There are several reasons to strengthen the cooperation between the integrated assessment (IA) and earth system (ES) modeling teams in order to better understand the joint development of environmental and human systems. This cooperation can take many different forms, ranging from information exchange between research communities to fully coupled modeling approaches. Here, we discuss the strengths and weaknesses of different approaches and try to establish some guidelines for their applicability, based mainly on the type of interaction between the model components (including the role of feedback), possibilities for simplification and the importance of uncertainty. We also discuss several important areas of joint IA-ES research, such as land use/land cover dynamics and the interaction between climate change and air pollution, and indicate the type of collaboration that seems to be most appropriate in each case. We find that full coupling of IA-ES models might not always be the most desirable form of cooperation, since in some cases the direct feedbacks between IA and ES may be too weak or subject to considerable process or scenario uncertainty. However, when local processes are important, it could be important to consider full integration. By encouraging cooperation between the IA and ES communities in the future more consistent insights can be developed.

h i g h l i g h t sWe estimate ozone impacts on crop yields using high and low air pollution scenarios. Substantial crop losses can be avoided by air pollution control, especially in Asia. Climate policies can have co-benefits for crops due to reduced co-emissions. Reduced ozone damage will also reduce land use and associated carbon emissions. a b s t r a c tExposure to surface ozone has detrimental impacts on vegetation and crop yields. In this study, we estimate ozone impacts on crop production and subsequent impacts on land use in the 2005e2050 period using results of the TM5 atmospheric chemistry and IMAGE integrated assessment model. For the crops represented in IMAGE, we compute relative yield losses based on published exposure-response functions. We examine scenarios with either constant or declining emission factors in a weak climate policy future (radiative forcing target of 6.0 W/m 2 at the end of the century), as well as co-benefits of stringent climate policy (targeted at 2.6 W/m 2 ). Without a large decrease in air pollutant emissions, higher ozone concentrations could lead to an increase in crop damage of up to 20% locally in 2050 compared to the situation in which the changes in ozone are not accounted for. This may lead to a 2.5% global increase in crop area, and a regional increase of 8.9% in Asia. Implementation of air pollution policies could limit crop yield losses due to ozone to maximally 10% in 2050 in the most affected regions. Similar effects can be obtained as a result of co-benefits from climate policy (reducing ozone precursor emissions). We also evaluated the impact of the corresponding land-use changes on the carbon cycle. Under the worst-case scenario analysed in this study, future ozone increases are estimated to increase the cumulative net CO 2 emissions between 2005 and 2050 by about 3.7 Pg C, which corresponds to about 10% of baseline land use emissions over the same period.

h i g h l i g h t s We have developed high and low air pollution emission scenario variants of the RCPs. Both climate and air pollution policies are important for future air quality. Pollution control can reduce the O 3 and aerosol direct RF under climate mitigation.