A signal reading that comes from a point in the sky is likely to be a satellite or the sun.
Assuming you have found a satellite, peak up on it in elevation and azimuth and read out the elevation angle on the scale and estimate the azimuth angle with a compass. That may give you a good idea which satellite it may be. The satellites are in a curved line across the sky. If you are near the top of the arc then you may see several satellites on a roughly horizontal line. Either side of the top of the arc you may detect just two satellites, a long way apart in azimuth e.g. in the south east and in the south west.
Complex satellite meters detect noise/signal power initially and this is fine for peaking up on any signal from any satellite. The meters also contain a simplified satellite TV receiver that is pre-programmed with some example TV carrier frequencies and symbol rates for each satellite. If one of these is recognised then it helps to identify the satellite.
This may not be reliable however for the following reasons:
You may have set the polarisation the wrong way - you will see a nice big signal but since the expected carrier is not present there will be no recognition.
You may have set the polarisation adjustment angle with insufficient accurately. In the worst case, a 45 deg error will give you a nice big signal but there will be so much interference that the receiver will fail to recognise the carrier.
The person who programmed the meter (using L band cable frequencies) may have made a different assumption about the type of LNB you are using now. If you are using a Ku band LNB with a local oscillator of 11.3 GHz and person who programmed the meter assumed a 9.75 GHz LO frequency then the expected L band frequency will never match what the meter is programmed for.
It is possible that there are similar carriers with approx same frequency and symbol rate, on other satellites. Your meter may give a false positive recognition on the wrong satellite.
The carrier that was programmed may have simply changed - new frequency, symbol rate or been turned off etc.
The programmed carrier may be in a different satellite downlink beam, aimed at countries well away from your location.
My experience is that these meters are fine if you are regularly using the same type LNB and lining up dishes on the same satellite. Attempting to use such a meter for the first time on a new satellite and with new LNB is very uncertain and not much better than a simple power meter. If you are in a first time situation I would recommend testing the whole "finding the satellite procedure" at a test/training location before attempting to do it in the field at a customer site. Use hardware, LNB, dish, meter etc that matches what will be used on site.
Check approx visibility of the satellite by putting your eye under the rear lower edge of the dish and sighting via the top rim of the feed horn radome window.
If you are detecting noise over a wide azimuth range corresponding to some buildings maybe your beam is pointed at the buildings rather than 20 deg above the buildings. I would expect noise from buildings to be similar to the thermal noise when you point the antenna or LNB at the ground.
Maybe there is interference. Are you using a C band system ? Is there a 4 GHz terrestrial radio relay link or radar system nearby ?
If you can peak up on some satellite, try to identify it. Then count along the orbit till you come to your NSS10 satellite.
Where are you ? What carrier frequency, symbol rate and polarisation are you trying to receive ? What nominal polarisation did you set as the starting position ? What polarisation adjustment angle did you then apply ? e.g. "I started by setting the receive LNB polarisation to horizontal; the broad faces of the rectangular LNB input waveguide were on either side. I then adjusted the feed 35 deg clockwise, while facing towards the satellite"
If C band, what is your polarisation type (linear or circular) ? NSS10 needs linear.