Download Free Short Range Forecasting Through Extrapolation Of Satellite Imagery Patterns Part 1 Book in PDF and EPUB Free Download. You can read online Short Range Forecasting Through Extrapolation Of Satellite Imagery Patterns Part 1 and write the review.

This report describes the development and testing of an objective technique to forecast cloudiness and precipitation through extrapolation of satellite imagery. By utilizing on objectively determined cloud-motion vector, the technique makes local forecasts of satellite parameters (brightness and IR temperature), with high temporal resolution, using simple linear extrapolation. Algorithms are then used to convert the satellite parameters to total cloud cover, probability of 1-hour precipitation, and presence of low, middle, and high clouds. The test program computed motion vectors and made forecasts out to 7 hours, in half-hour steps, at 30 locations. The program was tested on 12 spring and fall cases, using half-hourly GOES imagery. For periods beyond 2 hours, forecasts of cloud cover and precipitation were markedly better than persistence, which deficiencies in specification hindered short-period performance. Forecasts of cloud layers were worse than persistence due to inadequate specification algorithms. The net results were quite encouraging, and further refinements and developments are planned.
In the quixotic quest to reduce air pollution and fuel cars with alternative sources instead of gas, solar powered cars have emerged as one option. Although disagreements abound about the feasibility and practicality of these vehicles, this book presents the basics behind the idea of solar cars. From the construction of the engine to raising funds, the book is a valuable introduction to the present and future of the emission-free automobile.
This report summarizes the work accomplished during the first phase of an investigation concerning methods of introducing digitized satellite imagery into short-range, objective forecasting operations. The data archive being assembled for this study is described, with particular attention given to the steps taken to maximize the accuracy of the satellite imagery. These steps included 'fine tuning' the navigation and selecting procedures for 'normalizing' the data by correcting for the effects of Lambertian and anisotropic scattering. Consistency of the data, spatial and temporal, was tested by analysis of ground reflectance during cloudless days, and a pilot test of the specification of single layers of clouds was conducted. Both of these tests gave encouraging results. An investigation of specifying precipitation rate, using just the visible reflectance and infrared temperature of the cloud top, also produced good results. Nomograms for the average rate during the hour following the satellite observation, as well as for the probability of observing more than 0.01 in. and 0.10 in. of precipitation, are illustrated. Two appendices present the geometrical and optical equations relevant to the investigation. (Author).
A previously developed advection forecast technique was modified to include data extracted from satellite imagery. A forecast experiment was then conducted using a data base gathered at AFGL during March 1984. This experiment was designed to test the usefulness of : (a) 3-hour forecast updates, (b) a biquadratic interpolation, and (c) cloud and precipitation information from satellite imagery. The test results confirmed earlier tests in that advection using space-averaged 500-mb winds produced the best overall scores and that in general the scores for 1 - 15 hours were better than persistence. The age of the advection flow (3, 6 or 9 hours old) did not affect forecast score, making updates useful. The biquadratic interpolation procedure produced better fits to observation than bilinear and appears to have improved forecasts. There was but a small benefit from adding satellite information to surface observations when forecasting cloud cover and hourly precipitation. the difficulties of trying to forecast even 30 to 50 percent of the time-change variance suggest that alternative approaches such as mesoscale modeling will be needed for accurate, reliable short-range forecasts.