Search for: All records

Climate change alters how strongly the atmosphere draws water from the land, yet a consistent global assessment of this evaporative demand has been lacking. Here, we analyze 45 years of climate data and global models to quantify trends in the key drivers—air temperature, humidity, radiation, wind speed, and cloud cover—that determine the atmosphere’s drying power. We find that evaporative demand has increased worldwide, indicating a stronger atmospheric thirst, except in South Asia, where it has declined. There, widespread irrigation has increased soil and air moisture, enhanced cloud formation, and reduced sunlight reaching the surface, counteracting the global signal. These contrasting trends reveal how human water use can locally reshape the climate’s influence on the water cycle. 

Abstract: The magnitude of the terrestrial carbon sink remains a key uncertainty in future climate projections, in part due to poorly understood links between carbon uptake and its allocation to woody biomass in vegetation. Here, we show that photosynthesis and above-ground growth occur asynchronously across diel to seasonal scales in eight North American oak species. Across 137 tree-ring sites, current-year annual growth was insensitive to climate variability after mid-summer, despite 26-36% of annual gross primary productivity (GPP) occurring during this period. Hourly GPP flux and growth measurements at four sites spanning seven site-years further demonstrate that wood formation ceases earlier than photosynthesis and is restricted to periods of low atmospheric aridity and temperature. This photosynthesis-growth decoupling intensifies with inter-annual variability in vapour pressure deficit (r=0.86, p<0.05), suggesting that by assuming tight coupling between photosynthesis and woody biomass, current earth system models may overestimate long-term carbon sequestration in forests. TechnicalInfo: # Data and code from: Decoupled carbon assimilation and growth responses to aridity in temperate deciduous oaks DOI: 10.5061/dryad.m63xsj4h4 Last date updated, May 22 2026, by Rao MP. The dataset (r_markdown.zip) contains tree growth, carbon flux, and environmental data for different oak sites across North America This includes 4 high-resolution monitoring sites in the USA, which include (See Supplementary Figure 1) 1. Morton Arboretum, IL, USA 41.82°N, 88.07°W * Point dendrometers * Phenocams * SIF derived GPP data (computed from Turner et al 2021) * Environmental data 2. Lamont Sanctuary, NY, USA 41°N, 73.90°W * Point dendrometers * Phenocam - [https://phenocam.nau.edu/webcam/sites/ldeocam/](https://phenocam.nau.edu/webcam/sites/ldeocam/) * SIF derived GPP data (computed from Turner et al 2021) * Xylogenesis/Wood Anatomy data * Leaf level chlorophyll fluorescence data using a Jr. Pam instrument * Environmental data 3. Pace Forest, VA, USA 78°N, 33°W * Point dendrometers * Phenocam * In-situ eddy covariance and environmental data 4. Tonzi Ranch, CA, USA 38.43°N, 120.96°W * Point dendrometers * Phenocam - [https://phenocam.nau.edu/webcam/sites/tonzi/](https://phenocam.nau.edu/webcam/sites/tonzi/) * In-situ eddy covariance and environmental data - [https://ameriflux.lbl.gov/sites/siteinfo/US-Ton](https://ameriflux.lbl.gov/sites/siteinfo/US-Ton) Also included are raw (cross-dated) tree ring measurements from 137 oak sites across North America (134 - USA & 3 - Canada) ## Sharing/Access information You may contact the lead author Mukund Palat Rao [mukund@ldeo.columbia.edu](mailto:mukund@ldeo.columbia.edu) or [mukund24rao@gmail.com](mailto:mukund24rao@gmail.com) for assistance in interpreting and using the data. Please also cite the original publication if these data or codes are used. Rao MP et al (2026), Decoupled carbon assimilation and growth responses to aridity in temperate deciduous oaks, Science Advances, DOI: 10.1126/sciadv.ady7139 Links to other publicly accessible locations of the data: The data are available directly through the manuscript ## Code/Software Files in this folder can be read using R (.Rmd), and included code can be used to read these data and replicated all analysis included in the papers. Files 1-4 read in dendrometer, PhenoCam, flux, or remote sensing, and environmental data for the 4 high-resolution monitoring sites to compare the phenology and environmental sensitivity of growth vs gross primary productivity (GPP) 1. IL_Morton_Dendro-GPP-Pheno_20260126.Rmd 2. CA_Tonzi_Dendro-GPP-Pheno_20260208.Rmd 3. VA_Pace_Dendro-GPP-Pheno_20260218.Rmd 4. NY_Lamont_Dendro-GPP-Pheno_20260222.Rmd Files 5-6 are specific to Lamont-NY for leaf-level chlorophyll fluorescence data (using a Jr. PAM) and wood anatomy data 5. jrPAM_lamont_20240522.Rmd 6. NY_woodanatomy_20240401.Rmd File 7. Evaluates environmental data at all 4 sites and evaluates the environmental driver of the 'offset' variable 7. env_cv_vs_growth_gpp_20260223.Rmd File 8. Evaluates the environmental drivers of growth at 137 tree-ring sites 8. dendro_sensitivity_20240506.Rmd * climcorr_20260221.R is an embedded dependent function for the climate-growth response analysis File 9. Evaluates the phenology of Gross Primary Productivity (GPP) across all 137 tree-ring sites based on the Turner et al (2021) data set using solar-induced chlorophyll fluorescence (SIF). [https://doi.org/10.3334/ORNLDAAC/1875](https://doi.org/10.3334/ORNLDAAC/1875) Note that the code directly reads in downloaded data just for the 137 sites; however, commented code is included that would read in data from the original NETCDF files In this case, the user needs to download the Turner et al. data by themselves first. 9. turner_gpp_sites_20241104.R These files have been tested on R version 4.4.2. In each file, the working directory will need to be changed to your working directory in the first section. All required packages and dependencies will be auto-installed (if unavailable locally) # Knit HTML versions of each file are also included These files are HTML versions that show code and embedded figures in a web browser without needed to run the code 1. IL_Morton_Dendro-GPP-Pheno_20260126.html (from IL_Morton_Dendro-GPP-Pheno_20260126.Rmd) 2. NY_Lamont_Dendro-GPP-Pheno_20260222.html (from NY_Lamont_Dendro-GPP-Pheno_20260222.Rmd) 3. VA_Pace_Dendro-GPP-Pheno_20260218.html (from VA_Pace_Dendro-GPP-Pheno_20260218.Rmd) 4. CA_Tonzi_Dendro-GPP-Pheno_20260208.html (from CA_Tonzi_Dendro-GPP-Pheno_20260208.Rmd) 5. jrPAM_lamont_20240522.html (from jrPAM_lamont_20240522.Rmd) 6. NY_woodanatomy_20240401.html (from NY_woodanatomy_20240401.Rmd) ## Data This includes files included inside the data folder that are read into R 1. CRU * CRU Ts v 4.06 downloaded on Jan 18, 2023 * These data are used to compute the climate sensitivity of tree ring data in dendro_sensitivity_20240506.Rmd * Instructions to download data are provided inside the readme.text within the Data/CRU/ts.4.06/ folder. 2. tree-rings * Includes tree ring data in Tucson Format organised by US State (two letter code) and Canada (CAN). More information on each site is provided in Supplementary Table 1 * ITRDB_readme.txt - includes information on codes used to download tree ring data from the International Tree Ring Data Bank (ITRDB). * See Brewer et al 2011 for more information on the Tucson Format [https://bioone.org/journals/tree-ring-research/volume-67/issue-2/2010-12.1/Tricycle--A-Universal-Conversion-Tool-For-Digital-Tree-Ring/10.3959/2010-12.1.full](https://bioone.org/journals/tree-ring-research/volume-67/issue-2/2010-12.1/Tricycle--A-Universal-Conversion-Tool-For-Digital-Tree-Ring/10.3959/2010-12.1.full) 3. US_CA_Tonzi * phenocam - data downloaded from Phenocam Website, see [https://phenocam.nau.edu/webcam/tools/summary_file_format/](https://phenocam.nau.edu/webcam/tools/summary_file_format/) for descriptions of each column * dendrometers - Dendrometer data on site - US_CA_TON_20240125_clean.csv includes the following columns - time (local time in California, UTC-8) - year - tree (codes from TON1 to TON5) - species (4-letter species code, QUDO - Quercus douglasii), - val (change in tree radius in micrometers based on point dendrometer) - gr (tree radial growth in micrometers based on the zero-growth model) - battery (dendrometer battery voltage) * flux - data downloaded from the Ameriflux website, see README_AmeriFlux_BASE.txt for information on headers and data interpretation Information on column names is also available here [https://fluxnet.org/data/aboutdata/data-variables/](https://fluxnet.org/data/aboutdata/data-variables/) Note that we cannot provide these data directly, as they use to CC-BY-4.0 (By ATTRIBUTION). Instructions on how to download Ameriflux data are provided in data/US_CA_Tonzi/flux/20251124/AMF_US-Ton_BASE-BADM_25-5 in the readme.txt. Once data are downloaded, we use the following columns File name AMF_US-Ton_BASE_HH_25-5.csv Columns that are used - TIMESTAMP_START - start time of measurement - TIMESTAMP_END - end time of measurement - GPP_PI_F_1_1_1 - Gross Primary Productivity based on nightime method for the tall tower μmol/(m^2.s) - GPP_PI_F_1_2_1 - Gross Primary Productivity based on nightime method for the tall tower μmol/(m^2.s) - TA_PI_F_1_1_1 - Photosynthetic Photon Flux Density mol photons/(m2.s) - SWC_PI_F_1_2_1 SWC_PI_F_2_2_1, SWC_PI_F_3_2_1 SWC_PI_F_4_2_1, WC_PI_F_5_1_1, SWC_PI_F_6_3_1 SWC_PI_F_7_3_1 - Soil Water Content (m3 water/m3 soil), the numbers indicate horizontal, vertical, and repetitions H_V_R (Horizontal_Vertical_Replicate), ere each measurement is in a different horizontal location 1-7. 1. US_IL_Morton * sif - data on solar induced chlorophyll fluorescence (SIF) and SIF-Derived GPP around Morton Arboretum, as calculated by turner_gpp_sites_20241104.R Morton_Arborutum_Turner_SIF_GPP_2019_2021.csv includes the following columns - Year - Month (month from 1-12) - Day (day of month) - SIF (mW m−2 sr−1 nm−1) - GPP (in gC/(m2.day)) * phenocam - Phenological Camera information OakPhenology2019-2022.csv includes the following columns - Plot (4 plots for different species) - Datetime (Local datetime in Ilinois, USA, Central Time Zone UTC-6) - Date (date in yyyy-mm-dd) - Year (calendar year) - Day (Day of Year) - Jday (Julian Day) - Hour (Hour of Day) - Minute - Red (reflectance in red band in an 8-bit colour system from 0-255 for the region of interest - ROI) - Green (reflectance in green band in an 8-bit colour system from 0-255 for the region of interest - ROI) - Blue (reflectance in blue band in an 8-bit colour system from 0-255 for the region of interest - ROI) - gcc (Green Chromatic Coordinate - Green/(Red+Green+Blue) - ROI (Region of Interest, 3 per plot) * environmental - micrometeorological data at Morton Arboretum Please see the README weather_processe.docx within the same folder for description of the column names * dendrometers - Contains dendrometer data organised by Plot and Tree in micrometers in the file US_IL_MORTON_202103-202302_clean.csv Columns are defined as: - ts (Local datetime in Ilinois, USA, Central Time Zone UTC-6) - Series (unique code for 20 different trees monitored using dendrometers with codes for species, plot, and tree number within the plot) - Year (year) - Plot (code for Plot ID, 4 plots "QUAL-E" "QUBI-E" "QUBI-W" "QUPA-W") - Tree (tree number within plot) - val (change in tree radius in micrometers based on point dendrometer) - val_growth (tree radial growth in micrometers based on the zero-growth model) 1. US_NY_Lamont * phenocam - data downloaded from Phenocam Website, see [https://phenocam.nau.edu/webcam/tools/summary_file_format/](https://phenocam.nau.edu/webcam/tools/summary_file_format/) for descriptions of each column * anatomy - wood anatomical data at Lamont Sanctuary Forest, NY in 2021 based on xylogenesis sampling - For Column definitions, please see the embedded ROXAS_column_dictionary.xlsx file - This includes definitions for each column - Columns that are used in analyses are - RCTA - Relative Conductive Area Over Time - DOY - Day of year * dendrometers - dendrometer data on Lamond-Doherty Earth Observatory campus (site name Lamont Sanctuary-NY) File name hourly_dendro_env_20240401.csv with columns - variable ("dendro" for dendrometer, "airT" - for air temperature in degree C,"soilT" - for soil temperature in deg. C, "VPD" - for Vapour Pressure Deficit in kPa "VWC" - for volumetric water content at -10 cm) - tree (tree code for 8 trees with dendrometers, "oak1", "oak2", "oak3", "oak4", "redoak1", "redoak2", "redoak3", "whiteoak") - year (calendar year) - date (calendar date) - time (Local datetime in New York, USA, Central Time Zone UTC-4) - val (change in tree radius in micrometers based on point dendrometer) - battV (battery voltage 4 * 1.5V batteries, not used in analysis) * environmental - micrometeorological data on a campus Five different files with environmental data spanning different periods 1. 06-02000(06-02000)-Configuration 1-1710495780.0390148.csv 2. 06-02000(06-02000)-Configuration 2-1710495780.0390148.csv 3. 06-02000(06-02000)-Configuration 3-1710495780.0390148.csv 4. 06-02000(06-02000)-Configuration 4-1710495780.0390148.csv 5. 06-02000(06-02000)-Configuration 5-1710495780.0390148.csv Column definitions are "TA" = "degree_C Air Temperature", "TS" = "degree_C Soil Temperature", "VPD" = "kPa VPD - Vapour Pressure Deficit", "VP" = "kPa Vapor Pressure", "PA" = "kPa Atmospheric Pressure", "SWC" = "m3/m3 Water Content" * JrPAM - leave-level chlorophyll fluorescence data collected from a Jr. PAM on 3 trees between October-November 2021 Please see the Walz Jr. Pam handbook for a detailed explaination of all columns Available at [https://www.walz.com/products/junior-pam/](https://www.walz.com/products/junior-pam/) Column names are "F" = "1:F", "Fmp" = "1:Fm'", "PAR" = "1:PAR", Photosynthetically Active Radiation, Photosynthetic Photon Flux Density (PPFD), μmol/m2/s "Temp" = "1:Temp", Leaf temperature in degrees C. "YII" = "1:Y (II)", Effective quantum yield. Measures the proportion of light absorbed by PSII that is used for photochemistry in the light. "ETR" = "1:ETR", electron Transport Rate. Estimates the rate of electrons transported through the photosynthetic chain "FoP" = "1:Fo'", "ETRmf"= "1:ETR-F.", "qP" = "1:qP", Photochemical/Non-photochemical Quenching Coefficients. "qN" = "1:qN", Photochemical/Non-photochemical Quenching Coefficients. "qL" = "1:qL", Photochemical/Non-photochemical Quenching Coefficients. "NPQ" = "1:NPQ", Non-Photochemical Quenching. "Y_NO" = "1:Y (NO)", Quantum yield of non-regulated energy dissipation (energy lost as heat and fluorescence without being safely regulated). "Y_NPQ"= "1:Y (NPQ)", Quantum yield of regulated, light-induced energy dissipation (heat loss). "Fo" = "1:Fo", Minimal fluorescence. The baseline fluorescence level emitted when all Photosystem II (PSII) reaction centers are fully "open" (oxidized) after dark-adaptation. "Fm" = "1:Fm", Maximum fluorescence. The peak signal when a saturating flash of light temporarily closes all PSII reaction centers. "FvFm" = "1:Fv/Fm" Maximum quantum yield of Photosystem II * sif - data on solar induced chlorophyll fluorescence (SIF) and SIF-Derived GPP around Morton Arboretum, as calculated by turner_gpp_sites_20241104.R Turner_SIF_GPP_2019_2021.csv includes the following columns - Year - Month (month from 1-12) - Day (day of month) - SIF (mW m−2 sr−1 nm−1) - GPP (in gC/(m2.day)) 1. US_VA_Pace * phenocam - Green Chromatic Coordinate is calculated by Fluospec2 spectrometer on site as GCC = (green/(red+green+blue)) two files, Pace_GCC_2023_nostablefilter.txt & Pace_GCC_2023.txt Columns are - t_10min (local standard tiem in Pace Forest, Virgina, USA) - GCC_10min (green chromatic coordinate) * dendrometers - Dendrometer data on site US_VA_Pace_20251121.csv - time (Local time) - Tree (Tree code from Pace 1 to 5) - Species (which species, Quercus falcata, Pinus virginiana, Quercus alba, Quercus rubra) - Genus (Pine or oak) - val (change in tree radius in micrometers based on point dendrometer) * flux - data downloaded from Ameriflux website, see README_AmeriFlux_BASE.txt for information on headers and data interpretation Information on column names is also available here [https://fluxnet.org/data/aboutdata/data-variables/](https://fluxnet.org/data/aboutdata/data-variables/) File name FLX_US-Pac_FLUXNET_FULLSET_HH_2023_beta-5.csv Columns that are used - TIMESTAMP_START - start time of measurement - TIMESTAMP_END - end time of measurement - GPP_NT_VUT_REF - Gross Primary Productivity based on nightime method, μmol/(m^2.s) - GPP_DT_VUT_REF - Gross Primary Productivity based on daytime method, μmol/(m^2.s) - NEE_VUT_REF - Net Primary Productivity based on daytime method, μmol/(m^2.s) - PPFD_IN - Photosynthetic Photon Flux Density mol photons/(m2.s) 1. USTreeAtlas-main * Shapefiles for major tree species in the United States, see embedded README.md for more information data used for this study include shapefiles for study species - included in the shp folder. We use the following subfolders, querdoug, querrubr, queralba, querpalu, quercocc, querfalc, querprin for Quercus douglasii, Q. rubra, Q. alba, Q. palustris, Q. falcata, Q. coccinea, Q. prinus (now montana) 1. results * These results are generated from the intermediate files * Each line here comes from one of 1. IL_Pace_Dendro-GPP-Pheno_20260126.Rmd 2. CA_Tonzi_Dendro-GPP-Pheno_20260208.Rmd 3. VA_Pace_Dendro-GPP-Pheno_20260218.Rmd 4. NY_Lamont_Dendro-GPP-Pheno_20260222.Rmd Columns include - No - Serial Number; site - which site; year - which year - los.nd - length of season in number of days - off1.nd - offset during phase I in number of days (see table 1) - off2.nd - offset during phase II in number of days (see table 1) - Gr.Fr - Growth fraction in Phase I - Gr.P - Percent length of Phase I relative to active season (Phase I+II+III) - GPP.Fr - GPP fraction in Phase III - GPP.P - Percent length of Phase III relative to active season (Phase I+II+III) - Offset.N - Offset fraction as (Phases I+III)/Phase I + II + III - Remaining columns for micrometeorological variables at each site and eyar - Tmp - temperature in degrees Celcius - Tmp.K - temperature in degrees Kelvin - VPD - Vapour Pressure Deficit in kPa - Pr - Precipitation (mm) - SD - Standard Deviation - CV - Coefficient of Variation - QCV - Quartile CV 

The magnitude of the terrestrial carbon sink remains a key uncertainty in future climate projections, in part due to poorly understood links between carbon uptake and its allocation to woody biomass in vegetation. Here, in this study, we show that photosynthesis and aboveground growth occur asynchronously across diel to seasonal scales in eight North American oak species. Across 137 tree ring sites, current-year annual growth was insensitive to climate variability after midsummer despite 26 to 36% of annual gross primary productivity (GPP) occurring during this period. Hourly GPP flux and growth measurements at four sites spanning seven site years further demonstrate that wood formation ceases earlier than photosynthesis and is restricted to periods of low atmospheric aridity and temperature. This photosynthesis-growth decoupling intensifies with interannual variability in vapor pressure deficit (r= 0.86,P< 0.05), suggesting that by assuming tight coupling between photosynthesis and woody biomass, current earth system models may overestimate long-term carbon sequestration in forests. 

Abstract Terrestrial evapotranspiration is the second‐largest component of the land water cycle, linking the water, energy, and carbon cycles and influencing the productivity and health of ecosystems. The dynamics of ET across a spectrum of spatiotemporal scales and their controls remain an active focus of research across different science disciplines. Here, we provide an overview of the current state of ET science across in situ measurements, partitioning of ET, and remote sensing, and discuss how different approaches complement one another based on their advantages and shortcomings. We aim to facilitate collaboration among a cross‐disciplinary group of ET scientists to overcome the challenges identified in this paper and ultimately advance our integrated understanding of ET. 

Abstract Global warming increases ecosystem respiration (ER), creating a positive carbon-climate feedback. Thermal acclimation, the direct responses of biological communities to reduce the effects of temperature changes on respiration rates, is a critical mechanism that compensates for warming-induced ER increases and dampens this positive feedback. However, the extent and effects of this mechanism across diverse ecosystems remain unclear. By analyzing CO2 flux data from 93 eddy covariance sites worldwide, we observed thermal acclimation at 84 % of the sites. If sustained, thermal acclimation could reduce projected warming-induced nighttime ER increases by at least 25 % across most climate zones by 2041-2060. Strong thermal acclimation is particularly evident in ecosystems at high elevation, with low-carbon-content soils, and within tundra, semi-arid, and warm-summer Mediterranean climates, supporting the hypothesis that extreme environments favor the evolution of greater acclimation potential. Moreover, ecosystems with dense vegetation and high productivity such as humid tropical and subtropical forests generally exhibit strong thermal acclimation, suggesting that regions with substantial CO2 uptake may continue to serve as strong carbon sinks. Conversely, some ecosystems in cold continental climates show signs of enhancing thermal responses, the opposite of thermal acclimation, which could exacerbate carbon losses as climate warms. Our study underscores the widespread yet climate-specific patterns of thermal acclimation in global terrestrial ER, emphasizing the need to incorporate these patterns into Earth System Models for more accurate carbon-climate feedback projections. 

ABSTRACT Ecosystem exchanges of carbon, water, and energy are central to Earth system functioning, yet their sensitivities to environmental variability remain poorly constrained across biomes and climates. Here, we analyzed ≥ 5 years of eddy covariance data from 87 AmeriFlux sites (964 site‐years) spanning six vegetation types and a broad range of climatic conditions to examine the controls and multi‐year trends of gross primary productivity (GPP), evapotranspiration (ET), water‐use efficiency (WUE), and the Bowen ratio. We trained boosted regression tree ensembles with environmental (air temperature, vapor pressure deficit, soil water content, atmospheric CO₂, radiation, wind speed) and temporal (month, year) variables and used interpretable machine learning—SHapley Additive exPlanations (SHAP) and Accumulated Local Effects (ALE)—to quantify driver importance and nonlinear responses. Across biomes and along climatic gradients, environmental controls reorganized predictably: radiation dominated under warm and humid conditions, whereas soil moisture exerted a stronger influence in drier or warmer systems. GPP and ET were shaped by similar dominant drivers, with radiation and temperature generally strongest but soil moisture exerting comparable influence in some biomes. WUE was consistently constrained by vapor pressure deficit, indicating stomatal regulation under rising atmospheric dryness. In contrast, controls on the Bowen ratio diverged more across biomes, indicating heterogeneity in how ecosystems partition energy between latent and sensible heat. Multi‐year trend analysis revealed negative associations of GPP and ET with rising atmospheric demand, and positive associations of the Bowen ratio with drying, indicating reduced evaporative cooling and stronger land–atmosphere coupling. Together, these findings show that while carbon and water fluxes remain tightly coupled across timescales, their balances reorganize predictably along bioclimatic gradients. This framework underscores the value of long‐term flux networks and interpretable machine learning for benchmarking Earth system models and constraining projections of terrestrial carbon–water–energy dynamics under climate change. 

Abstract Methane (CH₄) is a potent greenhouse gas (GHG) with atmospheric concentrations that have nearly tripled since pre‐industrial times. Wetlands account for a large share of global CH₄emissions, yet the magnitude and factors controlling CH₄fluxes in tidal wetlands remain uncertain. We synthesized CH₄flux data from 100 chamber and 9 eddy covariance (EC) sites across tidal marshes in the conterminous United States to assess controlling factors and improve predictions of CH₄emissions. This effort included creating an open‐source database of chamber‐based GHG fluxes (https://doi.org/10.25573/serc.14227085). Annual fluxes across chamber and EC sites averaged 26 ± 53 g CH₄m⁻² year⁻¹, with a median of 3.9 g CH₄m⁻² year⁻¹, and only 25% of sites exceeding 18 g CH₄m⁻² year⁻¹. The highest fluxes were observed at fresh‐oligohaline sites with daily maximum temperature normals (MATmax) above 25.6°C. These were followed by frequently inundated low and mid‐fresh‐oligohaline marshes with MATmax ≤25.6°C, and mesohaline sites with MATmax >19°C. Quantile regressions of paired chamber CH₄flux and porewater biogeochemistry revealed that the 90th percentile of fluxes fell below 5 ± 3 nmol m⁻² s⁻¹at sulfate concentrations >4.7 ± 0.6 mM, porewater salinity >21 ± 2 psu, or surface water salinity >15 ± 3 psu. Across sites, salinity was the dominant predictor of annual CH₄fluxes, while within sites, temperature, gross primary productivity (GPP), and tidal height controlled variability at diel and seasonal scales. At the diel scale, GPP preceded temperature in importance for predicting CH₄flux changes, while the opposite was observed at the seasonal scale. Water levels influenced the timing and pathway of diel CH₄fluxes, with pulsed releases of stored CH₄at low to rising tide. This study provides data and methods to improve tidal marsh CH₄emission estimates, support blue carbon assessments, and refine national and global GHG inventories.