Temperature Humidity Infrared Radiometer Research Articles

Occurrence of ozone anomalies over cloudy areas in TOMS version-7 level-2 data

Abstract. This study investigates anomalous ozone distributions over cloudy areas in Nimbus-7 (N7) and Earth-Probe (EP) TOMS version-7 data and analyzes the causes for ozone anomaly formation. A 5°-longitude by 5°-latitude region is defined to contain a Positive Ozone Anomaly (POA) or Negative Ozone Anomaly (NOA) if the correlation coefficient between total ozone and reflectivity is > 0.5 or < -0.5. The average fractions of ozone anomalies among all cloud fields are 31.8 ± 7.7% and 35.8 ± 7.7% in the N7 and EP TOMS data, respectively. Some ozone anomalies are caused by ozone retrieval errors, and others are caused by actual geophysical phenomena. Large cloud-height errors are found in the TOMS version-7 algorithm in comparison to the Temperature Humidity Infrared Radiometer (THIR) cloud data. On average, cloud-top pressures are overestimated by ~200 hPa (THIR cloud-top pressure < 200 hPa) for high-altitude clouds and underestimated by ~150 hPa for low-altitude clouds (THIR cloud-top pressure > 750 hPa). Most tropical NOAs result from negative errors induced by large cloud-height errors, and most tropical POAs are caused by positive errors due to intra-cloud ozone absorption enhancement. However, positive and negative errors offset each other, reducing the ozone anomaly occurrence in TOMS data. Large ozone/reflectivity slopes for mid-latitude POAs show seasonal variation consistent with total ozone fluctuation, indicating that they result mainly from synoptic and planetary wave disturbances. POAs with an occurrence fraction of 30--60% occur in regions of marine stratocumulus off the west coast of South Africa and off the west coast of South America. Both fractions and ozone/reflectivity slopes of these POAs show seasonal variations consistent with that in the tropospheric ozone. About half the ozone/reflectivity slope can be explained by ozone retrieval errors over clear and cloudy areas. The remaining slope may result from there being more ozone production because of rich ozone precursors and higher photolysis rates over high-frequency, low-altitude clouds than in clear areas. Ozone anomalies due to ozone retrieval errors have important implications in TOMS applications such as tropospheric ozone derivation and analysis of ozone seasonal variation.

Upper tropospheric ozone derived from the cloud slicing technique: Implications for large‐scale convection

This study evaluates the spatial distributions and seasonal cycles in upper tropospheric ozone (pressure range 200–500 hPa) from low to high latitudes (60°S to 60°N) derived from the satellite retrieval method called “cloud slicing.” The cloud slicing method determines ozone profile information in the troposphere by combining colocated measurements of cloud top pressure and above‐cloud column ozone. Measurements of Nimbus 7 Total Ozone Mapping Spectrometer (TOMS) above‐cloud column ozone and Nimbus 7 Temperature Humidity Infrared Radiometer (THIR) cloud top pressure during 1979–1984 were merged to derive upper tropospheric ozone. Because of large footprint measurements from TOMS (∼100 km diameter on average), upper tropospheric ozone derived from cloud slicing coincides with large‐scale convection events. These events are not necessarily representative of average atmospheric conditions in regions near and poleward of the tropospheric wind jets (around ±30° latitude), especially in winter and spring seasons when dynamical wave activity in the troposphere and lower stratosphere is most intense. The cloud slicing method with Nimbus 7 geometry in any case provides a unique opportunity to investigate the behavior of upper tropospheric ozone under condition of intense broad‐scale convection. In the tropics the measured upper tropospheric ozone shows year‐round enhancement in the Atlantic region and evidence of a possible semiannual variability. Outside the tropics, upper tropospheric ozone from cloud slicing shows greatest abundance in winter and spring seasons in both hemispheres with largest variance and largest amounts in the northern hemisphere. This seasonal cycle behavior under conditions of intense convection is different from general ozonesonde climatology which shows instead upper tropospheric ozone maximizing around early to middle summer months. The seasonal cycles and spatial characteristics in upper tropospheric ozone from cloud slicing are similar to lower stratospheric ozone. It is suggested that the large‐scale convection events with cloud slicing coincide with an occurrence of stratosphere‐troposphere exchange (STE).

Areal distribution of the oxygen-isotope ratio in greenland

Mean values of the oxygen-isotope ratio relative to standard mean ocean water (δ 18O, in ‰) reported for 46 sites on the Greenland ice sheet are compiled together with data on mean annual surface temperature, latitude, δ 18O elevation, and mean annual shortest distance to the open ocean denoted by the 10% sea-ice concentration boundary. Stepwise regression analyses, with δ 18O as the dependent variable, define two robust models. In the forward mode at the 99.9% confidence level, only temperature enters the model. In the backward mode at the 95% confidence level, only temperature, latitude and distance to the open ocean remain in the model. Inversions of the models on the basis of 160 gridpoint locations 100 km apart in the area delimited by the surface equilibrium line produce four contoured distributions of δ 18O. Two distributions are based on the bivarrate model and two on the multivariate model. The second distribution for each model is obtained substituting mean annual surface-temperature values obtained from the Nimbus-7 Temperature Humidity Infrared Radiometer (THIR) database. All four distributions are considered valid, and differences between them are evaluated using contoured anomaly maps. It is suggested that the inversion of the multivariate model using THIR data provides the more reliable pattern for studies of atmospheric advection or for the derivation of ice-flow adjustments for δ 18O series obtained from deep-core or ablation-zone sites.

Areal distribution of the oxygen-isotope ratio in Greenland

Mean values of the oxygen-isotope ratio relative to standard mean ocean water (δ18O, in ‰) reported for 46 sites on the Greenland ice sheet are compiled together with data on mean annual surface temperature, latitude, δ18O elevation, and mean annual shortest distance to the open ocean denoted by the 10% sea-ice concentration boundary. Stepwise regression analyses, with δ18O as the dependent variable, define two robust models. In the forward mode at the 99.9% confidence level, only temperature enters the model. In the backward mode at the 95% confidence level, only temperature, latitude and distance to the open ocean remain in the model. Inversions of the models on the basis of 160 gridpoint locations 100 km apart in the area delimited by the surface equilibrium line produce four contoured distributions of δ18O. Two distributions are based on the bivarrate model and two on the multivariate model. The second distribution for each model is obtained substituting mean annual surface-temperature values obtained from the Nimbus-7 Temperature Humidity Infrared Radiometer (THIR) database. All four distributions are considered valid, and differences between them are evaluated using contoured anomaly maps. It is suggested that the inversion of the multivariate model using THIR data provides the more reliable pattern for studies of atmospheric advection or for the derivation of ice-flow adjustments for δ18O series obtained from deep-core or ablation-zone sites.

Surface temperatures in the polar regions from Nimbus 7 temperature humidity infrared radiometer

Monthly surface temperatures in the Arctic and Antarctic regions have been derived from the 11.5‐μM thermal infrared channel of the Nimbus 7 temperature humidity infrared radiometer (THIR) for a whole year in 1979 and for a winter and a summer month from 1980 through 1985. The data set shows interannual variability and provides spatial details that allow identification of temperature patterns over sea ice and ice sheet surfaces. For example, the coldest spot in the southern hemisphere is observed to be consistently in the Antarctic plateau in the southern hemisphere, while that in the northern hemisphere is usually located in Greenland, or one of three other general areas: Siberia, the central Arctic, or the Canadian Archipelago. Also, in the southern hemisphere, the amplitude of the seasonal fluctuation of ice sheet temperatures is about 3 times that of sea ice, while in the northern hemisphere, the corresponding fluctuations for the two surfaces are about the same. The main sources of error in the retrieval are cloud and other atmospheric effects. These were minimized by first choosing the highest radiance value from the set of measurements during the day taken within a 30 km by 30 km grid of each daily map. Then the difference of daily maps was taken, and where the difference is greater than a certain threshold (which in this case is 12°C), the data element is deleted. Overall, the monthly maps derived from the resulting daily maps are spatially and temporally consistent, are coherent with the topography of the Antarctic continent and the location of the sea ice edge, and are in qualitative agreement with climatological data. Quantitatively, THIR data are in good agreement with Antarctic ice sheet surface air temperature station data with a correlation coefficient of 0.997 and a standard deviation of 2.0°C. The absolute values are not as good over the sea ice edges, but a comparison with Russian 2‐m drift station temperatures shows very high correlation (with correlation coefficient at 0.998) and a standard deviation of 1.1°C. Overall, the rms error is estimated to be from 1° to 2°C, depending on the surface, while the average bias when compared with in situ data is less than 2°C.

Effect of marine stratocumulus on TOMS ozone

The algorithm used to correct total O3 from the total ozone mapping spectrometer (TOMS) for cloud effects is based on the measured reflectivity, a climatological cloud top height, and an assumed tropospheric O3 column amount below clouds. In regions of persistent subtropical marine stratocumulus it is assumed that this introduces a positive error into total O3 because these clouds are lower than the assumed mean cloud height used in the algorithm. This appears to be confirmed by high correlation between Nimbus 7 TOMS total O3 and reflectivity data for typical regimes of persistent stratus, as identified by the international satellite cloud climatology project (ISCCP) observations. The TOMS total O3 overestimate has been computed using Nimbus 7/solar backscattered ultraviolet total O3 derived using temperature humidity infrared radiometer (THIR) data for years 1979–1984. A functional relationship between the THIR/non‐THIR total O3 difference and reflectivity is used with TOMS reflectivity to modify Nimbus 7 TOMS O3 data for selected regions and periods. The correction diminishes or eliminates a number of apparent O3 maxima, with reductions of up to 20 Dobson units (DU) in total O3 on daily maps and ∼5 DU on monthly mean O3 maps. Significant correlation between corrected TOMS O3 and reflectivity data remains because low‐altitude O3 is retrieved more efficiently over a high‐albedo surface. It is also possible that dynamical influences leading to stratocumulus formation bring O3‐enriched air into the area. These results imply that although good arguments can be made for the use of TOMS total O3 as a proxy for tropospheric O3 in the tropics, caution must be exercised in the use of daily and even monthly O3 maps in the vicinity of clouds. Further research into the TOMS algorithm in cloudy regions is required to derive reliable estimates of tropospheric O3.

Clouds‐radiation interactions in a general circulation model: Impact upon the planetary radiation balance

The unique multimonth set of simultaneous Earth radiation budget observations and cloud amount estimates taken during the Nimbus 7 satellite mission from June 1979 to May 1980 was used to validate a long‐term climate simulation obtained with the latest version of the National Center for Atmospheric Research (NCAR) Community Climate Model. The comparison focused on the temporal variability of the model‐generated cloud and radiation fields versus satellite data with the aim to (1) test the model's ability to simulate short‐term fluctuations; and (2) examine the impact of the treatment of the interactions between clouds, radiation, and the hydrologie cycle on the model's performance. The Nimbus 7 data set comprised broad‐spectral‐band observations of the outgoing infrared radiation and planetary albedo taken by the Earth radiation budget scanners and total cloud amount estimates derived from radiances measured by the Temperature Humidity Infrared Radiometer and Total Ozone Mapping Spectrometer using a bispectral threshold method. Model‐simulated northern hemisphere summer and winter seasons were obtained from a 15‐year time integration including a seasonal cycle. The comparison helped identify a major difficulty in the simulation of climate with the NCAR Community Climate Model. Although the global distributions of the seasonal average and standard deviation of the model‐generated cloud and radiation fields agreed reasonably well with those obtained from satellite observations, the magnitude of the standard deviation of both fields was overestimated by about a factor of 2 over the whole globe. The high frequency of occurrence of clouds is responsible for the increased temporal variability in the model and partially results from reduced interactions between clouds and the hydrologic cycle. In view of the impact of clouds on the atmospheric circulation and its temporal variability, increased fluctuations in cloudiness may affect the sensitivity of the model‐simulated climate to external forcings and it is desirable to implement stronger couplings between the various physical processes in the NCAR Community Climate Model.

On the Interannual Variation of the Earth Radiation Balance

The interannual variability of the top of the atmosphere net radiation budget as measured from the Nimbus- 7 Earth Radiation Budget instruments was calculated for an eight yew period 1979–1986. The largest fluctuations are shown to occur in three tropical areas. the Atlantic off the west emit of Africa, the eastern Pacific near South America, and the western Pacific northeast of Indonesia. The variability in the Atlantic was 20% greater than in the eastern Pacific and 35% greater than in the Indonesian area. The maximum anomalies in these two Pacific regions occurred during the El Niño year 1982–1983, while the maximum Atlantic anomalies. south of the Gulf of Guinea, were during 1984. An independent dataset of derived cloud type and amount from the Temperature Humidity Infrared Radiometer (THIR) and the Total Ozone Mapping Spectrometer (TOMS) instruments shows interannual changes in multilevel convective cloud systems have a minimal effect of the net balance. However changes in middle and low clouds drastically effect the balance, and are the most likely cause of the maximum radiation balance variability in the Gulf of Guinea region. This observed interannual variation of the top of the atmosphere net balance, reported in the present study, denotes the most variable “cloud radiative forcing” situation observed to date.

Nimbus-7 Global Cloud Climatology. part I: Algorithms and Validation

Data from the Temperature Humidity Infrared Radiometer (THIR) and the Total Ozone Mapping Spectrometer (TOMS), both aboard the Nimbus-7 satellite, are used to determine cloudiness parameters for the globe. The 11.5 μm THIR radiances and the 0.36 μm and 0.38 μm TOMS reflectivities, along with concurrent surface temperature data from the Air Force 3-D nephanalysis, are the primary data sources. They are processed by an algorithm that determines total cloud amount, cloud amount in three altitude categories, cirrus cloud, deep convective cloud, warm cloud, and the radiance of radiation emitted by the clouds. and the underlying surface. The algorithm is of the bispectral threshold type, which yields two independent estimates of total cloud, one from the infrared algorithm and one from the UV reflectivity algorithm. For the daytime observations (local noon at the equator), these two independent estimates are combined to determine a composite estimate, while at night (local midnight at the equator), only the infrared threshold algorithm is used in the estimate. Quantitative validation of total cloud amount was performed by comparing the algorithm results with estimates derived by an analyst interpreting geosynchronous satellite (GOES) images, along with auxiliary meteorological data. It has been concluded that the systematic errors of the Nimbus-7 total cloud amount algorithm relative to the analyst are less than 10%, and that the random errors of daily estimates range between 7% and 16%, day or night. These empirical results are consistent with results from a theoretical sensitivity study. Qualitative validation has also been performed by making comparisons with GOES visible and infrared images for specific days. Results indicate that the TOMS cloud estimates improve the IR algorithm estimates of low cloud amount and provide for the identification of cirrus and deep convective cloud, but cloud amounts over humid tropical regions tend to be overestimated even with the use of TOMS. These results suggest that the spatial and temporal characteristics of daily and monthly averaged global cloud cover, including cirrus acid deep convective cloud types, which are presented in Part II, are generally well represented by the Nimbus-7 dataset, which covers a six-year period from April 1979 to March 1985.

Reflectivity of Earth's surface and clouds in ultraviolet from satellite observations

The Total Ozone Mapping Spectrometer (TOMS) on board Nimbus 7 is used to infer the ultraviolet (UV) surface and cloud reflectance at 370 nm. Cloudless surface reflectivity was analyzed on a global basis for all surface types for several months. The UV surface reflectivity varies from 2% for some forest and grassland regions to 14% for some sandy desert areas. A notable exception is the large salt flats of Bolivia, which have a reflectivity of ∼60%. Cloud reflectivity was also analyzed for clouds located at three levels in the atmosphere, as determined by the 11.5‐μm channel of the Temperature Humidity Infrared Radiometer. Average cloud reflectivity at 370 nm ranges from 52% for low clouds (tops <2 km) to 76% for high clouds (tops > 7 km at the equator, decreasing to >4 km at poles).

Evaluation of NIMBUS 7 THIR/CLE and Air Force three‐dimensional nephanalysis estimates of cloud amount

Three different estimates of the percent of fixed geographical regions (160×160 km) either free of cloud (clear) or covered by low, middle, and high (opaque) cloud have been intercompared. The estimates were derived by analysts interpreting geosynchronous satellite images, with concurrent meteorological observations; from NIMBUS 7 temperature humidity infrared radiometer (THIR) CLOUDS ERB (CLE) data; and from Air Force three‐dimensional (3‐D) nephanalysis (3DN) data. Standard linear regression and correlation statistics have been used to estimate the systematic and random errors of the CLE and 3DN cloud amounts (independent variable) at different altitudes relative to the analyst (dependent variable). It has been concluded that (1) Air Force 3‐D nephanalysis agrees better with the analyst than THIR/CLE, except for high cloud amount; (2) the CLE and 3DN results show the tendency to overestimate clear amount when clear amount is large and underestimate it when clear amount is small, by 10–20% for CLE and by 5–10% for 3DN, and both agree well with the analyst in the mean; (3) 3DN and CLE systematic errors relative to the analyst are largest for low cloud, typically underestimating the amount by 15%; (4) both 3DN and CLE random errors range between 6 and 30%, smallest for high clouds and greatest for low clouds; (5) CLE estimates of cloud amount over land at night should not be used for scientific investigations unless restricted to high cloud amount.

Satellite Determinations of the Relationship between Total Longwave Radiation Flux and Infrared Window Radiance

Nimbus-7 satellite observations are used to determine the relationship between the total longwave radiation flux and the radiance in the 10-12 μm infrared window. The total longwave fluxes are obtained from the earth radiation budget (ERB) narrow-field-of-view (NFOV) observations of total radiance; the IR window radiances are those measured by the Temperature Humidity Infrared Radiometer (THIR). Regression equations are obtained relating the total flux equivalent brightness temperatures to the radiance equivalent brightness temperature of the IR window. These empirical equations are compared to similar regression equations based on radiative transfer calculations for a large sample of atmospheric soundings. The latter theoretical equations are used by NOAA in the processing of IR window observations from operational polar orbiting satellites to obtain total longwave flux estimates. The observational results indicate that there is a very high correlation between the flux equivalent brightness temperature and the IR window radiance equivalent brightness temperature, and that the former can indeed be determined from measurements of the latter, thus validating the general NOAA approach. Tests on independent data suggest that rms flux errors of ∼11 w m−2 are to be expected for single applications of the empirical equations. The theoretical equations used by NOAA have an average positive bias of ∼13 wm−2 or a relative bias of ∼6% with respect to the ERB NFOV observations; the relative bias disappears at high flux values and increases with decreasing flux. A preliminary attempt to determine the cause of the discrepancy between the empirical and theoretical results indicates that a major factor may be the unrepresentativeness of the atmospheric soundings used in developing the theoretical regression coefficients.

Validation of nimbus-7 temperature-humidity infrared radiometer estimates of cloud type and amount

Estimates of clear and low, middle and high cloud amount in fixed geographical regions approximately (160km)2 are being made routinely from 11.5μm radiance measurements of the Nimbus-7 Temperature-Humidity Infrared Radiometer (THIR). The purpose of validation is to determine the accuracy of the THIR cloud estimates. Validation requires that a comparison be made between the THIR estimates of cloudiness and the “true” cloudiness. The validation results reported in this paper use human analysis of concurrent but independent satellite images with surface meteorological and radiosonde observations to approximate the “true” cloudiness. Regression and error analyses are used to estimate the systematic and random errors of THIR derived clear amount.

Wind Determination from Nimbus 5 Observations in the 6.3μm Water Vapor Band

Abstract Temperature Humidity Infrared Radiometer (THIR) data of adjacent Nimbus 5 orbits have been used to derive winds in the upper troposphere from the radiance measured in a wide spectral interval within the 6.3 μm water vapor band. The radiation in this spectral channel stems primarily from water vapor emission at ∼400 mb and/or emission of high clouds. The horizontal resolution at the subsatellite point is ∼23 km. For this purpose the digital data in overlapping swaths of consecutive orbits first have been enhanced to detect clear contours using the gradient method and the algorithm of Smith and Davis. Wind vectors computed from the displacements of the contours agree well with conventional data. The rms deviation between the derived wind velocities and the radiosonde data is 5–6 m s−1 for a mean wind speed of 30 m s−1. The rms deviation of the wind direction is about 21°. Systematic deviations can be explained by interfering cirrus clouds and by time differences between radiosonde and satellite mea...

Remote sensing of atmospheric water content from satellites using microwave radiometry

Analysis is presented which substantiates the high correlation achieved in relating integrated water vapor and liquid water to brightness temperatures at frequencies near the 22.235 GHz water vapor line. The influence of atmospheric and surface variability is shown to be minimal over low emissivity sea surfaces. Determination of atmospheric water content using regression techniques is shown to follow directly from radiation transfer theory. Satellite data from the Nimbus-E Microwave Spectrometer (NEMS) aboard Nimbus-5 are compared with radiosonde water vapor measurements and cloud images recorded by the Temperature Humidity Infrared Radiometer aboard Nimbus 5.

The Detection of Dust Storms Over Land and Water With Satellite Visible and Infrared Measurements

The position of a large area of dust was successfully located over the Sahara Desert and off the west coast of Africa with infrared measurements from the Temperature Humidity Infrared Radiometer (THIR) and visible information from the Image Dissector Camera System (IDCS) on Nimbus 4 between 20 and 25 April 1970. Over land, near midday, the dust areas were associated with lower equivalent blackbody temperatures (TBB's) than in the surrounding areas due to the mixing of the dust with the cooler air just above the surface. Near midnight dust areas appeared to be located where the TBB's were higher than the surrounding regions. The TBB contrast was insufficient for dust detection over the ocean, but the contrast in the visible was great enough to track the movement of the dust for several days off the African west coast. Measurements from the THIR 6.5-7.2 micron channel were useful for detecting the presence of cirrus clouds which could have caused a misinterpretation of the 10.5-12.5 micron TBB's.

Tropical Cyclone Rainfall as Measured By the Nimbus Electrically Scanning Microwave Radiometer

Abstract A selected group of 1973 North Pacific Ocean tropical cyclones was studied by using data from the Nimbus 5 Electrically Scanning Microwave Radiometer (ESMR), the Temperature-Humidity Infrared Radiometer (THIR), NOAA-2 and USAF DMSP imageries. From the unique combination of infrared, visible, and microwave data, it was possible during various stages of storm development to differentiate between dense cirrus outflow and rain areas, to identify centers of circulation and areas of low-level moisture, and by the use of a theoretical model to estimate semi-quantitatively areas of light, moderate, and heavy rainfall rates.

Application of Nimbus 4 THIR 6.7 μm Observations to Regional and Global Moisutre and Wind Field Analyses

The Nimbus 4 temperature-humidity infrared radiometer (THIR) monitors radiation in the 6.5 to 7.2-micron water vapor absorption region with a 23-km spatial resolution at the subsatellite point. Radiation monitored in this spectral region results primarily from emission in the 250 to 500-mb region of the upper troposphere. The THIR 6.7-micron observations are readily available in photofacsimile imagery form which shows very distinctive patterns associated with spatial variations in atmospheric water vapor. These radiometric observations have been combined in several instances with moisture values measured in the upper troposphere by the standard radiosonde network. In each instance, the result is a much more consistent analysis showing increased spatial detail that agrees with the radiometric observations and does not compromise the conventional data. The improved moisture analyses show relatively dry and moist tongues that are very difficult or impossible to infer from the conventional data alone.

Meteorological applications of the Nimbus 4 Temperature-Humidity Infrared Radiometer, 6.7 Micron Channel Data

Abstract Mid-tropospheric circulation features under essentially clear sky conditions have been noted in the data of the 6.7 μm channel of the Nimbus 4 Temperature-Humidity Infrared Radiometer (THIR) while at the same time these features were undetected by the 11.5 μm channel of the same instrument. The characteristic response of the 6.7 μm channel to atmospheric water vapor emission is primarily from the 250-mb (10.5 km) to 500-mb (5.5 km) levels with a peak contribution at 350 mb (8 km). Dry and moist patterns seen in the 6.7 μm data on 21 February 1971 have been integrated into a 400-mb moisture analysis over the United States. This analysis provided more detailed and timely information than was conventionally available about the advection of dry air aloft prior to development of the Mississippi Tornado of February 1971. The derivation of middle to upper atmosphere flow patterns from the Nimbus 4 THIR, 6.7 μm data under cloud-free conditions has a direct application on a global scale for the GARP and W...