The present study is aimed at identifying the wavelength regions, where the SED holds key information about the compact H II regions and quantifying their diagnostic values in terms of local slopes of the SEDs and dependence of the angular sizes on the wavelength. In addition, it presents approximate but quick conclusions about some basic parameters of compact H II regions, based on infrared photometric, imaging and / or spectroscopic measurements. This is of particular importance if the measurements are available for a large sample from a space mission like ISO. The idea being that of extracting crude first order parameters of compact H II regions before indulging into more sophisticated and detailed modelling of individual sources.

Self-consistent radiation transfer modelling have been carried
out for the dust and the ionized as well as the neutral gas, for a
large range of parameter space. Three types of exciting stars,
ZAMS O4, O7 and B0.5, have been considered to be embedded
at the centre of a spherical
cloud with a radial density distribution given by a power law
*n(r)* ~ *r*^{-m}; *m* = 0, 1.
The following observables, *viz.*, mid and far infrared
photometric colours; angular sizes (HWHM) as a function of
wavelength; and the radio continuum emission, have been predicted.

Since the total mass of the cloud
for these models is assumed to be
available through observations, the distance to the source must be known.
Hence, the total luminosity and the physical size are
also known. This implies that, for uniform density distribution, the
total _{100} can be
estimated and for r^{-1} an upper limit of
_{100}
may be derived. The observations need to be compared with a
family of models covering a range of
_{100}, to arrive at the best
description of the source.

The radio to far infrared ratio is a very good diagnostic for any
spectral type of the embedded star,
particularly for the r^{-1} density distribution law.

The structural information contained in radial profiles,
as expressed at various wavelengths of ISOCAM filters, when
parameterized by a best fit two parameter power law
(_{1/2} ~ *a* x
^{b}), also has a lot
of diagnostic value.

The next step to the present study relaxes further the assumptions about the compact H II regions as well as includes a more sophisticated treatment of the gas component to predict fine structure line emission of heavy elements that are observable using the ISO spectrometer (Mookerjea & Ghosh, 1999, Paper II).

It is a pleasure to thank D. Narasimha for many useful discussions.