精品SM在线影片

Skip to main content

Streams, Soils, Strategies And (Stressed Out?) Survivors - Ecohydrology In Seasonally Dry Climates

Thompson, Sally 1

1 University of California, Berkeley

Seasonally dry ecosystems (or SDEs) include Mediterranean, Monsoonal and Tropically Dry climates. SDEs occur worldwide and share a pronounced contrast in precipitation between a wet season, and a dry season in which little to no rain falls. Water availability in SDEs during the dry season is dependent on carry-over storage from the rainy season. This can lead to a pronounced bimodality in water availability, which poses several challenges - to those who study these systems, and to the organisms that inhabit them. I want to hit on 3 of these challenges: 1. Predictions about soils and streams: existing stochastic ecohydrologic theory requires that rainfall statistics be stationary in time in order to make a prediction about flow duration curves and soil moisture PDFs. Such predictions support risk-based approaches to understanding stream and terrestrial water supplies. Through simple depletion arguments, we extend these theories to SDEs and identify how the wet season controls water availability in the dry season. 2. Optimal strategies for plants: Plants in SDEs adopt an almost bewildering array of leaf phenological strategies (phenology in this case refers to the timing of when leaves appear on the plant, and under what conditions leaves will be shed). Simple models can be used to evaluate the conditions under which different strategies allow more carbon fixation over the growing season, and might offer insight into why so many different modes of behavior co-exist. 3. Surviving the drought: Strategies are only good up to a point - after which plants are exposed to severe water stress. Field observations suggest that water stress experienced by oak trees in the California drought to date is becoming extreme, but variable, amongst species, locations and individuals. Whether stress predicts mortality remains a critical challenge for linking theory to predictions of change.