scholarly journals The Metabolic Regimes at the Scale of an Entire Stream Network Unveiled Through Sensor Data and Machine Learning

Ecosystems ◽  
2021 ◽  
Author(s):  
Pier Luigi Segatto ◽  
Tom J. Battin ◽  
Enrico Bertuzzo
Keyword(s):  
Machine Learning ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Light Availability ◽  
Sensor Data ◽  
Stream Metabolism ◽  
River Continuum ◽  
Stream Network ◽  
Network Scale ◽  
Clear Peak

AbstractStreams and rivers form dense networks that drain the terrestrial landscape and are relevant for biodiversity dynamics, ecosystem functioning, and transport and transformation of carbon. Yet, resolving in both space and time gross primary production (GPP), ecosystem respiration (ER) and net ecosystem production (NEP) at the scale of entire stream networks has been elusive so far. Here, combining Random Forest (RF) with time series of sensor data in 12 reach sites, we predicted annual regimes of GPP, ER, and NEP in 292 individual stream reaches and disclosed properties emerging from the network they form. We further predicted available light and thermal regimes for the entire network and expanded the library of stream metabolism predictors. We found that the annual network-scale metabolism was heterotrophic yet with a clear peak of autotrophy in spring. In agreement with the River Continuum Concept, small headwaters and larger downstream reaches contributed 16% and 60%, respectively, to the annual network-scale GPP. Our results suggest that ER rather than GPP drives the metabolic stability at the network scale, which is likely attributable to the buffering function of the streambed for ER, while GPP is more susceptible to flow-induced disturbance and fluctuations in light availability. Furthermore, we found large terrestrial subsidies fueling ER, pointing to an unexpectedly high network-scale level of heterotrophy, otherwise masked by simply considering reach-scale NEP estimations. Our machine learning approach sheds new light on the spatiotemporal dynamics of ecosystem metabolism at the network scale, which is a prerequisite to integrate aquatic and terrestrial carbon cycling at relevant scales.

Data-based machine learning unveils ecosystem metabolic regimes at the scale of entire stream networks

2020 ◽  
Author(s):  
Pier Luigi Segatto ◽  
Tom J. Battin ◽  
Enrico Bertuzzo
Keyword(s):  
Machine Learning ◽  
Carbon Cycle ◽  
Ecosystem Respiration ◽  
Spatial Scales ◽  
Gross Primary Production ◽  
Data Availability ◽  
Ecosystem Metabolism ◽  
Sensor Technology ◽  
Stream Network ◽  
Metabolic Fluxes

<p>Inland waters are major contributors to the global carbon cycle. Nowadays, new sensor technology has changed the way we study ecosystem metabolism in streams. We are able to produce long-term time series of gross primary production (GPP) and ecosystem respiration (ER) to infer drivers of the stream ecosystem metabolic regime and its seasonal timing. Despite big data availability, most studies are limited to individual stream reaches and do not allow the appreciation of metabolic regimes at the scale of entire networks, which, however, would be fundamental to properly assess the relevance of metabolic fluxes within streams and rivers for carbon cycling at the regional and global scale. Machine learning (ML) has great potential in this direction. Firstly, ML could be used to extrapolate both in time and space heterogeneous forcings (e.g., streamwater temperature (T) and photosynthetic active radiation (PAR)) required to run a process-based model for reach-scale metabolism to the scale of an entire stream network. Secondly, the same procedure could be applied to reach-scale estimates of ecosystem metabolism to check whether available data contain enough information to explain the network scale variability. In this study, we used Random Forest to predict patterns of environmental forcings (T and PAR) and stream metabolism (GPP and ER) at the scale of an entire stream network. We used available high-frequency measurements of T and PAR, estimates of ecosystem metabolism and major proximal controls (e.g., incident light, discharge, stream-bed slope, drainage area, water level,  air temperature) from twelve reaches within the Ybbs River network (Austria) and explicitly trained our Random Forests by integrating distal factors, namely:  vegetation type, canopy cover, hydro-geomorphic properties, light,  precipitation, and other climatic variables. We designed two different training setups to assess spatial and temporal predicting model capabilities, respectively. This approach allowed us to reliably infer the target variables (T, PAR, GPP, and ER) on annual basis across a stream network, to filter the most important predictors, to assess the relative contribution of the metabolic fluxes from small to large streams, to estimate annual metabolic budgets at different spatial scales and to provide empirical evidence for long-standing theory predicting shifts of ecosystem metabolism along the stream continuum. Finally, we estimated autochthonous and allochthonous respiration for the entire stream network, which is crucial to integrate the role of ecosystem processes for the carbon cycle.</p>

scholarly journals Upscaling Northern Peatland CO2 Fluxes Using Satellite Remote Sensing Data

2021 ◽  
Vol 13 (4) ◽  
pp. 818
Author(s):  
Sofia Junttila ◽  
Julia Kelly ◽  
Natascha Kljun ◽  
Mika Aurela ◽  
Leif Klemedtsson ◽  
...  
Keyword(s):  
Remote Sensing ◽  
Land Surface ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Vegetation Indices ◽  
Sensor Data ◽  
Spatial And Temporal Variation ◽  
Model Parameters ◽  
Co2 Fluxes ◽  
Sentinel 2

Peatlands play an important role in the global carbon cycle as they contain a large soil carbon stock. However, current climate change could potentially shift peatlands from being carbon sinks to carbon sources. Remote sensing methods provide an opportunity to monitor carbon dioxide (CO2) exchange in peatland ecosystems at large scales under these changing conditions. In this study, we developed empirical models of the CO2 balance (net ecosystem exchange, NEE), gross primary production (GPP), and ecosystem respiration (ER) that could be used for upscaling CO2 fluxes with remotely sensed data. Two to three years of eddy covariance (EC) data from five peatlands in Sweden and Finland were compared to modelled NEE, GPP and ER based on vegetation indices from 10 m resolution Sentinel-2 MSI and land surface temperature from 1 km resolution MODIS data. To ensure a precise match between the EC data and the Sentinel-2 observations, a footprint model was applied to derive footprint-weighted daily means of the vegetation indices. Average model parameters for all sites were acquired with a leave-one-out-cross-validation procedure. Both the GPP and the ER models gave high agreement with the EC-derived fluxes (R2 = 0.70 and 0.56, NRMSE = 14% and 15%, respectively). The performance of the NEE model was weaker (average R2 = 0.36 and NRMSE = 13%). Our findings demonstrate that using optical and thermal satellite sensor data is a feasible method for upscaling the GPP and ER of northern boreal peatlands, although further studies are needed to investigate the sources of the unexplained spatial and temporal variation of the CO2 fluxes.

scholarly journals Resistance and resilience of stream metabolism to high flow disturbances

2020 ◽  
Author(s):  
Brynn O'Donnell ◽  
Erin R. Hotchkiss
Keyword(s):  
Dynamic Equilibrium ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Ecosystem Processes ◽  
High Flow ◽  
Running Waters ◽  
Stream Metabolism ◽  
Metabolic Resistance ◽  
Disturbance Intensity ◽  
Flow Disturbances

Abstract. Streams are ecosystems organized by disturbance. One of the most frequent and variable disturbances in running waters is elevated flow. Yet, we still have few estimates of how ecosystem processes, such as stream metabolism (gross primary production and ecosystem respiration; GPP and ER), respond to high flow events. Furthermore, we lack a predictive frame- work for understanding controls on within-site metabolic responses to flow disturbances. Using five years of high-frequency dissolved oxygen data from an urban- and agriculturally-influenced stream, we estimated daily GPP and ER and analyzed metabolic changes across 15 isolated high flow events. Metabolism was variable from day to day, even during lower flows. Thus, we calculated metabolic resistance as the magnitude of departure from the dynamic equilibrium during antecedent lower flows and quantified resilience from the days until GPP and ER returned to the range of antecedent dynamic equilibrium. We evaluated correlations between metabolic resistance and resilience with characteristics of each high flow event, antecedent conditions, and time since last flow disturbance. ER was more resistant and resilient than GPP. GPP was typically suppressed following flow disturbances, regardless of disturbance intensity. In contrast, the ER magnitude of departure increased with disturbance intensity. Additionally, GPP was less resilient and took longer to recover (0 to > 9 days, mean = 2.2) than ER (0 to 2 days, mean = 0.6). Given the flashy nature of streams draining human-altered landscapes and the variable consequences of flow for GPP and ER, testing how ecosystem processes respond to flow disturbances is essential to an integrative understanding of ecosystem function.

scholarly journals Moderate increases in channel discharge are positively related to ecosystem respiration in forested Ozark streams

2021 ◽  
Author(s):  
Allyn K. Dodd ◽  
Daniel D. Magoulick ◽  
Michelle A. Evans-White
Keyword(s):  
Primary Production ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Carbon Sources ◽  
Flow Regimes ◽  
Water Source ◽  
Stream Metabolism ◽  
Global Trends ◽  
Ecosystem Level ◽  
And Function

ABSTRACTThe natural flow regime is considered the “master variable” in lotic systems, controlling structure and function at organismal, population, community, and ecosystem levels. We sought to estimate forested headwater stream metabolism across two dominant flow regimes (Runoff and Groundwater) in northern Arkansas and evaluate potential differences in, and drivers of, gross primary production, ecosystem respiration, and net ecosystem metabolism. Flow regimes differed in intermittency, substrate heterogeneity, hyporheic connectivity, and dominant water source (subsurface runoff vs. groundwater), which we expected to result in differences in primary production and respiration. Average daily gross primary production (GPP) and ecosystem respiration (ER) estimated from field data collected from May 2015-June 2016 tended to be greater in Groundwater streams. Respiration was positively related to discharge (R2= 0.98 p< 0.0001) and net metabolism became more heterotrophic with increasing average annual discharge across sites (R2= 0.94, p= 0.0008). Characterizing ecosystem-level responses to differences in flow can reveal mechanisms governing stream metabolism and, in turn, provide information regarding trophic state and energy inputs as efforts continue to determine global trends in aquatic carbon sources and fates.

scholarly journals Hydrologic Setting Dictates the Sensitivity of Ecosystem Metabolism to Climate Variability in Lakes

Ecosystems ◽  
2021 ◽  
Author(s):  
Isabella A. Oleksy ◽  
Stuart E. Jones ◽  
Christopher T. Solomon
Keyword(s):  
Oxygen Sensor ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Climatic Variability ◽  
Ecosystem Metabolism ◽  
Sensor Data ◽  
Lake Area ◽  
Coefficients Of Variation ◽  
Bog Lakes ◽  
Carbon Concentrations

AbstractGlobal change is influencing production and respiration in ecosystems across the globe. Lakes in particular are changing in response to climatic variability and cultural eutrophication, resulting in changes in ecosystem metabolism. Although the primary drivers of production and respiration such as the availability of nutrients, light, and carbon are well known, heterogeneity in hydrologic setting (for example, hydrological connectivity, morphometry, and residence) across and within regions may lead to highly variable responses to the same drivers of change, complicating our efforts to predict these responses. We explored how differences in hydrologic setting among lakes influenced spatial and inter annual variability in ecosystem metabolism, using high-frequency oxygen sensor data from 11 lakes over 8 years. Trends in mean metabolic rates of lakes generally followed gradients of nutrient and carbon concentrations, which were lowest in seepage lakes, followed by drainage lakes, and higher in bog lakes. We found that while ecosystem respiration (ER) was consistently higher in wet years in all hydrologic settings, gross primary production (GPP) only increased in tandem in drainage lakes. However, interannual rates of ER and GPP were relatively stable in drainage lakes, in contrast to seepage and bog lakes which had coefficients of variation in metabolism between 22–32%. We explored how the geospatial context of lakes, including hydrologic residence time, watershed area to lake area, and landscape position influenced the sensitivity of individual lake responses to climatic variation. We propose a conceptual framework to help steer future investigations of how hydrologic setting mediates the response of metabolism to climatic variability.

scholarly journals Variation of stream metabolism along a tropical environmental gradient

2018 ◽  
Author(s):  
Wesley A. Saltarelli ◽  
Walter K. Dodds ◽  
Flavia Tromboni ◽  
Maria do Carmo Calijuri ◽  
Vinicius Neres-Lima ◽  
...  
Keyword(s):  
Land Use ◽  
Total Phosphorus ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Canopy Cover ◽  
Tropical Streams ◽  
Nutrient Concentrations ◽  
Stream Metabolism ◽  
Mixed Effect ◽  
Single Station

Stream metabolism is affected by both natural and human-induced processes. While metabolism has multiple implications for ecological processes, relatively little is known about how metabolic rates are influenced by land use in tropical streams. In this study, we assessed the metabolic characteristics and related environmental factors of six streams located in a transition area from Cerrado to Atlantic Forest (São Carlos/Brazil). Three streams were relatively preserved, while three were flowing through more agriculturally and/or urban impacted watersheds. Surface water samples were analyzed for biological and physico-chemical parameters as well as discharge and percentage of canopy cover. Metabolism was determined through the single-station method to estimate gross primary production (GPP), ecosystem respiration (ER) and net ecosystem production (NEP) with BAyesian Single-station Estimation (BASE). Nutrient concentrations tended to be higher in impacted versus preserved streams (e.g., average total phosphorus between 0.028-0.042 mg L-1 and 0.009-0.038 mg L-1, respectively). Average canopy cover varied between 58 and 77%, with no significant spatial or seasonal variation. All streams were net heterotrophic (ER exceeded GPP) in all sampling periods. GPP rates were always lower than 0.7 gO2 m-2 d-1 in all streams and ER varied from 0.6 to 42.1 gO2 m-2 d-1.  Linear Mixed-Effect models showed that depth, discharge, velocity and total phosphorus are the most important predictors for GPP. For ER, depth, velocity and canopy cover are significant potential predictors. Canopy cover was the main light limiting factor and influenced stream metabolism. Our findings reinforced the concepts that shifts in the shading effect provided by vegetation (e.g., through deforestation) or changes in discharge (e.g., through land use conversion or water abstractions) can impact freshwater metabolism. Our study suggests that human activities in low latitude areas can alter tropical streams’ water quality, ecosystem function, and the degree of riparian influence. Our data showed that tropical streams can be especially responsive to increases of organic matter inputs leading to high respiration rates and net heterotrophy, and this should be considered to support management and restoration efforts.

Longitudinal Structure of an Agricultural Prairie River System and its Relationship to Current Stream Ecosystem Theory

1990 ◽  
Vol 47 (2) ◽  
pp. 373-384 ◽  
Author(s):  
M. J. Wiley ◽  
L. L. Osborne ◽  
R. W. Larimore
Keyword(s):  
Primary Productivity ◽  
Gross Primary Production ◽  
Light Availability ◽  
River System ◽  
Community Respiration ◽  
River Continuum ◽  
Stream Ecosystem ◽  
Empirical Modelling ◽  
Ecosystem Theory ◽  
Central Illinois

The largescale structure of an agriculturally developed prairie river system in central Illinois was examined and compared with predictions from current stream ecosystem theory. High rates of primary productivity (> 15 g carbon∙m−2∙d−1) were characteristic of the watershed, although longitudinal patterns in riparian vegetation, stream temperature, and primary productivity were inverted relative to typical streams in forested uplands. Empirical models of gross primary production and community respiration were developed. Light availability, mediated by both channel shading and turbidity, appeared to be the principal factor limiting primary productivity. Both nitrate and orthophosphorus were found in high concentrations throughout the watershed. Largescale patterns in nutrient availability suggest that landuse patterns, and particularly urbanization, strongly affected spatial and temporal distributions of both nutrients. Differences between prairie river systems and "prototype" structures envisioned by the River Continuum Concept (RCC) derive from the descriptive nature of the RCC, and its inability to incorporate nonstandard distributions of key driving variables. The use of empirical modelling in stream ecosystem studies is discussed.

scholarly journals New Gap-Filling Strategies for Long-Period Flux Data Gaps Using a Data-Driven Approach

Atmosphere ◽  
2019 ◽  
Vol 10 (10) ◽  
pp. 568 ◽  
Author(s):  
Minseok Kang ◽  
Kazuhito Ichii ◽  
Joon Kim ◽  
Yohana M. Indrawati ◽  
Juhan Park ◽  
...  
Keyword(s):  
Machine Learning ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Data Driven ◽  
Series Data ◽  
Gap Filling ◽  
Long Period ◽  
Flux Data ◽  
Data Gaps ◽  
Data Driven Approach

In the Korea Flux Monitoring Network, Haenam Farmland has the longest record of carbon/water/energy flux measurements produced using the eddy covariance (EC) technique. Unfortunately, there are long gaps (i.e., gaps longer than 30 days), particularly in 2007 and 2014, which hinder attempts to analyze these decade-long time-series data. The open source and standardized gap-filling methods are impractical for such long gaps. The data-driven approach using machine learning and remote-sensing or reanalysis data (i.e., interpolating/extrapolating EC measurements via available networks temporally/spatially) for estimating terrestrial CO2/H2O fluxes at the regional/global scale is applicable after appropriate modifications. In this study, we evaluated the applicability of the data-driven approach for filling long gaps in flux data (i.e., gross primary production, ecosystem respiration, net ecosystem exchange, and evapotranspiration). We found that using a longer training dataset in the machine learning generally produced better model performance, although there was a greater possibility of missing interannual variations caused by ecosystem state changes (e.g., changes in crop variety). Based on the results, we proposed gap-filling strategies for long-period flux data gaps and used them to quantify the annual sums with uncertainties in 2007 and 2014. The results from this study have broad implications for long-period gap-filling at other sites, and for the estimation of regional/global CO2/H2O fluxes using a data-driven approach.

scholarly journals Machine learning estimates of eddy covariance carbon flux in a scrub in the Mexican highland

2021 ◽  
Vol 18 (2) ◽  
pp. 367-392
Author(s):  
Aurelio Guevara-Escobar ◽  
Enrique González-Sosa ◽  
Mónica Cervantes-Jiménez ◽  
Humberto Suzán-Azpiri ◽  
Mónica Elisa Queijeiro-Bolaños ◽  
...  
Keyword(s):  
Machine Learning ◽  
Eddy Covariance ◽  
Carbon Sink ◽  
Ecosystem Respiration ◽  
Gross Primary Production ◽  
Model Performance ◽  
Net Ecosystem Exchange ◽  
Carbon Dynamics ◽  
Ordinary Least Squares ◽  
Machine Learning Algorithms

Abstract. Arid and semiarid ecosystems contain relatively high species diversity and are subject to intense use, in particular extensive cattle grazing, which has favored the expansion and encroachment of perennial thorny shrubs into the grasslands, thus decreasing the value of the rangeland. However, these environments have been shown to positively impact global carbon dynamics. Machine learning and remote sensing have enhanced our knowledge about carbon dynamics, but they need to be further developed and adapted to particular analysis. We measured the net ecosystem exchange (NEE) of C with the eddy covariance (EC) method and estimated gross primary production (GPP) in a thorny scrub at Bernal in Mexico. We tested the agreement between EC estimates and remotely sensed GPP estimates from the Moderate Resolution Imaging Spectroradiometer (MODIS), and also with two alternative modeling methods: ordinary-least-squares (OLS) regression and ensembles of machine learning algorithms (EMLs). The variables used as predictors were MODIS spectral bands, vegetation indices and products, and gridded environmental variables. The Bernal site was a carbon sink even though it was overgrazed, the average NEE during 15 months of 2017 and 2018 was −0.78 gCm-2d-1, and the flux was negative or neutral during the measured months. The probability of agreement (θs) represented the agreement between observed and estimated values of GPP across the range of measurement. According to the mean value of θs, agreement was higher for the EML (0.6) followed by OLS (0.5) and then MODIS (0.24). This graphic metric was more informative than r2 (0.98, 0.67, 0.58, respectively) to evaluate the model performance. This was particularly true for MODIS because the maximum θs of 4.3 was for measurements of 0.8 gCm-2d-1 and then decreased steadily below 1 θs for measurements above 6.5 gCm-2d-1 for this scrub vegetation. In the case of EML and OLS, the θs was stable across the range of measurement. We used an EML for the Ameriflux site US-SRM, which is similar in vegetation and climate, to predict GPP at Bernal, but θs was low (0.16), indicating the local specificity of this model. Although cacti were an important component of the vegetation, the nighttime flux was characterized by positive NEE, suggesting that the photosynthetic dark-cycle flux of cacti was lower than ecosystem respiration. The discrepancy between MODIS and EC GPP estimates stresses the need to understand the limitations of both methods.

Sign in / Sign up

Export Citation Format

Share Document