Memory effects of climate and vegetation affecting net ecosystem CO2 fluxes in global forests

Universality of ecological memory for local and global net ecosystem exchange, atmospheric CO2, and sea surface temperature

Modeling global net ecosystem exchange is essential to understanding and quantifying the complex interactions between the Earth's terrestrial ecosystems and the atmosphere. Emphasizing the inter-relatedness between the global net ecosystem exchange, global sea surface temperature, and atmospheric CO 2 levels, intuitively suggests that all three systems may exhibit collective environmental memory. Motivated by this, we explicitly identified a collective memory function and showed a similar non-Markovian stochastic behavior for these systems exhibiting superdiffusive behavior in short time intervals. We obtained the values of the memory parameter, μ , and the characteristic frequencies, ν , for global net ecosystem exchange (GNEE) ( μ = 0.94 ± 0.03 , ν = 0.67 ± 0.08 / m o . ), global sea surface temperature (GSST) ( μ = 0.68 ± 0.11 , ν = 0.30 ± 0.18 / m o . ), and atmospheric CO 2 ( μ = 0.78 ± 0.08 , ν = 0.66 ± 0.13 / w k . ). The values of the memory parameter are within the range, 0 < μ < 1 , and thus all three systems are in the superdiffusive regime. We emphasize, further, that these results were consistent with our previous analyses at the ecosystem level (i.e. Great Barrier Reef) suggesting scale invariance for these phenomena. Thus, the observed superdiffusive behavior operating at different scales suggests universality of the collective memory function for these systems.

Long-term changes in forest response to extreme atmospheric dryness

Atmospheric dryness, as indicated by vapor pressure deficit (VPD), has a strong influence on forest greenhouse gas exchange with the atmosphere. In this study, we used long-term (10-30 years) net ecosystem productivity (NEP) measurements from 60 forest sites across the world (1003 site-years) to quantify long-term changes in forest NEP resistance and NEP recovery in response to extreme atmospheric dryness. We tested two hypotheses: first, across sites differences in NEP resistance and NEP recovery of forests will depend on both the biophysical characteristics (i.e., leaf area index [LAI] and forest type) of the forest as well as on the local meteorological conditions of the site (i.e., mean VPD of the site), and second, forests experiencing an increasing trend in frequency and intensity of extreme dryness will show an increasing trend in NEP resistance and NEP recovery over time due to emergence of long-term ecological stress memory. We used a data-driven statistical learning approach to quantify NEP resistance and NEP recovery over multiple years. Our results showed that forest types, LAI, and median local VPD conditions explained over 50% of variance in both NEP resistance and NEP recovery, with drier sites showing higher NEP resistance and NEP recovery compared to sites with less atmospheric dryness. The impact of extreme atmospheric dryness events on NEP lasted for up to 3 days following most severe extreme events in most forests, indicated by an NEP recovery of less than 100%. We rejected our second hypothesis as we found no consistent relationship between trends of extreme VPD with trends in NEP resistance and NEP recovery across different forest sites, thus an increase in atmospheric dryness as it is predicted might not increase the resistance or recovery of forests in terms of NEP.

Greenhouse gas fluxes (CO2, N2O and CH4) of pea and maize during two cropping seasons: Drivers, budgets, and emission factors for nitrous oxide

Agriculture contributes considerably to the increase of global greenhouse gas (GHG) emissions. Hence, magnitude and drivers of temporal variations in carbon dioxide (CO2), nitrous oxide (N2O) and methane (CH4) fluxes in croplands are urgently needed to develop sustainable, climate-smart agricultural practices. However, our knowledge of GHG fluxes from croplands is still very limited. The eddy covariance technique was used to quantify GHG budgets and N2O emission factors (EF) for pea and maize in Switzerland. The random forest technique was applied for gap-filling N2O and CH4 fluxes as well as to determine the relevance of environmental, vegetation vs. management drivers of the GHG fluxes during two cropping seasons. Environmental (i.e., net radiation, soil water content, soil temperature) and vegetation drivers (i.e., vegetation height) were more important drivers for GHG fluxes at field scale than time since management for the two crop species. Both crops acted as GHG sinks between sowing and harvest, clearly dominated by net CO2 fluxes, while CH4 emissions were negligible. However, considerable N2O emissions occurred in both crop fields early in the season when crops were still establishing. N2O fluxes in both crops were small later in the season when vegetation was tall, despite high soil water contents and temperatures. Results clearly show a strong and highly dynamic microbial-plant competition for N driving N2O fluxes at the field scale. The total loss was 1.4 kg N2O-N ha-1 over 55 days for pea and 4.8 kg N2O-N ha-1 over 127 days for maize. EFs of N2O were 1.5 % (pea) and 4.4 % (maize) during the cropping seasons, clearly exceeding the IPCC Tier 1 EF for N2O. Thus, sustainable, climate-smart agriculture needs to consider crop phenology and better adapt N supply to crop N demand for growth, particularly during the early cropping season when competition for N between establishing crops and soil microorganisms modulates N2O losses.

An Outlook for Deep Learning in Ecosystem Science

Rapid advances in hardware and software, accompanied by public- and private-sector investment, have led to a new generation of data-driven computational tools. Recently, there has been a particular focus on deep learning—a class of machine learning algorithms that uses deep neural networks to identify patterns in large and heterogeneous datasets. These developments have been accompanied by both hype and scepticism by ecologists and others. This review describes the context in which deep learning methods have emerged, the deep learning methods most relevant to ecosystem ecologists, and some of the problem domains they have been applied to. Deep learning methods have high predictive performance in a range of ecological contexts, leveraging the large data resources now available. Furthermore, deep learning tools offer ecosystem ecologists new ways to learn about ecosystem dynamics. In particular, recent advances in interpretable machine learning and in developing hybrid approaches combining deep learning and mechanistic models provide a bridge between pure prediction and causal explanation. We conclude by looking at the opportunities that deep learning tools offer ecosystem ecologists and assess the challenges in interpretability that deep learning applications pose.

Google Earth Engine and Artificial Intelligence (AI): A Comprehensive Review

Remote sensing (RS) plays an important role gathering data in many critical domains (e.g., global climate change, risk assessment and vulnerability reduction of natural hazards, resilience of ecosystems, and urban planning). Retrieving, managing, and analyzing large amounts of RS imagery poses substantial challenges. Google Earth Engine (GEE) provides a scalable, cloud-based, geospatial retrieval and processing platform. GEE also provides access to the vast majority of freely available, public, multi-temporal RS data and offers free cloud-based computational power for geospatial data analysis. Artificial intelligence (AI) methods are a critical enabling technology to automating the interpretation of RS imagery, particularly on object-based domains, so the integration of AI methods into GEE represents a promising path towards operationalizing automated RS-based monitoring programs. In this article, we provide a systematic review of relevant literature to identify recent research that incorporates AI methods in GEE. We then discuss some of the major challenges of integrating GEE and AI and identify several priorities for future research. We developed an interactive web application designed to allow readers to intuitively and dynamically review the publications included in this literature review.

Tropical extreme droughts drive long-term increase in atmospheric CO2 growth rate variability

The terrestrial carbon sink slows the accumulation of carbon dioxide (CO2) in the atmosphere by absorbing roughly 30% of anthropogenic CO2 emissions, but varies greatly from year to year. The resulting variations in the atmospheric CO2 growth rate (CGR) have been related to tropical temperature and water availability. The apparent sensitivity of CGR to tropical temperature ([Formula: see text]) has changed markedly over the past six decades, however, the drivers of the observation to date remains unidentified. Here, we use atmospheric observations, multiple global vegetation models and machine learning products to analyze the cause of the sensitivity change. We found that a threefold increase in [Formula: see text] emerged due to the long-term changes in the magnitude of CGR variability (i.e., indicated by one standard deviation of CGR; STDCGR), which increased 34.7% from 1960-1979 to 1985-2004 and subsequently decreased 14.4% in 1997-2016. We found a close relationship (r2 = 0.75, p < 0.01) between STDCGR and the tropical vegetated area (23°S - 23°N) affected by extreme droughts, which influenced 6-9% of the tropical vegetated surface. A 1% increase in the tropical area affected by extreme droughts led to about 0.14 Pg C yr-1 increase in STDCGR. The historical changes in STDCGR were dominated by extreme drought-affected areas in tropical Africa and Asia, and semi-arid ecosystems. The outsized influence of extreme droughts over a small fraction of vegetated surface amplified the interannual variability in CGR and explained the observed long-term dynamics of [Formula: see text].

Model Selection for Ecosystem Respiration Needs to Be Site Specific: Lessons from Grasslands on the Mongolian Plateau

Selecting an appropriate model for simulating ecosystem respiration is critical in modeling the carbon cycle of terrestrial ecosystems due to their magnitude and high variations in time and space. There is no consensus on the ideal model for estimating ecosystem respiration in different ecosystems. We evaluated the performances of six respiration models, including Arrhenius, logistic, Gamma, Martin, Concilio, and time series model, against measured ecosystem respiration during 2014–2018 in four grassland ecosystems on the Mongolian Plateau: shrubland, dry steppe, temperate steppe, and meadow ecosystems. Ecosystem respiration increased exponentially with soil temperature within an apparent threshold of ~19.62 °C at shrubland, ~16.05 °C at dry steppe, ~16.92 °C at temperate steppe, and ~15.03 °C at meadow. The six models explained approximately 50–80% of the variabilities of ecosystem respiration during the study period. Both soil temperature and soil moisture played considerable roles in simulating ecosystem respiration with R square, ranging from 0.5 to 0.8. The Martin model performed better than the other models, with a relatively high R square, i.e., R2 = 0.68 at shrubland, R2 = 0.57 at dry steppe, R2 = 0.74 at temperate steppe, and R2 = 0.81 at meadow. These models achieved good performance for around 50–80% of the simulations. No single model performs best for all four grassland types, while each model appears suitable for at least one type of ecosystem. Models that oil moisture include models, especially the Martin model, are more suitable for the accurate prediction of ecosystem respiration than Ts-only models for the four grassland ecosystems.

Characterizing and forecasting the responses of tropical forest leaf phenology to El Nino by machine learning algorithms

Climate change and global warming have serious adverse impacts on tropical forests. In particular, climate change may induce changes in leaf phenology. However, in tropical dry forests where tree diversity is high, species responses to climate change differ. The objective of this research is to analyze the impact of climate variability on the leaf phenology in Thailand's tropical forests. Machine learning approaches were applied to model how leaf phenology in dry dipterocarp forest in Thailand responds to climate variability and El Niño. First, we used a Self-Organizing Map (SOM) to cluster mature leaf phenology at the species level. Then, leaf phenology patterns in each group along with litterfall phenology and climate data were analyzed according to their response time. After that, a Long Short-Term Memory neural network (LSTM) was used to create model to predict leaf phenology in dry dipterocarp forest. The SOM-based clustering was able to classify 92.24% of the individual trees. The result of mapping the clustering data with lag time analysis revealed that each cluster has a different lag time depending on the timing and amount of rainfall. Incorporating the time lags improved the performance of the litterfall prediction model, reducing the average root mean square percent error (RMSPE) from 14.35% to 12.06%. This study should help researchers understand how each species responds to climate change. The litterfall prediction model will be useful for managing dry dipterocarp forest especially with regards to forest fires.

Seasonal biological carryover dominates northern vegetation growth

The state of ecosystems is influenced strongly by their past, and describing this carryover effect is important to accurately forecast their future behaviors. However, the strength and persistence of this carryover effect on ecosystem dynamics in comparison to that of simultaneous environmental drivers are still poorly understood. Here, we show that vegetation growth carryover (VGC), defined as the effect of present states of vegetation on subsequent growth, exerts strong positive impacts on seasonal vegetation growth over the Northern Hemisphere. In particular, this VGC of early growing-season vegetation growth is even stronger than past and co-occurring climate on determining peak-to-late season vegetation growth, and is the primary contributor to the recently observed annual greening trend. The effect of seasonal VGC persists into the subsequent year but not further. Current process-based ecosystem models greatly underestimate the VGC effect, and may therefore underestimate the CO2 sequestration potential of northern vegetation under future warming.

Climate-Driven Variability and Trends in Plant Productivity Over Recent Decades Based on Three Global Products

Variability in climate exerts a strong influence on vegetation productivity (gross primary productivity; GPP), and therefore has a large impact on the land carbon sink. However, no direct observations of global GPP exist, and estimates rely on models that are constrained by observations at various spatial and temporal scales. Here, we assess the consistency in GPP from global products which extend for more than three decades; two observation-based approaches, the upscaling of FLUXNET site observations (FLUXCOM) and a remote sensing derived light use efficiency model (RS-LUE), and from a suite of terrestrial biosphere models (TRENDYv6). At local scales, we find high correlations in annual GPP among the products, with exceptions in tropical and high northern latitudes. On longer time scales, the products agree on the direction of trends over 58% of the land, with large increases across northern latitudes driven by warming trends. Further, tropical regions exhibit the largest interannual variability in GPP, with both rainforests and savannas contributing substantially. Variability in savanna GPP is likely predominantly driven by water availability, although temperature could play a role via soil moisture-atmosphere feedbacks. There is, however, no consensus on the magnitude and driver of variability of tropical forests, which suggest uncertainties in process representations and underlying observations remain. These results emphasize the need for more direct long-term observations of GPP along with an extension of in situ networks in underrepresented regions (e.g., tropical forests). Such capabilities would support efforts to better validate relevant processes in models, to more accurately estimate GPP.

Identifying Dynamic Memory Effects on Vegetation State Using Recurrent Neural Networks

Vegetation state is largely driven by climate and the complexity of involved processes leads to non-linear interactions over multiple time-scales. Recently, the role of temporally lagged dependencies, so-called memory effects, has been emphasized and studied using data-driven methods, relying on a vast amount of Earth observation and climate data. However, the employed models are often not able to represent the highly non-linear processes and do not represent time explicitly. Thus, data-driven study of vegetation dynamics demands new approaches that are able to model complex sequences. The success of Recurrent Neural Networks (RNNs) in other disciplines dealing with sequential data, such as Natural Language Processing, suggests adoption of this method for Earth system sciences. Here, we used a Long Short-Term Memory (LSTM) architecture to fit a global model for Normalized Difference Vegetation Index (NDVI), a proxy for vegetation state, by using climate time-series and static variables representing soil properties and land cover as predictor variables. Furthermore, a set of permutation experiments was performed with the objective to identify memory effects and to better understand the scales on which they act under different environmental conditions. This was done by comparing models that have limited access to temporal context, which was achieved through sequence permutation during model training. We performed a cross-validation with spatio-temporal blocking to deal with the auto-correlation present in the data and to increase the generalizability of the findings. With a full temporal model, global NDVI was predicted with R² of 0.943 and RMSE of 0.056. The temporal model explained 14% more variance than the non-memory model on global level. The strongest differences were found in arid and semiarid regions, where the improvement was up to 25%. Our results show that memory effects matter on global scale, with the strongest effects occurring in sub-tropical and transitional water-driven biomes.

Direct Prediction of the Toxic Gas Diffusion Rule in a Real Environment Based on LSTM

Predicting the diffusion rule of toxic gas plays a distinctly important role in emergency capability assessment and rescue work. Among diffusion prediction models, the traditional artificial neural network has exhibited excellent performance not only in prediction accuracy but also in calculation time. Nevertheless, with the continuous development of deep learning and data science, some new prediction models based on deep learning algorithms have been shown to be more advantageous because their structure can better discover internal laws and external connections between input data and output data. The long short-term memory (LSTM) network is a kind of deep learning neural network that has demonstrated outstanding achievements in many prediction fields. This paper applies the LSTM network directly to the prediction of toxic gas diffusion and uses the Project Prairie Grass dataset to conduct experiments. Compared with the Gaussian diffusion model, support vector machine (SVM) model, and back propagation (BP) network model, the LSTM model of deep learning has higher prediction accuracy (especially for the prediction at the point of high concentration values) while avoiding the occurrence of negative concentration values and overfitting problems found in traditional artificial neural network models.