Which analysis to use for comparison of species proportions?

#1
I have samples of bee populations from 2011 and 2013. There were 5 different species collected. I need a test that can tell me if there is significant differences between proportions of species between years (ie, did one species start to decline and one species start to become more abundant).

There are also two sites (high alpine site, low alpine site). I would like to analyze my data at BOTH sites first (combined) and then separate it out further by site.

I have looked at Bray-Curtis tests, however I don't know how to compare it by year, species, and SITE using this test. I don't know if i can use a chi square since I have uneven numbers of collections between years (2011 = total 188 bees, 2013 = 335 bees)? I can't use a distance matrix because I would lose species data. Any suggestions?

This is how I have my data arranged:

Year Site Species Count
2013 1 Balteatus 42
2013 1 Bifarious 11
2013 1 Flavifrons 120
2013 1 Frigidus 12
2013 1 Sylvicola 22
2013 2 Balteatus 35
2013 2 Bifarious 10
2013 2 Flavifrons 46
2013 2 Frigidus 10
2013 2 Sylvicola 27
2011 1 Balteatus 30
2011 1 Bifarious 9
2011 1 Flavifrons 14
2011 1 Frigidus 24
2011 1 Sylvicola 17
2011 2 Balteatus 43
2011 2 Bifarious 4
2011 2 Flavifrons 7
2011 2 Frigidus 12
2011 2 Sylvicola 28