There is a very dated book by Cone and Hayes (if I remember correctly) called Environmental Problems/Behavioral Solutions, that takes a behaviourist perspective and evaluates the success or otherwise of several forms of pro-environmental schemes. This may be a start.

This may help:

Factors influencing car use for commuting and the intention to reduce it: A question of self-interest or morality?

Wokje Abrahamse, Linda Steg, Robert Gifford, Charles Vlek

Department of Psychology, University of Groningen, The Netherlands; Department of Psychology, University of Victoria, Canada

Transportation Research Part F: Traffic Psychology and Behaviour 01/2009; DOI:10.1016/j.trf.2009.04.004

ABSTRACT Car use for commuting contributes to various environmental and traffic problems, such as pollution and congestion. Policies aimed at reducing commuter car use will be more effective when they target important determinants of car use and willingness to reduce it. This study examined whether variables reflecting self-interest (from the theory of planned behavior [Ajzen, I. (1985). From intentions to actions: A theory of planned behavior. In J. Kuhl & J. Beckmann (Eds.), Action control: From cognition to behavior (pp. 11–39). Berlin: Springer]) and variables reflecting moral considerations (from the norm-activation model [Schwartz, S. H. (1977). Normative influences on altruism. In L. Berkowitz (Ed.). Advances in experimental social psychology (Vol. 10, pp. 221–279). New York: Academic Press]) were able to explain self-reported car use for commuting and intentions to reduce it in a sample of Canadian office workers. Car use for commuting was mostly explained by variables related to individual outcomes (perceived behavioral control and attitudes) whereas the intention to reduce car use was mostly explained by variables related to morality (personal norms). The study also found that perceived behavioral control moderated the relation between personal norms and behavioral intentions: stronger personal norms were associated with stronger behavioral intentions, but only when perceived behavioral control was low. Some issues evoked by these results are discussed.

[