When a target source is observed,a noise level for each baseline can be expressed as

(3) |

where *k* is Boltzmann constant, is quantization efficiency (), is system noise temperature, is system equivalent flux density, is antenna effective aperture area ( in which and are the aperture efficiency and antenna diameter, respectively), *B* is the bandwidth, and is on-source integration time. Note that for an integration time beyond 3 minutes (in the K band), the noise level expected by equation (3) cannot be attained because of the coherence loss due to the atmospheric fluctuation. Thus, for finding fringe within a coherence time, the integration time cannot be longer than 3 minutes. For VLBI observations, signal-to-noise ratio (*S/N*) of at least 5 and usually 7 is generally required for finding fringes.

A resultant image noise level can be expressed as

(4) |

If the array consists of identical antennas, an image noise levels can be expressed as

(5) |

where is the number of antennas. Using the typical antenna parameters (Table 9), baseline and image sensitivity values of KaVA are calculated as listed in Table 10.

KVN | VERA | ||||||

Band | Tsys[K] | SEFD [Jy] | Tsys[K] | SEFD [Jy] | |||

K | 100 | 0.6 | 1328 | 120 | 0.5 | 2343 | |

Q | 150 | 0.6 | 1992 | 250 | 0.5 | 4881 |

Baseline sensitivity [mJy] | Image sensitivity [mJy] | ||||||

Band | VERA-VERA | KVN-KVN | VERA-KVN | VERA | KVN | KaVA | |

K | 10.7 | 6.1 | 8.1 | 0.4 | 0.3 | 0.2 | |

Q | 22.4 | 9.1 | 14.3 | 0.8 | 0.5 | 0.3 | |

Sensitivities (1) are listed in unit of mJy. | |||||||

Integration time of 120 seconds and 4 hours are assumed for baseline and image sensitivities, respectively. Total bandwidth of 256 MHz (for continuum emission) is assumed for all the calculations. In the case of narrower bandwidth of 15.625 KHz (for maser emission), sensitivities can be calculated by multiplying a factor of 128. |

Figures 10 and 11 show the system noise temperature at Mizusawa and Ulsan, respectively. For Mizusawa, receiver noise temperatures are also plotted.

Note that the receiver temperature of the VERA antenna includes the temperature increase due to the feedome loss and the spill-over effect. In Mizusawa, typical system temperature in the K band is K in fine weather of winter season, but sometimes rises above K in summer season. The system temperature at Iriki station shows a similar tendency to that in Mizusawa. In Ogasawara and Ishigakijima, typical system temperature is similar to that for summer in Mizusawa site, with typical optical depth of . The typical system temperature in the Q band in Mizusawa is K in fine weather of winter season, and K in summer season. The typical system temperature in Ogasawara and Ishigakijima in the Q band is larger than that in Mizusawa also. The typical system temperature in the K band at all KVN stations is around 100 K in winter season. In summer season, it increases up to 300 K. In the Q band, the typical system temperature is around 150 K in winter season and 250 K in summer season at Yonsei and Tamna. The system temperature of Ulsan in the Q band is about 40 K lower than the other two KVN stations. This is mainly due to the difference in receiver noise temperature (see Table 5).**KaVA ( KVN and VERA Array )**