Because of the diversity of breeding habitats, using container index and breteau index used in the case of Aedes aegypti need not give a realistic picture in the case of Aedes albopictus.
In my oppinion Ovitraps is a very common method, but as you mention, the diversity of breeding habitats difusse this measurement method. Other method is the adult capture, prefereably with BG Sentinel trap. And the last one and more realistic is the use of human volunteers as bait, but it is very laborious. In my oppnion a dureable monitoring with ovitraps is the best cost/effective method. This way you can obtain the number og eggs/week/trap and fix the tolerance index. Some mosquito control services also uses the number of complaits received as index
see Weeraratne, T.C., Perera, M.D.B., Mansoor, M.A.C.M. and Karunaratne, S.H.P.P. (2013) Prevalence and the breeding habitats of dengue vectors Aedes aegypti and Aedes albopictus in semi-urban areas in two different climatic zones in Sri Lanka. International Journal of Tropical Insect Science 33 (4): 216-226.
To estimate 'levels' BG trap is probably a better estimate than eggs or larvae because it measures adults. Pupal counts is also better than oviposition but if you have a lot of natural habitats it is difficult. Oviposition is great for presence absence or monitoring temporal trends, but can be heavily influenced by competing habitats.
Thank you all for your suggestions and expert opinions. I think the best index for Aedes albopictus would be some kind of adult index. I have designed a cheap electrical mosquito collection device for outdoor mosquito sampling and currently am testing its efficiency. My intention is to come out with an adult index.
The BG traps are expensive but the most widely used. If cost is an issue you could also look into constructing sticky traps. Something like a modified oviposition trap but with a sticky substrate (see below, link #1). You can buy the glue (e.g tanglefoot) or purchase sticky cards (e.g. alphascents). I work with a group that does large scale surveillance using sticky traps in Brazil (see below, link #2)
Bare-leg-catch method using human volunteers can give better view of adult mosquito population. It's a bit traditional and laborious but gives reliable results .
Carrieri et al. (www.bioone.org/doi/abs/10.1603/ME10259) compared different estimates of adult Ae. albopictus abundance (pupal index, human landing collection, and # of bites declared by citizens) to the mean number of eggs laid in ovitraps, finding a positive correlation between the 4 methods.