利用树轮数据改进长期气候变化中的森林模拟

在气候变化背景下，全球很多地区森林受到了越来越频繁和严重的干旱威胁。对于未来森林的将如何变化，我们依靠模型来给我们答案，因此需要保证模型的质量可以胜任这一项工作。然而，一般森林模拟仅用短期的观测数据（数小时数天或者数年的）调校或验证其模拟。这样的方式即便能正确模拟短时期里的森林动态，在超过半个世纪乃至更长的气候变化中它们的模拟是否还能有效呢？为了弄清这个问题，除了利用短期观测数据为模型设置参数外，我们还将利用数十年至上百年的树轮宽度和碳稳定同位素(δ13C)序列调校和验证森林模型。当森林模型能够复盘过去几十年气候变化中树轮宽度和δ13C的变化时，该模型也应该有能力去模拟未来气候中森林对气候变化的响应。最后，我们将利用调校好的模型模拟森林在未来气候中生产力及水分状况的变化。本项目应该能提供新的森林模拟方式和可信的森林模拟，其结果定将推动森林生态生理，树轮研究，森林模拟等多学科的发展。

As the climate changes proceed, many forests across the globe have been suffering from more frequent and severe droughts. Such impacts will continue in the future. As we rely on models to predict how forests will change in the future, it is important to have reliable modeling practices to make meaningful predictions. However, most forest modeling practices used short-term observations data (i.e. minutes, hours, or at most several years) to calibrate or validate models. Such practices should reasonably simulate the forest dynamics in the short period, but could it still well represent the forests in the timeframe of half a century or longer when the dramatic climate change has occured? Besides improving modeling quality by using as many short-term observations as possible, we will apply long-term observations of tree-ring width and stable carbon isotope ratios (δ13C) to calibrate and validate models. If calibrated models could reasonably represent the variations of ring width and δ13C during dramatic climate changes in the past decades, such models should have sufficient ability to predict the changes of the forests in the future. At last, we will use the well-calibrated model to predict how forest would change in productivity and hydraulics in response to the climate change, and how forest management practices (thinning as the example) will benefit forests in the future climate. Two process-based models will be used in this study to test if such procedure can benefit both simple and advanced forest/vegetation models, including a simple model 3-PG (Physiological Principles in Predicting Growth) and an advanced model FATES（Functionally Assembled Terrestrial Ecosystem Simulator). The proposed study should provide a new modeling approach and reliable forest simulations; such results would improve our knowledge in forest ecophysiology, tree-ring study, and forest modeling.