Chlordecone was applied between 1972 and 1993 in banana fields of the French West Indies. This resulted in long-term pollution of soils and contamination of waters, aquatic biota, and crops. To assess pollution level and duration according to soil type, WISORCH, a leaching model based on first-order desorption kinetics, was developed and run. Its input parameters are soil organic carbon content (SOC) and SOC/water partitioning coefficient (Koc). It accounts for current chlordecone soil contents and drainage water concentrations. The model was valid for andosol, which indicates that neither physicochemical nor microbial degradation occurred. Dilution by previous deep tillages makes soil scrapping unrealistic.Lixiviation appeared the main way to reduce pollution. Besides the SOC and rainfall increases, Koc increased from nitisol to ferralsol and then andosol while lixiviation efficiency decreased. Consequently, pollution is bound to last for several decades for nitisol, centuries for ferralsol, and half a millennium for andosol.

In the context of nonpoint source pollution the regulator can not attribute individually the responsibility of pollution because of informational asymmetry which makes the costs of monitoring of individual emission very high. This grounds a moral hazard problem. We analyse group performance based instruments to regulate this kind of informational problem. In particular, we assess randomand collective fining schemes with respect to their deterrence and efficiency. We show that a collective fine scheme is more deterrent than a random fine scheme. However, the analysis of efficiency is less categorical between these two schemes. The efficiency depends on the number of non-compliant agents. If the number of non-compliant agents is high it is better to implement a collective fine scheme. If the number of non-compliant agents is small it is better to implement a random fine scheme.

The objective was to describe and model varia- tion patterns in individual fish responses to contaminants among estuaries, season and gender. Two hundred twenty- seven adult European flounders were collected in two sea- sons (winter and summer) in four estuaries along the Bay of Biscay (South West France), focusing on a pristine system (the Ster), vs. three estuaries displaying contrasted levels of contaminants (the Vilaine, Loire and Gironde). Twenty-three variables were measured by fish, considering the load of contaminants (liver metals, liver and muscle persistent organic pollutants, muscle polycyclic aromatic hydrocar- bons); the gene expression (Cyt C oxydase, ATPase, BHMT, Cyt P450 1A1, ferritin); the blood genotoxicity (Comet test); and liver histology (foci of cellular alteration–tumour, steatosis, inflammation, abnormal glycogen storage). Ca- nonical redundancy analysis (RDA) was used to model these variables using gender, season and estuary of origin as explanatory variables. The results underlined the homo- geneity of fish responses within the pristine site (Ster) and more important seasonal variability within the three contam- inated systems. The complete model RDA was significant and explained 35 % of total variance. Estuary and season respectively explained 30 and 5 % of the total independent variation components, whilst gender was not a significant factor. The first axis of the RDA explains nearly 27 % of the total variance and mostly represents a gradient of contami- nation. The links between the load of contaminants, the expression of several genes and the biomarkers were ana- lysed considering different levels of chemical stress and a possible multi-stress, particularly in the Vilaine estuary.

This paper offers a new mechanism in order to Nash-implement a Pareto optimal level of ambient pollution. As usuas in the literature on non point source pollution, the proposed scheme is not conditional on individual emissions, since they are not observable; rather it is conditional on aggregate emission. But the novelty here is that we do not assume the regulator knows the agents'preferences, with which he could identify the target level of aggregate emission. Our mechanism dispenses with this information, yet it achieves Pareto optimality provided that the number of agents involved in the problem is known.

Nous développons un modèle à générations emboîtées dans lequel l’évolution de la pollutionet la formation de la préoccupation environnementale sont endogènes. D’un côté, despersonnes concernées de façon hétérogène par les questions environnementales participentvolontairement à la dépollution. D’un autre côté, la transmission des attitudes vis-à-vis del’environnement résulte d’un choix économique qui dépend lui-même de la pollution. Lemodèle prédit que la proportion à long terme d’individus préoccupés par l’environnement seraforte dans tous les cas. Cependant, selon la nature de la technologie qui génère la pollution, latransition d’une société peu préoccupée par l’environnement vers une société fortementpréoccupée par l’environnement s’accompagne de deux effets différents sur le niveau depollution à long terme. Si la technologie est « propre », on aboutit à un niveau d’équilibrestable de pollution. A l’inverse, si elle est « sale », la pollution connaît une croissanceillimitée qui finit par causer un désastre environnemental. Ce résultat reproduit les faitsstylisés observés quant à l’évolution conjointe de la préoccupation environnementale et lapollution dans les pays en développement. Dans le deuxième cas, nous montrons que destransferts intergénérationnels de la génération âgée vers la génération active jeune rétablissentla possibilité d’atteindre un état stationnaire de pollution. | We develop an overlapping generations model within which the evolution of pollution and theformation of environmental concern are endogenous. On the one hand, peopleheterogeneously concerned with environmental issues contribute to pollution which is apublic bad. On the other hand, the transmission of environmental attitudes is the result ofsome economic choice which is affected by pollution. The model predicts that the long runproportion of environmentally concerned individuals will always be high. Though, dependingon the pollution-generating technology, the transition from a low-environmentally concernedsociety to a high-environmentally concerned one is accompanied by two different outcomesregarding the long run level of pollution. If the technology is “clean”, there is a stable steadystate level of pollution. However, if it is “dirty”, pollution experiences an unlimited growthwhich eventually causes an environmental disaster. This result captures some stylized factsregarding the joint evolution of environmental concern and pollution in developing nations. Inthe latter case, we show that intergenerational transfers from the older generation to the youngworking one restore the possibility to reach a stationary level of pollution.