Ronald M. Welch

Climatic Impact of Tropical Lowland Deforestation on Nearby Montane Cloud Forests

Tropical montane cloud forests (TMCFs) depend on predictable, frequent, and prolonged immersion in cloud. Clearing upwind lowland forest alters surface energy budgets in ways that influence dry season cloud fields and thus the TMCF environment. Landsat and Geostationary Operational Environmental Satellite imagery show that deforested areas of Costa Ricas Caribbean lowlands remain relatively cloud-free when forested regions have well-developed dry season cumulus cloud fields. Further, regional atmospheric simulations show that cloud base heights are higher over pasture than over tropical forest areas under reasonable dry season conditions. These results suggest that land use in tropical lowlands has serious impacts on ecosystems in adjacent mountains.

A neural network approach to cloud classification

It is shown that, using high-spatial-resolution data, very high cloud classification accuracies can be obtained with a neural network approach. A texture-based neural network classifier using only single-channel visible Landsat MSS imagery achieves an overall cloud identification accuracy of 93%. Cirrus can be distinguished from boundary layer cloudiness with an accuracy of 96%, without the use of an infrared channel. Stratocumulus is retrieved with an accuracy of 92%, cumulus at 90%. The use of the neural network does not improve cirrus classification accuracy. Rather, its main effect is in the improved separation between stratocumulus and cumulus cloudiness. The present study is based on a nonlinear, nonparametric four-layer neural network approach. A three-layer neural network architecture, the nonparametric K-nearest neighbor approach, and the linear stepwise discriminant analysis procedure are compared. >

Cumulus Cloud Properties Derived Using Landsat Satellite Data

Abstract Landsat Multispectral Scanner (MSS) digital data are used to remotely sense cumulus cloud properties such as cloud fraction and cloud reflectance, along with the distribution of cloud number and cloud fraction as a function of cloud size. The analysis is carried out for four cumulus fields covering regions approximately 150 km square. Results for these initial cloud fields indicate that: (i) the common intuitive model of clouds as nearly uniform reflecting surfaces is a poor representation of cumulus clouds, (ii) the cumulus clouds were often multicelled, even for clouds as small as 1 km in diameter, (iii) cloud fractional coverage derived using a simple reflectance threshold is sensitive to the chosen threshold even for 57-meter resolution Landsat data, (iv) the sensitivity of cloud fraction to changes in satellite sensor resolution is less sensitive than suggested theoretically, and (v) the Landsat derived cloud size distributions show encouraging similarities among the cloud fields examined.

Global Survey of the Relationships of Cloud Albedo and Liquid Water Path with Droplet Size Using ISCCP

The most common approach used to model the aerosol indirect effect on clouds holds the cloud liquid water path constant. In this case, increasing aerosol concentration increases cloud droplet concentration, decreases cloud droplet size, and increases cloud albedo. The expected decrease in cloud droplet size associated with larger aerosol concentrations has been found to be larger over land than over water and larger in the Northern than in the Southern Hemisphere, but the corresponding cloud albedo increase has not been found. Many previous studies have shown that cloud liquid water path varies with changing cloud droplet size, which may alter the behavior of clouds when aerosols change. This study examines the relationship between geographic and seasonal variations of cloud effective droplet size and cloud albedo, as well as cloud liquid water path, in low-level clouds using International Satellite Cloud Climatology Project data. The results show that cloud albedo increases with decreasing droplet size for most clouds over continental areas and for all optically thicker clouds, but that cloud albedo decreases with decreasing droplet size for optically thinner clouds over most oceans and the tropical rain forest regions. For almost all clouds, the liquid water path increases with increasing cloud droplet size.

Automated cloud classification of global AVHRR data using a fuzzy logic approach

Abstract A fuzzy logic classification (FLC) methodology is proposed to achieve the two goals of this paper: 1) to discriminate between clear sky and clouds in a 32 × 32 pixel array, or sample, of 1.1-km Advanced Very High Resolution Radiometer (AVHRR) data, and 2) if clouds are present, to discriminate between single-layered and multilayered clouds within the sample. To achieve these goals, eight FLC modules are derived that are based broadly on airmass type and surface type (land or water): equatorial over land, marine tropical over land, marine tropical/equatorial over water, continental tropical over land, marine polar over land, marine polar over water, continental polar over land, and continental polar/arctic over water. Derivation of airmass type is performed using gridded analyses provided by the National Centers for Environmental Prediction. The training and testing data used by the FLC are collected from more than 150 daytime AVHRR local area coverage scenes recorded between 1991 and 1994 over al...

First estimates of the radiative forcing of aerosols generated from biomass burning using satellite data

Collocated measurements from the Advanced Very High Resolution Radiometer (AVHRR) and the Earth Radiation Budget Experiment (ERBE) scanner are used to examine the radiative forcing of atmospheric aerosols generated from biomass burning for 13 images in South America. Using the AVHRR, Local Area Coverage (LAC) data, a new technique based on a combination of spectral and textural measures is developed for detecting these aerosols. Then, the instantaneous shortwave, longwave, and net radiative forcing values are computed from the ERBE instantaneous scanner data. Results for the selected samples from 13 images show that the mean instantaneous net radiative forcing for areas with heavy aerosol loading is about -36 W/sq m and that for the optically thin aerosols are about -16 W/sq m. These results, although preliminary, provide the first estimates of radiative forcing of atmospheric aerosols from biomass burning using satellite data.

Global variation of column droplet concentration in low-level clouds

Cloud droplet concentration is a very important parameter in model studies. However, no global observation is available because it is hard to retrieve by current satellite remote sensing techniques. This study introduces another parameter, column droplet concentration, which can be retrieved by satellite data and used in models. The column droplet concentration (Nc) is the product of cloud geometrical thickness and droplet volume number concentration. This paper presents a method and the results of retrieving column droplet concentration for low-level clouds. The first near-global survey (50°S to 50°N) of Nc reveals more clearly the effect of aerosol concentration variations on clouds. The survey shows the expected increase of column droplet concentrations between ocean and continental clouds and in tropical areas during dry seasons where biomass burning is prevalent. Therefore, column droplet concentration is demonstrated as a good indication of available CCN populations in certain areas.

Validation of satellite retrievals of cloud microphysics and liquid water path using observations from FIRE

Abstract Cloud effective radii (re) and cloud liquid water path (LWP) are derived from ISCCP spatially sampled satellite data and validated with ground-based pyranometer and microwave radiometer measurements taken on San Nicolas Island during the 1987 FIRE IFO. Values of re derived from the ISCCP data are also compared to values retrieved by a hybrid method that uses the combination of LWP derived from microwave measurement and optical thickness derived from GOES data. The results show that there is significant variability in cloud properties over a 100 km×80 km area and that the values at San Nicolas Island are not necessarily representative of the surrounding cloud field. On the other hand, even though there were large spatial variations in optical depth, the re values remained relatively constant (with σ≤2–3 µm in most cases) in the marine stratocumulus. Furthermore, values of re derived from the upper portion of the cloud generally are representative of the entire stratiform cloud. When LWP values are...

The 27-28 October 1986 FIRE IFO cirrus case study: Comparison of radiative transfer theory with observations by satellite and aircraft

Abstract Observations of cirrus and altocumulus clouds during the First International Satellite Cloud Climatology Project Regional Experiment (FIRE) are compared to theoretical models of cloud radiative properties. Three tests are performed. First, Landsat radiances are used to compare the relationship between nadir reflectance at 0.83 μm and beam emittance at 11.5 μm with that predicted by model calculations using spherical and nonspherical phase functions. Good agreement is found between observations and theory when water droplets dominate. Poor agreement is found when ice particles dominate, especially if scattering phase functions for spherical particles am used. Even when compared to a laboratory measured ice particle phase function (Volkovitskiy et al. 1980), the observations show increased side scattered radiation relative to the theoretical calculations. Second, the anisotropy of conservatively scattered radiation is examined using simultaneous multiple-angle views of the cirrus from Landsat and E...

Three Different Behaviors of Liquid Water Path of Water Clouds in Aerosol-Cloud Interactions

Abstract Estimates of the indirect aerosol effect in GCMs assume that either cloud liquid water path is constant (Twomey effect) or increases with increased droplet number concentration (drizzle-suppression or Albrecht effect). On the other hand, if cloud thermodynamics and dynamics are considered, cloud liquid water path may also decrease with increasing droplet number concentration, which has been predicted by model calculations and observed in ship track and urban influence studies. This study examines the different changes of cloud liquid water path associated with changes of cloud droplet number concentration. Satellite data (January, April, July, and October 1987) are used to determine the cloud liquid water sensitivity, defined as the ratio of changes of liquid water path and changes of column droplet number concentration. The results of a global survey for water clouds (cloud-top temperature >273 K, optical thickness 1 ≤ τ < 15) reveal all three behaviors of cloud liquid water path with aerosol ch...