Institute of Theoretical and Experimental Biophysics, Russian Academy of Sciences, Pushchino, Russia, 142290.
Belarusian State University, 220010, Minsk, Belarus.
Sci Rep. 2023 Jun 22;13(1):10124. doi: 10.1038/s41598-023-36950-3.
We present an approach (knowledge-and-data-driven, KDD, modeling) that allows us to get closer to understanding the processes that affect the dynamics of plankton communities. This approach, based on the use of time series obtained as a result of ecosystem monitoring, combines the key features of both the knowledge-driven modeling (mechanistic models) and data-driven (DD) modeling. Using a KDD model, we reveal the phytoplankton growth-rate fluctuations in the ecosystem of the Naroch Lakes and determine the degree of phase synchronization between fluctuations in the phytoplankton growth rate and temperature variations. More specifically, we estimate a numerical value of the phase locking index (PLI), which allows us to assess how temperature fluctuations affect the dynamics of phytoplankton growth rates. Since, within the framework of KDD modeling, we directly include the time series obtained as a result of field measurements in the model equations, the dynamics of the phytoplankton growth rate obtained from the KDD model reflect the behavior of the lake ecosystem as a whole, and PLI can be considered as a holistic parameter.
我们提出了一种（知识和数据驱动的、KDD、建模）方法，使我们能够更深入地了解影响浮游生物群落动态的过程。这种方法基于生态系统监测结果获得的时间序列，结合了知识驱动建模（机理模型）和数据驱动（DD）建模的关键特征。使用 KDD 模型，我们揭示了纳罗奇湖生态系统中浮游植物生长率的波动，并确定了浮游植物生长率波动与温度变化之间的相位同步程度。更具体地说，我们估计了相位锁定指数（PLI）的数值，这使我们能够评估温度波动如何影响浮游植物生长率的动态。由于在 KDD 建模框架内，我们直接将现场测量结果获得的时间序列包含在模型方程中，因此从 KDD 模型获得的浮游植物生长率动态反映了整个湖泊生态系统的行为，并且 PLI 可以被认为是一个整体参数。
