An official website of the United States government
Abstract Resilience is the ability of ecosystems to maintain function while experiencing perturbation. Globally, forests are experiencing disturbances of unprecedented quantity, type, and magnitude that may diminish resilience. Early warning signals are statistical properties of data whose increase over time may provide insights into decreasing resilience, but there have been few applications to forests. We quantified four early warning signals (standard deviation, lag-1 autocorrelation, skewness, and kurtosis) across detrended time series of multiple ecosystem state variables at the Hubbard Brook Experimental Forest, New Hampshire, USA and analyzed how these signals have changed over time. Variables were collected over periods from 25 to 55 years from both experimentally manipulated and reference areas and were aggregated to annual timesteps for analysis. Long-term (>50 year) increases in early warning signals of stream calcium, a key biogeochemical variable at the site, illustrated declining resilience after decades of acid deposition, but only in watersheds that had previously been harvested. Trends in early warning signals of stream nitrate, a critical nutrient and water pollutant, likewise exhibited symptoms of declining resilience but in all watersheds. Temporal trends in early warning signals of some of groups of trees, insects, and birds also indicated changing resilience, but this pattern differed among, and even within, groups. Overall, ∼60% of early warning signals analyzed indicated decreasing resilience. Most of these signals occurred in skewness and kurtosis, suggesting ‘flickering’ behavior that aligns with emerging evidence of the forest transitioning into an oligotrophic condition. The other ∼40% of early warning signals indicated increasing or unchanging resilience. Interpretation of early warning signals in the context of system specific knowledge is therefore essential. They can be useful indicators for some key ecosystem variables; however, uncertainties in other variables highlight the need for further development of these tools in well-studied, long-term research sites.
more »
« less
Evaluating the performance of temporal and spatial early warning statistics of algal blooms
Abstract Regime shifts have large consequences for ecosystems and the services they provide. However, understanding the potential for, causes of, proximity to, and thresholds for regime shifts in nearly all settings is difficult. Generic statistical indicators of resilience have been proposed and studied in a wide range of ecosystems as a method to detect when regime shifts are becoming more likely without direct knowledge of underlying system dynamics or thresholds. These early warning statistics (EWS) have been studied separately but there have been few examples that directly compare temporal and spatial EWS in ecosystem‐scale empirical data. To test these methods, we collected high‐frequency time series and high‐resolution spatial data during a whole‐lake fertilization experiment while also monitoring an adjacent reference lake. We calculated two common EWS, standard deviation and autocorrelation, in both time series and spatial data to evaluate their performance prior to the resulting algal bloom. We also applied the quickest detection method to generate binary alarms of resilience change from temporal EWS. One temporal EWS, rolling window standard deviation, provided advanced warning in most variables prior to the bloom, showing trends and between‐lake patterns consistent with theory. In contrast, temporal autocorrelation and both measures of spatial EWS (spatial SD, Moran's I) provided little or no warning. By compiling time series data from this and past experiments with and without nutrient additions, we were able to evaluate temporal EWS performance for both constant and changing resilience conditions. True positive alarm rates were 2.5–8.3 times higher for rolling window standard deviation when a lake was being pushed towards a bloom than the rate of false positives when it was not. For rolling window autocorrelation, alarm rates were much lower and no variable had a higher true positive than false positive alarm rate. Our findings suggest temporal EWS provide advanced warning of algal blooms and that this approach could help managers prepare for and/or minimize negative bloom impacts.
more »
« less
- PAR ID:
- 10371137
- Publisher / Repository:
- Wiley Blackwell (John Wiley & Sons)
- Date Published:
- Journal Name:
- Ecological Applications
- Volume:
- 32
- Issue:
- 5
- ISSN:
- 1051-0761
- Format(s):
- Medium: X
- Sponsoring Org:
- National Science Foundation
More Like this
-
-
Abstract Research aimed at identifying indicators of persistent abrupt shifts in ecological communities, a.k.a regime shifts, has led to the development of a suite of early warning signals (EWSs). As these often perform inaccurately when applied to real-world observational data, it remains unclear whether critical transitions are the dominant mechanism of regime shifts and, if so, which EWS methods can predict them. Here, using multi-trophic planktonic data on multiple lakes from around the world, we classify both lake dynamics and the reliability of classic and second generation EWSs methods to predict whole-ecosystem change. We find few instances of critical transitions, with different trophic levels often expressing different forms of abrupt change. The ability to predict this change is highly processing dependant, with most indicators not performing better than chance, multivariate EWSs being weakly superior to univariate, and a recent machine learning model performing poorly. Our results suggest that predictive ecology should start to move away from the concept of critical transitions, developing methods suitable for predicting resilience loss not limited to the strict bounds of bifurcation theory.more » « less
-
Abstract This study explores the behavior of machine-learning-based flare forecasting models deployed in a simulated operational environment. Using Georgia State University’s Space Weather Analytics for Solar Flares benchmark data set, we examine the impacts of training methodology and the solar cycle on decision tree, support vector machine, and multilayer perceptron performance. We implement our classifiers using three temporal training windows: stationary, rolling, and expanding. The stationary window trains models using a single set of data available before the first forecasting instance, which remains constant throughout the solar cycle. The rolling window trains models using data from a constant time interval before the forecasting instance, which moves with the solar cycle. Finally, the expanding window trains models using all available data before the forecasting instance. For each window, a number of input features (1, 5, 10, 25, 50, and 120) and temporal sizes (5, 8, 11, 14, 17, and 20 months) were tested. To our surprise, we found that, for a window of 20 months, skill scores were comparable regardless of the window type, feature count, and classifier selected. Furthermore, reducing the size of this window only marginally decreased stationary and rolling window performance. This implies that, given enough data, a stationary window can be chosen over other window types, eliminating the need for model retraining. Finally, a moderately strong positive correlation was found to exist between a model’s false-positive rate and the solar X-ray background flux. This suggests that the solar cycle phase has a considerable influence on forecasting.more » « less
-
Abstract In this era of global environmental change and rapid regime shifts, managing core areas that species require to survive and persist is a grand challenge for conservation. Wildlife monitoring data are often limited or local in scale. The emerging ability to map and track spatial regimes (i.e., the spatial manifestation of state transitions) using advanced geospatial vegetation data has the potential to provide earlier warnings of habitat loss because many species of conservation concern strongly avoid spatial regime boundaries. Using 23 yr of data for the lek locations of Greater Prairie‐Chicken (Tympanuchus cupido; GPC) in a remnant grassland ecosystem, we demonstrate how mapping changes in the boundaries between grassland and woodland spatial regimes provide a spatially explicit early warning signal for habitat loss for an iconic and vulnerable grassland‐obligate known to be highly sensitive to woody plant encroachment. We tested whether a newly proposed metric for the quantification of spatial regimes captured well‐known responses of GPC to woody plant expansion into grasslands. Resource selection functions showed that the grass:woody spatial regime boundary strength explained the probability of 80% of relative lek occurrence, and GPC strongly avoided grass:woody spatial regime boundaries at broad scales. Both findings are consistent with well‐known expectations derived from GPC ecology. These results provide strong evidence for vegetation‐derived delineations of spatial regimes to serve as generalized signals of early warning for state transitions that have major consequences to biodiversity conservation. Mapping spatial regime boundaries over time provided interpretable early warnings of habitat loss. Woody plant regimes displaced grassland regimes starting from the edges of the study area and constricting inward. Correspondingly, the relative probability of lek occurrence constricted in space. Similarly, the temporal trajectory of spatial regime boundary strength increased over time and moved closer to the observed limit of GPC lek site usage relative to grass:woody boundary strength. These novel spatial metrics allow managers to rapidly screen for early warning signals of spatial regime shifts and adapt management practices to defend and grow habitat cores at broad scales.more » « less
-
Climate change is predicted to intensify lake algal blooms globally and result in regime shifts. However, observed increases in algal biomass do not consistently correlate with air temperature or precipitation, and evidence is lacking for a causal effect of climate or the nonlinear dynamics needed to demonstrate regime shifts. We modeled the causal effects of climate on annual lake chlorophyll (a measure of algal biomass) over 34 y for 24,452 lakes across broad ecoclimatic zones of the United States and evaluated the potential for regime shifts. We found that algal biomass was causally related to climate in 34% of lakes. In these cases, 71% exhibited abrupt but mostly temporary shifts as opposed to persistent changes, 13% had the potential for regime shifts. Climate was causally related to algal biomass in lakes experiencing all levels of human disturbance, but with different likelihood. Climate causality was most likely to be observed in lakes with minimal human disturbance and cooler summer temperatures that have increased over the 34 y studied. Climate causality was variable in lakes with low to moderate human disturbance, and least likely in lakes with high human disturbance, which may mask climate causality. Our results explain some of the previously observed heterogeneous climate responses of lake algal biomass globally and they can be used to predict future climate effects on lakes.more » « less
