtained when comparing with all days of eddy-flux measurements, days without interception losses with eddy-flux measurements and sap-flow measurements of canopy transpiration, respectively. The total evaporation from May to the end of October was estimated to be 314mm, of which interception losses, canopy transpiration and soil evaporation accounted for 70, 181 and 63mm respectively. However, the eddy-flux measurements indicated that the total sum had been underestimated, and the sap-flow measurements suggested that canopy transpiration had been slightly overestimated. The high correspondence between simulated and measured fluxes was surprising considering the heterogeneity involved and simplistic parameterization of the model. In future studies the resolution of input data to the model will be higher, and spatial variation will be explicity represented when simulating evaporation from this ecosystem.

Assessment of annual carbon exchange in a water‐stressed Pinus radiata plantation: an analysis based on eddy covariance measurements and an integrated biophysical model

AbstractWe used a combination of eddy flux, chamber and environmental measurements with an integrated suite of models to analyse the seasonality of net ecosystem carbon uptake (FCO2) in an 8‐year‐old, closed canopy Pinus radiata D.Don plantation in New Zealand (42°52′ S, 172°45′ E). The analyses utilized a biochemically based, big‐leaf model of tree canopy photosynthesis (Ac), coupled to multiplicative environmental‐constraint functions of canopy stomatal conductance (Gc) via environmental measurements, a temperature‐dependent model of ecosystem respiration (Reco), and a soil water balance model. Available root zone water storage capacity at the measurement site is limited to about 50 mm for the very stony soil, and annual precipitation is only 660 mm, distributed evenly throughout the year. Accordingly the site is prone to soil moisture deficit throughout the summer. G c and Ac obtained maximum rates early in the growing season when plentiful soil water supply was associated with sufficient quantum irradiance (Qabs), and moderate air saturation deficit (D) and temperature (T). From late spring onwards, soil water deficit and D confined Gc and Ac congruously, which together with the solely temperature dependency of Reco resulted in the pronounced seasonality in FCO2. Reflecting a light‐limitation of Ac in the closed canopy, modelled annual carbon (C) uptake was most sensitive to changes in Qabs. However, Qabs did not vary significantly between years, and changes in annual FCO2 were mostly due to variability in summer rainfall and D. Annual C‐uptake of the forest was 717 g C m–2 in a near‐average rainfall year, exceeding by one third the net uptake in a year with 20% less than average rainfall (515 g C m–2).

Analysis of carbon and water fluxes from the NOPEX boreal forest: comment

In a recently published paper on simulation and analyses of water and carbon fluxes at the NOPEX central site, a modeling exercise indicated a rather strong sink of carbon for the studied forest stand and year (1994). At the 100-m high tower, the eddy-flux measurements stretch out to the present day. They show that the surrounding forest area is a rather weak source of carbon. This note clarifies the previously published numbers in view of several new findings at the site since the paper preparation (1995) up to the publishing/present date (1998/1999). The key issues that could not be adopted three years ago, were (i) footprint analyses of the eddy-covariance measurements, (ii) correction of eddy-flux data for night-time, and (iii) information on carbon flux from soil. These complementary findings address the previous uncertainties in carbon budget assessment and model parameterization and suggest improvements to model application and better match the measured data.

Simulating forest productivity and surface-atmosphere carbon exchange in the BOREAS study region.

A process-based, general ecosystem model (BIOME-BGC) was used to simulate daily gross primary production, maintenance and heterotrophic respiration, net primary production and net ecosystem carbon exchange of boreal aspen, jack pine and black spruce stands. Model simulations of daily net carbon exchange of the ecosystem (NEE) explained 51.7% (SE = 1.32 g C m(-2) day(-1)) of the variance in daily NEE derived from stand eddy flux measurements of CO(2) during 1994. Differences between measured and simulated results were attributed to several factors including difficulties associated with measuring nighttime CO(2) fluxes and model assumptions of site homogeneity. However, comparisons between simulations and field data improved markedly at coarser time-scales. Model simulations explained 66.1% (SE = 0.97 g C m(-2) day(-1)) of the variance in measured NEE when 5-day means of daily results were compared. Annual simulations of aboveground net primary production ranged from 0.6-2.4 Mg C ha(-1) year(-1) and were concurrent with results derived from tree increment core measurements and allometric equations. Model simulations showed that all of the sites were net sinks (0.1-4.1 Mg C ha(-1) year(-1)) of atmospheric carbon for 1994. Older conifer stands showed narrow margins between uptake of carbon by net photosynthesis and carbon release through respiration. Younger stands were more productive than older stands, primarily because of lower maintenance respiration costs. However, all sites appeared to be less productive than temperate forests. Productivity simulations were strongly linked to stand morphology and site conditions. Old jack pine and aspen stands showed decreased productivity in response to simulated low soil water contents near the end of the 1994 growing season. Compared with the aspen stand, the jack pine stand appeared better adapted to conserve soil water through lower daily evapotranspiration losses but also exhibited a narrower margin between daily net photosynthesis and respiration. Stands subjected to water stress during the growing season may exist on the edge between being annual sources or sinks for atmospheric carbon.

Oceanic heat fluxes under thin sea ice in Saroma-ko Lagoon, Hokkaido, Japan

Copyright (c) 1996 Elsevier Science B.V. All rights reserved. Measurements of turbulent eddy fluxes of momentum and heat under growing thin sea ice in a pool artificially constructed in the ice sheet were made with a three-dimensional ultrasonic current meter and a fast-response temperature−conductivity meter in Saroma-ko Lagoon, Hokkaido, Japan, during the joint Japanese and Canadian Saroma−Resolute Study (SARESr project. The conductive heat flux across the ice sheet and the latent heat of newly formed ice in the pool were estimated from observing the changes in temperature gradients through the ice sheet and in ice thickness. The mean friction velocity was 0.13 cm s −1 over the one-day sampling period. The mean oceanic heat flux from the ocean to the growing thin ice, obtained from the eddy correlation method, was 75 W m −2 over the ice growing period at night and 33 W m −2 over the non-ice growing period in the day.

Application software for a data acquisition and control system for measuring atmospheric trace gas and energy fluxes

Abstract Software for a data acquisition and control system for measuring tower-based fluxes by micrometeorological techniques is described. The system can record up to 16 channels and also creates pseudo-channels from mathematical expressions. Fluxes can be measured with this system by either the eddy-correlation or eddy-accumulation techniques. The measuring system controls valves for conditional sampling, reservoir selection based on a preselected schedule for eddy accumulation systems and alignment of sensors to orient them into the wind. The software simplifies the interfacing of micrometeorological sensors to data acquisition systems for collecting, storing and processing flux data. Many software features described in this publication can be adapted to other hardware systems for eddy flux measurements. It has been extensively used for more than three years in various experiments related to biosphereatmosphere interactions.

Latent and sensible heat flux predictions from a uniform pine forest using surface renewal and flux variance methods

A surface renewal model that links organized eddy motion to the latent and sensible heat fluxes is tested with eddy correlation measurements carried out in a 13m tall uniform Loblolly pine plantation in Duke Forest, Durham, North Carolina. The surface renewal model is based on the occurance of ramp-like patterns in the scalar concentration measurements. To extract such ramp-like patterns from Eulerian scalar concentration measurements, a newly proposed time-frequency filtering scheme is developed and tested. The time-domain filtering is carried out using compactly-supported orthonormal wavelets in conjunction with the “Universal Wavelet Thresholding” approach of Donoho and Johnstone, while the frequency filtering is carried out by a band-pass sine filter centered around the ramp-occurrence frequency as proposed by other studies. The method was separately tested for heat and water vapour with good agreement between eddy correlation flux measurements and model predictions. The usefulness of the flux-variance method to predict sensible and latent heat fluxes is also considered. Our measurements suggest that the simple flux-variance method reproduces the measured heat and momentum fluxes despite the fact that the variances were measured within the roughness sublayer and not in the surface layer. Central to the predictions of water vapour fluxes using the flux-variance approach is the similarity between heat and water vapour transport by the turbulent air flow. This assumption is also investigated for this uniform forest terrain.

Atmospheric deposition of reactive nitrogen oxides and ozone in a temperate deciduous forest and a subarctic woodland: 1. Measurements and mechanisms

We present 5 years of NOy and O3 eddy flux and concentration measurements and NOx concentration measurements at Harvard Forest (1990–1994), a mixed deciduous forest in central Massachusetts, and 2 months of data for a spruce woodland near Schefferville, Quebec, during the NASA ABLE3B/Northern Wetlands Study (1990). Mean midday values of net dry NOy flux from atmosphere to canopy were 3.4 and 3.2 μmole m−2 hr−1 at Harvard Forest in summer and winter, respectively, and 0.5 μmole m−2hr−1 at Schefferville during summer. Nighttime values were 1.3, 2.0, and 0.15 μmole m−2 hr−1, respectively. For 1990–1994, the net annual dry deposition of nitrogen oxides was 17.9 mmole m−2 yr−1 (2.49 kgN ha−1 y−1). Oxidized species such as HNO3 dominated N deposition, with minor contributions from direct deposition of NO2. Emissions of NO from the forest soil were negligible compared to deposition. Comparison of NOy deposition at Harvard Forest and Schefferville and analysis of the dependence on meteorological parameters show that anthropogenic sources dominate the nitrogen oxide inputs over much of North America. Heterogeneous reactions account for >90% of the conversion of NO2 to HNO3 in winter, leading to rates for dry deposition of NOy similar to fluxes in summer despite 10‐fold decrease in OH concentrations. In summer, formation of HNO3 by heterogeneous reactions (mainly at night) could provide 25–45% of the NO2 oxidation.

Scaling gross ecosystem production at Harvard Forest with remote sensing: a comparison of estimates from a constrained quantum‐use efficiency model and eddy correlation

ABSTRACTTwo independent methods of estimating gross ecosystem production (GEP) were compared over a period of 2 years at monthly integrals for a mixed forest of conifers and deciduous hardwoods at Harvard Forest in central Massachusetts. Continuous eddy flux measurements of net ecosystem exchange (NEE) provided one estimate of GEP by taking day to night temperature differences into account to estimate autotrophic and heterotrophic respiration. GEP was also estimated with a quantum efficiency model based on measurements of maximum quantum efficiency (Qmax), seasonal variation in canopy phenology and chlorophyll content, incident PAR, and the constraints of freezing temperatures and vapour pressure deficits on stomatal conductance. Quantum efficiency model estimates of GEP and those derived from eddy flux measurements compared well at monthly integrals over two consecutive years (R2= 0–98).Remotely sensed data were acquired seasonally with an ultralight aircraft to provide a means of scaling the leaf area and leaf pigmentation changes that affected the light absorption of photosynthetically active radiation to larger areas. A linear correlation between chlorophyll concentrations in the upper canopy leaves of four hardwood species and their quantum efficiencies (R2= 0–99) suggested that seasonal changes in quantum efficiency for the entire canopy can be quantified with remotely sensed indices of chlorophyll. Analysis of video data collected from the ultralight aircraft indicated that the fraction of conifer cover varied from < 7% near the instrument tower to about 25% for a larger sized area. At 25% conifer cover, the quantum efficiency model predicted an increase in the estimate of annual GEP of < 5% because unfavourable environmental conditions limited conifer photosynthesis in much of the non‐growing season when hardwoods lacked leaves.

Measurements of internal wave band eddy fluxes above a sloping bottom

The boundary layer near a sloping bottom may have a major influence on the ocean’s interior density structure (due to “boundary mixing”) and on its circulation (because of the arrest of the Ekman layer by buoyancy forces). As a first attempt to measure eddy fluxes of momentum and buoyancy, in order to quantify the mixing in this region, we have carried out a S-day pilot experiment on a sloping side of Emerald Basin on the Scotian Shelf. A moored upward-looking 1.2 MHz ADCP and a thermistor chain mounted along its vertical axis returned analyzable data between 8 and 17 m above the bottom at one-minute intervals. An extensive set of microstructure profiles was also obtained. The predominantly tidal flow regime causes the bottom boundary layer thickness to vary between 3 < z < 30 m, with most high frequency activity during the upslope phase. A bottom-normal momentum flux significantly different from zero is found in the crossisobath direction only. The main contribution comes from a band near the buoyancy frequency N, possibly indicative of advective or Kelvin-Helmholtz instability. When cast in terms of mean-flow shear, the stress yields an eddy viscosity A = 9 x 10e3 m* s-l within the boundary layer and twice this value at z = 15 m, the average height of the pycnocline that caps the boundary layer. The buoyancy flux also seems to be dominated by fluctuating signals near N, but is countergradient and only significantly different from zero at a height of about 15 m. The associated restratification occurs in short periods of approximately one hour when isotherms rise rapidly. Indirect evidence for the importance of the tertiary circulation within the boundary layer is found from the gradient of stress divergence and the mean bottom-normal velocity. An approximate turbulent kinetic energy balance has been investigated, with the currents split into three parts (mean, tidal, and the high frequency part of the internal waveband (“turbulence”)). Production balances viscous dissipation within a factor of 2. Turbulent kinetic energy production by interaction between the turbulent Reynolds stress and the mean flow shear and tidal shear are of the same order of magnitude, but the buoyancy term appears to be of equal importance at the pycnocline.

Eddy momentum flux and its contribution to the Southern Ocean momentum balance

A LARGE amount of momentum is transferred to the Southern Ocean by strong westerly winds. Analytical and numerical models have suggested that transient eddies may be important in transporting this momentum away from the region of wind forcing, either horizontally1or vertically downwards where it is balanced by bottom topographic drag2–5. There are, however, few long-term in situ observations of horizontal eddy momentum flux6–8, and no large-scale measurements of vertical eddy fluxes, to test these models. As a result, the momentum balance of the Antarctic circumpolar current (ACC) remains uncertain, and the role of eddies controversial. Here we use Geosat satellite altimeter data to resolve directional eddy kinetic energy and horizontal eddy momentum flux in the ACC on fine spatial and temporal scales. The complex spatial distribution of surface eddy momentum flux is strongly influenced by bottom topography. The horizontal eddy momentum flux tends generally to concentrate the mean flow, although some regions of divergence are observed. Our results show that the zonally averaged horizontal eddy momentum flux from transient eddies is an order of magnitude too small, and in the wrong direction to directly balance the eastward momentum input from wind1.

A sensitive fast-response probe to measure turbulence and heat flux from any airplane

The theory, configuration, and accuracy of an inexpensive probe to measure turbulence from a small airplane are presented. The probe employs a nine-hole pressure-sphere design along with inprobe high-frequency pressure, temperature, and acceleration sensors. This sensor suite is specifically designed to extend mass, momentum and energy eddy-flux measurement to the higher frequencies characteristic of marine and nocturnal boundary layers. The probe is part of a mobile flux system, independent of the conveyance, which does not require a separate Inertial Navigation System.

A novel ozone sensor for direct eddy flux measurements

A small, lightweight (1.5 kg) and fast-response ozone sensor for direct eddy flux measurements has been built. The basis for detection is the chemiluminescence of an organic dye adsorbed on dry silica gel in the reaction with ozone. The chemiluminescence is monitored with a cheap and small blue-sensitive photomultiplier. At a flow rate of 100 l min-1 the ozone sensor has a 90% response time of significantly better than 0.1 s with a detection limit lower than 50 ppt at S/N=3. There are no interferences from other atmospheric trace gases like NOx, H2O2 and PAN. Water vapour and SO2 enhance the chemiluminescence efficiency of the ozone sensor. Since their response times are 22 seconds and 30 minutes, respectively, no correlation between rapid ozone fluctuations and those of these two trace gases is noticed by the ozone sensor when operating at a frequency of 10 Hz.

Mean flux estimation and cospectra of airborne carbon dioxide and water vapour eddy flux measurements in the planetary surface layer

CO2 and H2O eddy-flux measurements derived from airborne vertical wind and concentration data are examined to determine the accuracy of the eddy-flux estimate in moderately unstable conditions within the surface layer. Integral scale estimation is used to determine the minimum length of a sample required to achieve a given accuracy. Cospectral analysis is used to examine the scales responsible for the transport of water vapour and carbon dioxide in the surface layer. Cospectra are found to broaden as the sampling altitude increases, a well known result. At low altitudes, the cospectrum is found to become negative for large scales. A significant fraction of the flux is carried by this range of scales, which suggests that scales larger than the boundary-layer height may play an important role in the transport at low altitudes.

Measuring ambient ozone with high sensitivity and bandwidth

An ozone (O3) analyzer for use in eddy-correlation measurements is described which has an unfiltered bandwidth of at least 20 Hz and sensitivity of at least 0.1 ppbv at 12-Hz bandwidth. Based on chemiluminescence with nitric oxide, it is designed primarily for use on meteorological aircraft, but it also works well for tower measurements. It is significant that the detecting mechanism responds linearly to time-dependent inputs, which is a definite advantage for applications involving measurements of eddy fluxes and spectra of O3. The unit is highly specific, the only known interference being a slight quenching of the chemiluminescence by water vapor. The unit does not respond to cloud liquid water or to precipitation, and excellent performance has been obtained in the presence of these two conditions. The entire package weighs 110 kg and requires 360 W at 110 V ac (60 or 400 Hz), and 1200 W at 28 V dc.