The effects of experimental irrigation on plant productivity, insect abundance and the non-breeding season performance of a migratory songbird.

Migratory bird populations are often limited by food during the non-breeding season. Correlative evidence suggests that food abundance on territories varies among years in relation to rainfall, which affects plant productivity and arthropod biomass. At the Font Hill Nature Preserve in Jamaica, we used an irrigation experiment to test the hypothesis that rainfall affects the condition of wintering American redstarts (Setophaga ruticilla) via intermediate effects on plant productivity and arthropod abundance. Experimental plots were irrigated in late February and early March to simulate a mid-season pulse of 200 mm of rain. Irrigation maintained soil moisture levels near saturation and had immediate effects on plant productivity. Cumulative leaf abscission over the dry season was 50% lower on experimental plots resulting in greater canopy cover, and we observed significantly higher ground level shoot growth and the flushing of new leaves on about 58% of logwood (Haematoxylon campechianum) individuals. Arthropod biomass was 1.5 times higher on irrigated plots, but there was considerable inter-plot variability within a treatment and a strong seasonal decline in biomass. Consequently, we found no significant effect of irrigation on arthropod abundance or redstart condition. We suspect that the lack of an irrigation effect for taxa higher on the trophic chain was due to the small spatial scale of the treatment relative to the scale at which these taxa operate. Although redstart condition was not affected, we did observe turnover from subordinate to dominant territorial individuals on experimental plots suggesting a perceived difference in habitat quality that influenced individual behavior.