An official website of the United States government
Identifying hidden interactions within complex systems is key to unlocking deeper insights into their operational dynamics, including how their elements affect each other and contribute to the overall system behavior. For instance, in neuroscience, discovering neuron-to-neuron interactions is essential for understanding brain function; in ecology, recognizing interactions among populations is key to understanding complex ecosystems. Such systems, often modeled as dynamical systems, typically exhibit noisy high-dimensional and non-stationary temporal behavior that renders their identification challenging. Existing dynamical system identification methods typically yield operators that accurately capture short-term behavior but fail to predict long-term trends, suggesting an incomplete capture of the underlying process. Methods that consider extended forecasts (e.g., recurrent neural networks) lack explicit representations of element-wise interactions and require substantial training data, thereby failing to capture interpretable network operators. Here we introduce Lookahead-driven Inference of Networked Operators for Continuous Stability (LINOCS), a robust learning procedure for identifying hidden dynamical interactions in noisy time-series data. LINOCS integrates several multi-step predictions with adaptive weights during training to recover dynamical operators that can yield accurate long-term predictions. We demonstrate LINOCS’ ability to recover the ground truth dynamical operators underlying synthetic time-series data for multiple dynamical systems models (including linear, piece-wise linear, time-changing linear systems’ decomposition, and regularized linear time-varying systems) as well as its capability to produce meaningful operators with robust reconstructions through various real-world examples
more »
« less
Assessing the predictability of nonlinear dynamics under smooth parameter changes
Short-term forecasts of nonlinear dynamics are important for risk-assessment studies and to inform sustainable decision-making for physical, biological and financial problems, among others. Generally, the accuracy of short-term forecasts depends upon two main factors: the capacity of learning algorithms to generalize well on unseen data and the intrinsic predictability of the dynamics. While generalization skills of learning algorithms can be assessed with well-established methods, estimating the predictability of the underlying nonlinear generating process from empirical time series remains a big challenge. Here, we show that, in changing environments, the predictability of nonlinear dynamics can be associated with the time-varying stability of the system with respect to smooth changes in model parameters, i.e. its local structural stability. Using synthetic data, we demonstrate that forecasts from locally structurally unstable states in smoothly changing environments can produce significantly large prediction errors, and we provide a systematic methodology to identify these states from data. Finally, we illustrate the practical applicability of our results using an empirical dataset. Overall, this study provides a framework to associate an uncertainty level with short-term forecasts made in smoothly changing environments.
more »
« less
- PAR ID:
- 10204792
- Date Published:
- Journal Name:
- Journal of The Royal Society Interface
- Volume:
- 17
- Issue:
- 162
- ISSN:
- 1742-5689
- Page Range / eLocation ID:
- 20190627
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Forecasting the El Niño-Southern Oscillation (ENSO) has been a subject of vigorous research due to the important role of the phenomenon in climate dynamics and its worldwide socioeconomic impacts. Over the past decades, numerous models for ENSO prediction have been developed, among which statistical models approximating ENSO evolution by linear dynamics have received significant attention owing to their simplicity and comparable forecast skill to first-principles models at short lead times. Yet, due to highly nonlinear and chaotic dynamics (particularly during ENSO initiation), such models have limited skill for longer-term forecasts beyond half a year. To resolve this limitation, here we employ a new nonparametric statistical approach based on analog forecasting, called kernel analog forecasting (KAF), which avoids assumptions on the underlying dynamics through the use of nonlinear kernel methods for machine learning and dimension reduction of high-dimensional datasets. Through a rigorous connection with Koopman operator theory for dynamical systems, KAF yields statistically optimal predictions of future ENSO states as conditional expectations, given noisy and potentially incomplete data at forecast initialization. Here, using industrial-era Indo-Pacific sea surface temperature (SST) as training data, the method is shown to successfully predict the Niño 3.4 index in a 1998–2017 verification period out to a 10-month lead, which corresponds to an increase of 3–8 months (depending on the decade) over a benchmark linear inverse model (LIM), while significantly improving upon the ENSO predictability “spring barrier”. In particular, KAF successfully predicts the historic 2015/16 El Niño at initialization times as early as June 2015, which is comparable to the skill of current dynamical models. An analysis of a 1300-yr control integration of a comprehensive climate model (CCSM4) further demonstrates that the enhanced predictability afforded by KAF holds over potentially much longer leads, extending to 24 months versus 18 months in the benchmark LIM. Probabilistic forecasts for the occurrence of El Niño/La Niña events are also performed and assessed via information-theoretic metrics, showing an improvement of skill over LIM approaches, thus opening an avenue for environmental risk assessment relevant in a variety of contexts.more » « less
-
null (Ed.)Prompt surveillance and forecasting of COVID-19 spread are of critical importance for slowing down the pandemic and for the success of any public mitigation efforts. However, as with any infectious disease with rapid transmission and high virulence, lack of COVID-19 observations for near-real-time forecasting is still the key challenge obstructing operational disease prediction and control. In this context, we can follow the two approaches to forecasting COVID-19 dynamics: based on mechanistic models and based on machine learning. Mechanistic models are better in capturing an epidemiological curve, using a low amount of data, and describing the overall trajectory of the disease dynamics, hence, providing long-term insights into where the disease might go. Machine learning, in turn, can provide more precise data-driven forecasts especially in the short-term horizons, while suffering from limited interpretability and usually requiring backlog history on the infectious disease. We propose a unified reinforcement learning framework that combines the two approaches. That is, long-term trajectory forecasts are used in machine learning techniques to forecast local variability which is not captured by the mechanistic model.more » « less
-
Abstract We investigated the predictability (forecast skill) of short-term droughts using the Palmer drought severity index (PDSI). We incorporated a sophisticated data training (of decadal range) to evaluate the improvement of forecast skill of short-term droughts (3 months). We investigated whether the data training of the synthetic North American Multi-Model Ensemble (NMME) climate has some influence on enhancing short-term drought predictability. The central elements are the merged information among PDSI and NMME with two postprocessing techniques. 1) The bias correction–spatial disaggregation (BC-SD) method improves spatial resolution by using a refined soil information introduced in the available water capacity of the PDSI calculation to assess water deficit that better estimates drought variability. 2) The ensemble model output statistic (EMOS) approach includes systematically trained decadal information of the multimodel ensemble simulations. Skill of drought forecasting improves when using EMOS, but BC-SD does not increase the forecast skill when compared with an analysis using BC (low spatial resolution). This study suggests that predictability forecast of drought (PDSI) can be extended without any change in the core dynamics of the model but instead by using the sophisticated EMOS postprocessing technique. We pointed out that using NMME without any postprocessing is of limited use in the suite of model variations of the NMME, at least for the U.S. Northeast. From our analysis, 1 month is the most extended range we should expect, which is below the range of the seasonal scale presented with EMOS (2 months). Thus, we propose a new design of drought forecasts that explicitly includes the multimodel ensemble signal.more » « less
-
Abstract This study illustrates the considerable improvement in accuracy achievable for long‐lead forecasts (18 months) of the Ocean Niño Index (ONI) through the utilization of a long short‐term memory (LSTM) machine learning algorithm. The research assesses the predictive potential of eight predictors from both tropical and extratropical regions constructed based on sea surface temperature, outgoing longwave radiation, sea surface height and zonal and meridional wind anomalies. In comparison to linear regression model forecasts, the LSTM model outperforms them for both the tropical and extratropical predictor sets. Among all the predictors, the western North Pacific (WNP) index demonstrates the highest prediction skill in ONI forecasts, followed by the North Tropical Atlantic (NTA) index and then the sea surface height index. While other predictors help the LSTM model to forecast either the phase variation of the amplitude variation of the observed ONI, the extratropical WNP predictor enables the LSTM model to forecast both variations. This superiority can be attributed to the involvement of SST anomalies in the WNP region in both tropical and extratropical El Niño–Southern Oscillation (ENSO) dynamics, allowing for the utilization of predictive potential from both components of ENSO dynamics. The study also concludes that the extratropical ENSO dynamics provide a robust source of predictability for long‐lead ENSO forecasts, which can be effectively harnessed using the LSTM model.more » « less
- Free Publicly Accessible Full Text
-
Accepted Manuscript1.0
- Journal Article:
- https://doi.org/10.1098/rsif.2019.0627
