The aim of this paper is to estimate the impact of weather related income shocks on child health in rural Burkina Faso where rain fed agriculture is the dominant production system. We combine health data originating from the 2008 household survey with meteorological data to define shocks at the child level. We first estimate the marginal effect of rainfall at various ages on the child's health in order to identify the critical period during which deprivation has the most severe consequences. Then we look for a different impact of shocks on girls and boys that would reflect a gender bias in intra household resource allocation. We also assess the household ability to smooth consumption by testing for an asymmetric effect of rainfall shocks according to their size and by testing the impact of shocks according to household endowments. Results evidence a strong relationship between rainfall shocks during the prenatal period and child health. Households are not able to dampen small but negative rainfall shocks. Unexpectedly, girls are less severely affected by shocks than boys. The robustness of results is tested by using the sibling and difference-in-differences estimators as well as placebo regressions.