Heights of Formation

There are well-known difficulties associated with attributing specific HOFs to emission from broadband filters. Due to the complexity of line emission and continuum, any HOF will vary in both time (e.g., during a flare) and space (e.g., plage/network - internetwork). Whilst realising the limitations of any assigned HOF we suggest the following. Previous work with TRACE UV data (Judge et al. 2001; Krijger et al. 2001; Handy et al. 1999a) place the 1700 Å band lowest in the atmosphere (UV continuum at ~4-10 x 10³K, below temperature minimum), followed by the 1600 Å passband (UV continuum at ~4-10 x 10³K plus Fe II at ~1.3 x 10⁴K, temperature minimum). Up to 50% of the signal in the 1216 Å passband is also believed to be UV continuum (Worden et al. 1999) suggesting a HOF around the mid chromosphere (50% Lyα at ~1-3 x 10⁴K, plus 50% UV continuum). The 1550 Å passband has the highest HOF, at ~1 x 10⁵K, but with sigificant contributions from other lines (e.g., Si II at ~1 x 10⁴K) and continuum. These temperatures are approximate and there is expected to be a degree of overlapping emission between passbands. However in order to carry out a qualitative discussion of our results, we assume a sequence of increasing HOF of 1700 Å, 1600 Å, 1216 Å, 1550 Å.

Techniques have been developed to separate both the Lyα (Handy et al. 1999b) and C IV (Handy et al. 1998) emission from the continuum. However in both cases it has been shown that these techniques are only reliable for more intense solar features (e.g., Figure 6 of Handy et al. 1998). The bright network will be at the lower limit of this intensity, and the techniques are certainly not applicable to the internetwork without long exposure times. As we wish to compare network and internetwork oscillations we decided not to apply any continuum removal to the data.