Abstract Heatwaves threaten ecosystem carbon balances, yet the mechanisms driving short-term carbon flux responses remain poorly understood. Here, integrating high-frequency eddy covariance (EC) data from 140 global flux tower sites (872 site-years) with detailed process-based modeling, we examine ecosystem responses during and immediately after heatwaves. We show that heatwaves caused a −40% (range [−29%, −128%]) reduction in net ecosystem productivity (NEP) compared to pre-heatwave values, with this reduction persisting over the following two weeks (−38% range [+3%, −154%]). We attributed NEP decreases to photosynthesis decreases more than to ecosystem respiration (RE) increases. Forest sites had greater NEP decreases during heatwaves than non-forest sites, but remained carbon sinks afterwards, indicating resilience. Our modeling analysis of extreme heatwaves at selected EC sites shows that decreased photosynthesis, increased maintenance respiration, and decreased plant non-structural carbon reserves during heatwaves drive carbon cycle changes that persist for weeks. Consistent with phenocam observations, we modeled a reduction in leaf area index caused by reduced non-structural carbon reserves, leading to early leaf senescence and longer-term impacts. Ongoing increases in heatwaves are therefore likely to reduce NEP across a range of ecosystems, exacerbating carbon cycle feedback.

Abstract Ongoing climate change has increased the frequency and intensity of climate extremes such as heatwaves, impacting terrestrial ecosystems. Grasslands are often shaped by human activities such as mowing, which may modulate their responses to climate extremes. However, the mechanisms underlying such responses and the factors important in stabilizing grassland functioning under environmental disturbance are currently poorly understood. In this study, we experimentally compared the effects of heatwaves and mowing on the functioning (based on CO2 exchange) of two different grassland types, Stipa krylovii typical grassland (Sti‐Tpl) and Leymus chinensis meadow steppe (Ley‐Mdw), in Inner Mongolia. In each grassland, ecosystem CO2 fluxes and plant community characteristics (biomass, community structure and biodiversity indices) were recorded. We specifically focused on the stability of grassland CO2 exchange during heatwave events (resistance), the capacity to regain functionality afterwards (recovery) and the plant factors influencing these resilience metrics in both grasslands. The results indicate non‐linear temporal trajectories in carbon flux recovery, with a weak correlation between resistance and recovery. Ecosystem respiration (RE) generally exhibited greater resistance and recovery to heatwaves than gross ecosystem production (GEP); GEP reduction led to a decrease in net ecosystem production (NEP). However, local mowing practices partially mitigated these negative effects. The importance value (Iv) of dominant species and biodiversity both positively influenced NEP resistance, but their effects on NEP recovery were opposite, as Iv enhanced recovery while biodiversity reduced it. Finally, the contribution level of dominant species to grassland stability was closely related to their Iv in the plant community. Synthesis. In this study, we investigated the complex factors influencing ecosystem resistance and recovery to heatwaves and mowing in two distinct grasslands. We found that the physiology, morphology and regeneration traits of the dominant species in each grassland community explained most of the divergence of grassland function stability. Thus, to buffer ecosystems against adverse impacts of climate extremes in conjunction with land management, it can be advantageous to focus on the maintenance or selection of dominant species rather than solely on increasing species richness.

ABSTRACT In this study, the rates of ingestion (IR), respiration (RR), ammonia excretion (AR), faecal production (FR), and calcification (CR) of green mussel (Perna viridis) and pearl oyster (Pinctada fucata) were determined across four different temperature levels. Additionally, the carbon budgets of these two bivalves under different temperatures were also modelled. The IR, RR, AR, FR, and CR of both green mussels and pearl oysters were significantly affected by temperature changes. The peak values for these parameters were observed at 26°C, except for the CR of the green mussel and the AR of the pearl oyster. At temperatures ranging from 22°C to 30°C, the IR, RR, AR, and FR of the pearl oyster were significantly higher than those of the green mussel, whereas the opposite trend was observed for the CR. Specifically, negative CR values were recorded in both species at 18°C. The growth carbon (GC) ratio of green mussels initially increased and then decreased as the temperature rose, whereas the GC ratio of pearl oysters remained relatively stable across varying temperatures. The GC ratios of pearl oysters were higher than those of green mussels, while green mussels displayed higher faecal production carbon ratios than pearl oysters.

Prolonged inundations are altering coastal forest ecosystems of the southeastern US, causing extensive tree die-offs and the development of ghost forests. This hydrological stressor also alters carbon fluxes, threatening the stability of coastal carbon sinks. This study was conducted to investigate the interactions between hydrological drivers and ecosystem responses by analyzing daily eddy covariance flux data from a wetland forest in North Carolina, USA, spanning 2009–2019. We analyzed temporal patterns of net ecosystem exchange (NEE), gross primary productivity (GPP), and ecosystem respiration (RE) under both flooded and non-flooded conditions and evaluated their relationships with observed tree mortality. Generalized Additive Modeling (GAM) revealed that groundwater table depth (GWT), leaf area index (LAI), NEE, and net radiation (Rn) were key predictors of mortality transitions (R2 = 0.98). Elevated GWT induces root anoxia; declining LAI reduces productivity; elevated NEE signals physiological breakdown; and higher Rn may amplify evapotranspiration stress. Receiver Operating Characteristic (ROC) analysis revealed critical early warning thresholds for tree mortality: GWT = 2.23 cm, LAI = 2.99, NEE = 1.27 g C m−2 d−1, and Rn = 167.54 W m−2. These values offer a basis for forecasting forest mortality risk and guiding early warning systems. Our findings highlight the dominant role of hydrological variability in ecosystem degradation and offer a threshold-based framework for early detection of mortality risks. This approach provides insights into managing coastal forest resilience amid accelerating sea level rise.

Soil respiration (RS) comprises terrestrial ecosystems’ second-largest carbon flux. Yet, methodological errors in RS partitioning and uncertainties in seasonal responses of RS make it difficult to predict future RS. Here, we tested the assumption of RS partitioning (similar microbial respiration between planted and root-free soils), and explored two components of RS, autotrophic and heterotrophic respiration (RA, RH, respectively), in a temperate grassland under monsoon continental climate. Microbial respiration in soils from planted plots was 3.88 times higher than that from root-free plots during lab incubation. In field, RH:RS ratio was relatively low during non-monsoon, but increased during monsoon. The RH was more sensitive to temperature than RS, indicating a greater Q10 of RH than that of estimated RA. The annual RH:RS excluding the monsoon period was comparable to those reported in the global Soil Respiration Database (SRDB) and other Korean literature. This study highlights that the assumption of RS partitioning can be violated, that RH exhibits a greater sensitivity to changes in temperature and soil water content than RA, and that annual RH:RS may be similar across the globe when extreme precipitation (e.g., monsoon) is excluded.

Abstract Coastal wetlands are increasingly vital carbon sinks, helping mitigate atmospheric CO2 and slow global warming. However, we have limited knowledge about the carbon sink capacity of coastal wetlands, whereby developing advanced skills for predicting CO2 fluxes of coastal wetlands is critical. Here, by employing recent cutting-edge achievements in artificial intelligence, we evaluated three automated machine learning (AutoML) platforms, including Lazy Predict, H2O AutoML and fast and lightweight automated machine learning, for predicting monthly gross primary production (GPP), ecosystem respiration (RE), and net ecosystem exchange (NEE) in China’s of mangrove and saltmarsh coastal wetlands with multi-source satellite observations. Our results indicate that these AutoML platforms effectively predicted GPP, RE, and NEE, with superior performance for GPP and RE compared to NEE. For individual predictions across 14 sites, the testing set yielded average determination coefficient (R 2) values of 0.74, 0.79, and 0.63, and root mean square error values of 0.83, 0.45, and 0.76 gC m−2s−1 for GPP, RE, and NEE, respectively. Cross-site predictions performed better for saltmarsh (average R 2: 0.86, 0.84, and 0.76 for GPP, RE, and NEE) than mangrove ecosystems (average R 2: 0.72, 0.76, and 0.59). In addition, ensemble ML models, particularly on the Lazy Predict platform, significantly outperformed individual models. Feature important analyses revealed that vegetation variables (leaf area index and fraction of absorbed photosynthetically active radiation) play pronouncedly important roles in mangrove ecosystems, followed by climate variables (air temperature (Ta) and precipitation) with considerably important roles, while Ta dominated in saltmarsh ecosystems, with vegetation variables but playing a lesser role. Our study offers valuable insights for utilizing AutoML techniques to enhance CO2 flux predictions and regional budget estimations for coastal wetlands, potentially advancing strategies for monitoring large-scale coastal ‘blue carbon’ dynamics.

Forest–atmosphere interactions through mass and energy fluxes significantly influence climate processes. However, due to anthropogenic actions, native Araucaria forests in southern Brazil, part of the Atlantic Forest biome, have been drastically reduced. This study quantifies CO2 and energy flux contributions from each forest stratum to improve understanding of surface–atmosphere interactions. Eddy covariance data from November 2009 to April 2012 were used to assess fluxes in an Araucaria forest in Paraná, Brazil, across the ecosystem, understory, and overstory strata. On average, the ecosystem acts as a carbon sink of −298.96 g C m−2 yr−1, with absorption doubling in spring–summer compared to autumn–winter. The understory primarily acts as a source, while the overstory functions as a CO2 sink, driving carbon absorption. The overstory contributes 63% of the gross primary production (GPP) and 75% of the latent heat flux, while the understory accounts for 94% of the ecosystem respiration (RE). The energy fluxes exhibited marked seasonality, with higher latent and sensible heat fluxes in summer, with sensible heat predominantly originating from the overstory. Annual ecosystem evapotranspiration reaches 1010 mm yr−1: 60% of annual precipitation. Water-use efficiency is 2.85 g C kgH2O−1, with higher values in autumn–winter and in the understory. The influence of meteorological variables on the fluxes was analyzed across different scales and forest strata, showing that solar radiation is the main driver of daily fluxes, while air temperature and vapor pressure deficit are more relevant at monthly scales. This study highlights the overstory’s dominant role in carbon absorption and energy fluxes, reinforcing the need to preserve these ecosystems for their crucial contributions to climate regulation and water-use efficiency.

From lab to real-life: A three-stage validation of wearable technology for stress monitoring.

The rapid expansion of tea plantations in the hilly regions of southeastern China significantly impacts regional carbon cycle. The Biome-BGC model, commonly used to quantify carbon fluxes, lacks sufficient representation of artificial management processes. We integrated the measured and remote-sensed leaf area index (LAI) to improve the Biome-BGC model, enhancing its simulation capabilities for the artificial management processes in tea plantations. The results showed that LAI was a crucial intermediate variable in the Biome-BGC model. Accurate simulation of LAI was the key to improve the model's precision in simulating carbon fluxes in tea plantations. The improved model significantly enhanced the simulation accuracy of gross primary productivity (GPP) and ecosystem respiration (RE), with 5-year average GPP and RE values of 1.26 and 1.19 kg C·m-2, respectively. The daily-scale R2 values reached 0.55 and 0.80, representing an increase of 44.5% for GPP and a decrease of 0.9% for RE compared to the original model. The root mean square error (RMSE) values were 0.887 and 1.030 g C·m-2·d-1, representing reductions of 50.3% for GPP and 68.4% for RE compared to the original model, respectively. At the month scale, the improved model significantly reduced the overestimation of original model resulted from insufficient representation of artificial pruning for tea plantations. The improved model could dynamically depict the impact of LAI fluctuations caused by pruning on the carbon cycle and its applicability across different time scales had been verified, which would provide technical support for quantitative research on carbon cycling in tea plantations with high-intensity anthropogenic management.

Identifying driver inattention is crucial for road safety, driver well-being and can be enhanced using multimodal physiological signals. However, effective fusion of multimodal data is highly challenging, particularly with intermediate fusion, where data fusion may become less informative. In this study, we address this challenge by employing a 1D Convolutional Neural Network (CNN) with Cross Hierarchical Attention Fusion on multimodal data. For this purpose, electrocardiogram (ECG) (256 Hz) and respiration (RESP) (128 Hz) signals were obtained from subjects (N=10) using textile electrodes while driving in various scenarios, specifically normal and calling. The acquired multimodal data were preprocessed, hierarchically fused, and subjected to a cross-attention mechanism to identify driver inattention. Experiments were conducted using Leave-One-Subject-Out Cross-Validation (LOSOCV). The proposed approach is able to classify driver inattention states. It was observed that shorter data segments yielded higher accuracy compared to longer segments. Additionally, multimodal data from textile electrodes effectively discriminated between driver inattention states. Therefore, the proposed approach utilizing wearable smart shirts enables non-intrusive monitoring in real-world driving scenarios.

The disposal of urban tree litter as waste has significant implications for material cycles, energy flows, and global climate change within urban ecosystems. However, the species-specific contributions of urban trees to atmospheric CO2 emissions through soil respiration (RS) remain poorly understood. This study investigates the effects of litter management on RS dynamics in urban green spaces, focusing on six common species (Mangifera indica, Ficus microcarpa, Cinnamomum camphora, Bauhinia purpurea, Triadica sebifera, and Celtis sinensis) in Fuzhou, China. Three litter treatments—litter retention (CK), litter removal (RL), and litter doubling (DL)—were established to monitor monthly RS fluctuations. Results indicate that DL significantly increased RS rates, while RL reduced them. The increase in RS due to litter addition was more pronounced than the decrease caused by litter removal for most species. RS rates exhibited a unimodal seasonal pattern, peaking in summer. Furthermore, litter treatments influenced the temperature sensitivity coefficient (Q10), with F. microcarpa showing the highest average Q10 (4.16) and M. indica the lowest (1.88). This study underscores the critical role of litter input in modulating RS in urban green spaces and highlights the joint but asymmetric effects of soil temperature and moisture on RS dynamics.

The extreme conditions in arid ecosystems make these environments sensitive to environmental changes. Particularly, land use and seasonal changes are determinants of their soil carbon dynamics. The effect of those elements on soil respiration (RS) is still poorly known in several arid regions of the world. This study investigates the seasonal effect on the RS and its controlling factors throughout different land use systems in northeastern Mexico. RS and 34 biotic and abiotic variables were measured across agricultural crops, natural shrublands, livestock farms, walnut orchards, and industrially influenced soils during the dry and wet seasons. Six variables (soil water content, soil organic matter, soil temperature, silt, and pH) were found as drivers of RS on both local and regional scales. Seasonal and land use had a transversal effect on RS and its controlling factors. RS dynamics were primarily modulated by soil water content, with the wet season and managed lands showing increased sensitivity to climatic and anthropogenic changes. These results indicate that land management strategies are critical for carbon cycling, particularly in water-limited regions like northeastern Mexico, where land use changes are occurring at an accelerated pace.

Narrowing uncertainties associated with land–atmosphere carbon (C) fluxes is critical for projecting climate futures, but large uncertainties in modeling soil respiration (RS) hinder progress. Difficulties accounting for how biological communities will respond to altered precipitation contribute to those uncertainties, but remain underexplored in situ. In a rainfall and grassland diversity manipulation experiment altering both plant richness and community composition, we measured RS monthly for four growing seasons, along with multiple physical (soil moisture and temperature) and biological drivers (aboveground, root, and microbial biomass) of RS. Relationships between plant richness and RS were dependent on plant community composition and soil moisture conditions. Elevated RS was associated with grass diversity, likely governed by enhanced soil moisture at 12 cm. Microbial biomass was the strongest independent predictor of RS. Though soil moisture was a strong predictor of RS, covariance with precipitation treatments and microbial biomass suggests it operated through multiple indirect pathways. Even after accounting for several RS drivers, plant community composition and richness still accounted for a nontrivial amount of variation in RS. This suggests that unexplored pathways associated with biological complexity (for example, microbial community composition) influence RS. Finally, altered precipitation changed diversity-RS relationships over time, suggesting that soil microbes can respond relatively rapidly to altered precipitation, perhaps due to the diversity of specialist microbes in our initial common soils. Our work demonstrates how biological complexity can interact with physical drivers and changing climates to influence RS in ways currently unaccounted for in models.

An in-depth understanding of carbon dynamics and ecosystem productivity is essential for conservation and management of different ecosystems. Ecosystem dynamics and carbon budget are assessed by estimating net ecosystem production (NEP) across different global ecosystems. An ecological productivity assessment of forest and floating meadow ecosystems in Keibul Lamjao National Park (KLNP), Manipur, North East India, was conducted using the multi-criteria decision-making process namely, gray relational analysis (GRA). The analysis was performed on 24 selected criterions classified either as "higher-the-better" or "lower-the-better" based on their degree of influence on the carbon budget. Floating meadows exhibited a higher production of aboveground and belowground biomass and a higher total mortality and decay. Furthermore, the study found that floating meadows exhibited a higher soil organic carbon (SOC) and net soil organic matter (SOM) than the forest ecosystem. The forest ecosystem showed higher total respiration (RT), heterotrophic respiration (RH), and autotrophic respiration (RA) than floating meadows. Floating meadows exhibited a higher net primary productivity (NPP) of 616.49 ± 33.87 gCm-2year-1 than the forest ecosystem, which has a NPP of 566.64 ± 65.26 gCm-2year-1. Similarly, floating meadows have higher NEP (495.25 ± 36.46 gCm-2year-1) than forest ecosystems (418.39 ± 65.76 gCm-2year-1). These characteristics have a significant influence on the carbon budget in floating meadows as compared to forest ecosystems, as shown by larger values of gray relational coefficient (GRC) in GRA. The floating meadows ecosystem (0.82) obtained 54.72% gain in gray relational grades (GRG) value with the forest ecosystem (0.53). This study might help in improving KLNP and other adjutant areas for conservation and management policies from the vital information given on the importance of wetlands in carbon dynamics and ecosystem productivity.

Biological signals such as respiration (RSP) and heart rate (HR) are oscillatory and physiologically coupled, maintaining homeostasis through regulatory mechanisms. This report models the dynamic relationship between RSP and HR in 45 healthy volunteers at rest. Cross-correlation between RSP and HR was computed, along with regression analysis to predict HR from RSP and its first-order time derivative in continuous signals. A simulation model tested the possibility of replicating the RSP-HR relationship. Cross-correlation results showed a time lag in the sub-second range of these signals (849.21 ms ± SD 344.84). The possible modulation of HR by RSP was mediated by the RSP amplitude and its first-order time derivative (in 45 of 45 cases). A simulation of this process allowed us to replicate the physiological relationship between RSP and HR. These results provide support for understanding the dynamic interactions in cardiorespiratory coupling at rest, showing a short time lag between RSP and HR and a modulation of the HR signal by the first-order time derivative of the RSP. This dynamic would optionally be incorporated into dynamic models of resting cardiopulmonary coupling and suggests a mechanism for optimizing respiration in the alveolar system by promoting synchrony between the gases and hemoglobin in the alveolar pulmonary system.

Abstract Assessment of driver stress, crucial for road safety, can greatly benefit from the analysis of multimodal physiological signals. However, fusing such heterogeneous data poses significant challenges, particularly in intermediate fusion where noise can also be fused. In this study, we address this challenge by exploring a 1D convolutional neural network (CNN) with self-attention mechanisms on multimodal data. Electrocardiogram (ECG) signals (256 Hz) and respiration (RESP) signals (128 Hz) were obtained from ten subjects using textile electrodes while driving in different scenarios, namely normal driving and phone usage (calling). The obtained multimodal data is preprocessed and then applied to a self-attention mechanism (SAM) CNN (SAMcNN) to identify driver stress. Experiments are validated using Leave-one-outsubject cross validation. The proposed approach is capable of classifying driver stress. It is observed that shorter segments yield an accuracy of 64.16% compared to longer segment lengths. Thus, exploring self-attention mechanisms for multimodal signals using wearable shirts facilitates non-intrusive monitoring in real-world driving scenarios.

Abstract To further evaluate the effect of water stress on soil respiration (RS), reveal the influencing factors of daily and seasonal RS, and systematically evaluate and compare the sensibility of different machine learning algorithms (multiple nonlinear regression [MNR], support vector machine regression [SVR], backpropagation artificial neural network [BPNN]) to estimate RS from a maize field under water stress condition, the field experiments were conducted within a maize field in Inner Mongolia, China, during the entire 2019 growing season. Various levels of deficit irrigation were conducted in the vegetative, reproductive, and mature stages. Our research indicated that soil CO2 fluxes from 100% evapotranspiration treatment (Tr1) were significantly greater than various deficit irrigation treatments (Tr2, Tr3, Tr4) during each growth stage of summer maize. The cumulative soil CO2 fluxes of Tr2, Tr3, and Tr4 decreased 24.8%, 30.3%, and 43.7% compared with Tr1, respectively. We determined that the drivers affecting the daily RS were soil temperature at 5 cm depth (TS,5) and soil surface temperature (TSF), followed by water‐filled porosity (WFPS) at 5 cm depth, but no significant correlations were observed at 25 cm depths. TS,5 and TSF also performed similar correlation with seasonal RS with R greater than 0.753 among all water treatments, followed by chlorophyll content with R greater than 0.726. During the whole growing season, the BPNN model exhibited the best predicting result, and could explain the 60%–80% and 87.8% of the variations of RS at the daily and seasonal scales, with root mean square error of 48.7–100.9 mg m−2 h−1 and 91.5 mg m−2 h−1, respectively. The SVR and MNR models could estimate the 47.9%–57% and 39.9%–52.1% of the daily RS and 81.4% and 78.6% of the seasonal RS, respectively. Overall, our study indicated the machine learning algorithms could be successfully applied to estimate RS at daily and seasonal scales from a maize field under water stress condition.

ABSTRACTHigh‐temperature limits early season vegetative growth of cotton, and the physiological response of cotton (Gossypium hirsutum L.) to high daytime or nighttime temperature needs to be explored. The objectives of the current study were to determine (1) plant growth response, (2) physiological contributors to variation in biomass production and (3) mechanisms driving variation in net photosynthetic rate (AN) in response to different combinations of high daytime and nighttime temperatures. Beginning at planting, cotton was exposed to four different growth temperature regimes: (1) optimum (30/20°C day/night), (2) high nighttime (30/30°C), (3) high daytime (40/20°C) and combined high daytime and nighttime (40/30°C) for 4 weeks. Relative to the 30/20°C treatment, plant growth was positively affected by high nighttime temperature and negatively affected by high daytime temperature and combined high day and night temperature. Increased leaf area mainly contributed to increased biomass production in high nighttime temperature; higher nighttime respiration (RN) drove reductions in biomass in combined high daytime and nighttime temperature; and decreased leaf area and AN and increased RN drove reductions in biomass under high daytime temperature alone. AN was not impacted by high nighttime temperature, while decreased under high daytime temperature and increased with combined high daytime and nighttime temperature. Adjustments in leaf traits contributed to increases in AN in combined high daytime and nighttime temperature, and increased photorespiration and respiration contributed to reductions in AN under high daytime temperature. Overall, early season vegetative growth of cotton exhibited differential responses to high daytime and nighttime temperatures.

Assessing the olfactory preferences of consumers is an important aspect of fragrance product development and marketing. With the advancement of wearable device technologies, physiological signals hold great potential for evaluating olfactory preferences. However, there is currently a lack of relevant studies and specific explanatory procedures for preference assessment methods that are based on physiological signals. In response to this gap, a synchronous data acquisition system was established using the ErgoLAB multi-channel physiology instrument and olfactory experience tester. Thirty-three participants were recruited for the olfactory preference experiments, and three types of autonomic response data (skin conductance, respiration, and heart rate) were collected. The results of both individual and overall analyses indicated that olfactory preferences can lead to changes in skin conductance (SC), respiration (RESP), and heart rate (HR). The trends of change in both RESP and HR showed significant differences (with the HR being more easily distinguishable), while the SC did not exhibit significant differences across different olfactory perception preferences. Additionally, gender differences did not result in significant variations. Therefore, HR is more suitable for evaluating olfactory perception preferences, followed by RESP, while SC shows the least effect. Moreover, a logistic regression model with a high accuracy (84.1%) in predicting olfactory perception preferences was developed using the changes in the RESP and HR features. This study has significant implications for advancing the assessment of consumer olfactory preferences.

A hybrid transposed attention based deep learning model for wearable and explainable stress recognition