Jon Cranko Page

Hydroclimatic extremes are intensifying, escalating risks to ecosystems, infrastructure, and vulnerable communities. Rapid technological innovation and scientific discovery offer powerful tools to address these challenges, yet a gap remains between hydroclimate knowledge production and its use in decision‐making. Bridging this divide requires democratizing hydroclimate science through expanded participation in research design and implementation, as well as in the distribution of recognition, resources, and decision power. Democratizing science can foster outcomes that are (a) more credible, through transparency and integration of decision‐maker perspectives; (b) more just, through fair distribution of research opportunities, resources, recognition, and decision power; and (c) more resilient, by strengthening the capacity of communities and institutions to adapt, learn, and recover from climate pressures. Drawing on insights from the 2024 Global Energy and Water EXchanges (GEWEX) Early Career Researcher Workshop, which gathered participants from 19 countries across disciplines, we outline pathways toward more democratic hydroclimate science grounded in diverse participation and shared leadership. We highlight three priorities: building trust through co‐production, improving the usability of hydroclimate information, and promoting responsible research and innovation in emerging technologies. Case studies illustrate how early‐ and mid‐career researchers and practitioners are advancing these priorities and modeling practices that connect innovation with equity. Achieving lasting and effective change requires research and funding institutions to recognize and support collaborative leadership, and to value relevance and accountability alongside technical excellence. Strengthening such democratic frameworks can help ensure that hydroclimate science advances more just research practices and supports the resilience of the communities it serves.

Sub-Arctic peatlands are often delicately poised at the carbon source-sink threshold. With peatlands among the most carbon-dense ecosystems on Earth, they are critical players in global climate regulation, with land–atmosphere feedbacks that can disproportionately influence climate change trajectories. However, peatland carbon dynamics, and whether they act as sources or sinks for carbon, are strongly shaped by local conditions underscoring the need for site-specific measurements of turbulent fluxes and meteorology to predict their future role in the carbon cycle. While eddy-covariance is a common and critical in-situ measurement technique, the choice of pre-processing algorithms has the potential to interfere in the clear interpretation of source or sink classification in transitional peatland regimes . Here, we present two years of eddy-covariance observations from a newly established eddy-covariance tower in a fen peatland in northeastern Finland. Our analysis characterises the carbon dynamics at the site and addresses a key methodological challenge that is often overlooked: the uncertainty introduced by the subjective choices inherent in eddy-covariance data processing. By generating multiple datasets using alternative processing algorithms, we quantify the sensitivity of flux estimates at the peatland to these decisions, where processing methods affect conclusions regarding its source-sink status. The results provide motivation for a framework for more robust interpretation of peatland carbon fluxes.

Hydroclimatic extremes are intensifying, escalating risks to ecosystems, infrastructure, and vulnerable communities. Rapid technological innovation and scientific discovery offer powerful tools to address these challenges, yet a gap remains between hydroclimate knowledge production and its use in decision-making. Bridging this divide requires democratizing hydroclimate science through expanded participation in research design and implementation, as well as in the distribution of recognition, resources, and decision power. Democratizing science can foster outcomes that are (1) more credible, through transparency and integration of decision-maker perspectives; (2) more just, through fair distribution of research opportunities, resources, recognition, and decision power; and (3) more resilient, by strengthening the capacity of communities and institutions to adapt, learn, and recover from climate pressures. Drawing on insights from the 2024 Global Energy and Water EXchanges (GEWEX) Early Career Researcher Workshop, which gathered participants from 19 countries and across disciplines, we outline pathways toward more democratic hydroclimate science grounded in diverse participation and shared leadership. We highlight three priorities: building trust through co-production, improving the usability of hydroclimate information, and promoting ethical use of emerging technologies. Case studies illustrate how early- and mid-career researchers and practitioners are advancing these priorities and modeling practices that connect innovation with equity. Achieving lasting and effective change requires research and funding institutions to recognize and support collaborative leadership, and to value relevance and accountability alongside technical excellence. Strengthening such democratic frameworks can help ensure that hydroclimate science advances more just research practices and supports the resilience of the communities it serves.

The exchange of carbon, water, and energy fluxes between the land and the atmosphere plays a vital role in shaping global change and extreme events. Yet our understanding of the theory of this surface-atmosphere exchange, represented via land surface models (LSMs), continues to be limited, highlighted by marked biases in model-data benchmarking exercises. Here, we leveraged the PLUMBER2 dataset of observations and model simulations of terrestrial sensible heat, latent heat, and net ecosystem exchange fluxes from 153 international eddy-covariance sites to identify the meteorological conditions under which land surface models are performing worse than independent benchmark expectations. By defining performance relative to three sophisticated out-of-sample empirical models, we generated a lower bound of performance in turbulent flux prediction that can be achieved with the input information available to the land surface models during testing at flux tower sites. We found that land surface model performance relative to empirical models is worse at edge conditions – that is, LSMs underperform in timesteps where the meteorological conditions consist of coinciding relative extreme values. Conversely, LSMs perform much better under “typical” conditions within the centre of the meteorological variable distributions. Constraining analysis to exclude the edge conditions results in the LSMs outperforming strong empirical benchmarks. Encouragingly, we show that refinement of the performance of land surface models in these edge conditions, consisting of only 12 %–31 % of all site-timesteps, would see large improvements (22 %–114 %) in an aggregated performance metric. Better performance in the edge conditions could see mean relative improvements in the aggregated metric of 77 % for the latent heat flux, 48 % for the sensible heat flux, and 36 % for the net ecosystem exchange on average across all LSMs and sites. Precise targeting of model development towards these meteorological edge conditions offers a fruitful avenue to focus model development, ensuring future improvements have the greatest impact.

The exchange of carbon, water, and energy fluxes between the land and the atmosphere plays a vital role in shaping our understanding of global change and how climate change will affect extreme events. Yet our understanding of the theory of this surface-atmosphere exchange, represented via land surface models, continues to be limited, highlighted by marked biases in model-data benchmarking exercises. Here, we leveraged the PLUMBER2 dataset of observations and model simulations of terrestrial fluxes from 153 international eddy-covariance sites to identify the meteorological conditions under which land surface models are performing worse than a priori expectations. By defining performance relative to three sophisticated out-of-sample empirical models, we generated a lower bound of performance in turbulent flux prediction that can be achieved with the input information available to the land surface models during testing at flux tower sites. We found that land surface model (LSM) performance relative to empirical models is worse at boundary conditions – that is, LSMs underperform in timesteps where the meteorological conditions consist of coinciding relative extreme values. Conversely, LSMs perform much better under "typical" conditions within the centre of the meteorological variable distributions. Constraining analysis to exclude the boundary conditions results in the LSMs outperforming strong empirical benchmarks. Encouragingly, we show that refinement of the performance of land surface models in these boundary conditions, consisting of only 12 % to 31 % of time steps, would see large improvements (22 % to 114 %) in an aggregated performance metric. Precise targeting of model development towards these meteorological boundary conditions offers a fruitful avenue to focus model development, ensuring future improvements have the greatest impact.

The exchange of carbon, water, and energy fluxes between the land and the atmosphere plays a vital role in shaping our understanding of global change and how climate change will affect extreme events. Yet our understanding of the theory of this surface-atmosphere exchange, represented via land surface models, continues to be limited, highlighted by marked biases in model-data benchmarking exercises. Here, we leveraged the PLUMBER2 dataset of observations and model simulations of terrestrial fluxes from 153 international eddy-covariance sites to identify the meteorological conditions under which land surface models are performing worse than a priori expectations. By defining performance relative to three sophisticated out-of-sample empirical models, we generated a lower bound of performance in turbulent flux prediction that can be achieved with the input information available to the land surface models during testing at flux tower sites. We found that land surface model (LSM) performance relative to empirical models is worse at boundary conditions – that is, LSMs underperform in timesteps where the meteorological conditions consist of coinciding relative extreme values. Conversely, LSMs perform much better under "typical" conditions within the centre of the meteorological variable distributions. Constraining analysis to exclude the boundary conditions results in the LSMs outperforming strong empirical benchmarks. Encouragingly, we show that refinement of the performance of land surface models in these boundary conditions, consisting of only 12 % to 31 % of time steps, would see large improvements (22 % to 114 %) in an aggregated performance metric. Precise targeting of model development towards these meteorological boundary conditions offers a fruitful avenue to focus model development, ensuring future improvements have the greatest impact.

The exchange of carbon, water, and energy fluxes between the land and the atmosphere plays a vital role in shaping our understanding of global change and how climate change will affect extreme events. Yet our understanding of the theory of this surface-atmosphere exchange, represented via land surface models, continues to be limited, highlighted by marked biases in model-data benchmarking exercises. Here, we leveraged the PLUMBER2 dataset of observations and model simulations of terrestrial fluxes from 153 international eddy-covariance sites to identify the meteorological conditions under which land surface models are performing worse than a priori expectations. By defining performance relative to three sophisticated out-of-sample empirical models, we generated a lower bound of performance in turbulent flux prediction that can be achieved with the input information available to the land surface models during testing at flux tower sites. We found that land surface model (LSM) performance relative to empirical models is worse at boundary conditions – that is, LSMs underperform in timesteps where the meteorological conditions consist of coinciding relative extreme values. Conversely, LSMs perform much better under "typical" conditions within the centre of the meteorological variable distributions. Constraining analysis to exclude the boundary conditions results in the LSMs outperforming strong empirical benchmarks. Encouragingly, we show that refinement of the performance of land surface models in these boundary conditions, consisting of only 12 % to 31 % of time steps, would see large improvements (22 % to 114 %) in an aggregated performance metric. Precise targeting of model development towards these meteorological boundary conditions offers a fruitful avenue to focus model development, ensuring future improvements have the greatest impact.

The exchange of carbon, water, and energy fluxes between the land and the atmosphere plays a vital role in shaping our understanding of global change and how climate change will affect extreme events. Yet our understanding of the theory of this surface-atmosphere exchange, represented via land surface models, continues to be limited, highlighted by marked biases in model-data benchmarking exercises. Here, we leveraged the PLUMBER2 dataset of observations and model simulations of terrestrial fluxes from 153 international eddy-covariance sites to identify the meteorological conditions under which land surface models are performing worse than a priori expectations. By defining performance relative to three sophisticated out-of-sample empirical models, we generated a lower bound of performance in turbulent flux prediction that can be achieved with the input information available to the land surface models during testing at flux tower sites. We found that land surface model (LSM) performance relative to empirical models is worse at boundary conditions – that is, LSMs underperform in timesteps where the meteorological conditions consist of coinciding relative extreme values. Conversely, LSMs perform much better under "typical" conditions within the centre of the meteorological variable distributions. Constraining analysis to exclude the boundary conditions results in the LSMs outperforming strong empirical benchmarks. Encouragingly, we show that refinement of the performance of land surface models in these boundary conditions, consisting of only 12 % to 31 % of time steps, would see large improvements (22 % to 114 %) in an aggregated performance metric. Precise targeting of model development towards these meteorological boundary conditions offers a fruitful avenue to focus model development, ensuring future improvements have the greatest impact.

The exchange of carbon, water, and energy fluxes between the land and the atmosphere plays a vital role in shaping our understanding of global change and how climate change will affect extreme events. Yet our understanding of the theory of this surface-atmosphere exchange, represented via land surface models, continues to be limited, highlighted by marked biases in model-data benchmarking exercises. Here, we leveraged the PLUMBER2 dataset of observations and model simulations of terrestrial fluxes from 153 international eddy-covariance sites to identify the meteorological conditions under which land surface models are performing worse than a priori expectations. By defining performance relative to three sophisticated out-of-sample empirical models, we generated a lower bound of performance in turbulent flux prediction that can be achieved with the input information available to the land surface models during testing at flux tower sites. We found that land surface model (LSM) performance relative to empirical models is worse at boundary conditions – that is, LSMs underperform in timesteps where the meteorological conditions consist of coinciding relative extreme values. Conversely, LSMs perform much better under "typical" conditions within the centre of the meteorological variable distributions. Constraining analysis to exclude the boundary conditions results in the LSMs outperforming strong empirical benchmarks. Encouragingly, we show that refinement of the performance of land surface models in these boundary conditions, consisting of only 12 % to 31 % of time steps, would see large improvements (22 % to 114 %) in an aggregated performance metric. Precise targeting of model development towards these meteorological boundary conditions offers a fruitful avenue to focus model development, ensuring future improvements have the greatest impact.

The exchange of carbon, water, and energy fluxes between the land and the atmosphere plays a vital role in shaping our understanding of global change and how climate change will affect extreme events. Yet our understanding of the theory of this surface-atmosphere exchange, represented via land surface models, continues to be limited, highlighted by marked biases in model-data benchmarking exercises. Here, we leveraged the PLUMBER2 dataset of observations and model simulations of terrestrial fluxes from 153 international eddy-covariance sites to identify the meteorological conditions under which land surface models are performing worse than a priori expectations. By defining performance relative to three sophisticated out-of-sample empirical models, we generated a lower bound of performance in turbulent flux prediction that can be achieved with the input information available to the land surface models during testing at flux tower sites. We found that land surface model (LSM) performance relative to empirical models is worse at boundary conditions – that is, LSMs underperform in timesteps where the meteorological conditions consist of coinciding relative extreme values. Conversely, LSMs perform much better under "typical" conditions within the centre of the meteorological variable distributions. Constraining analysis to exclude the boundary conditions results in the LSMs outperforming strong empirical benchmarks. Encouragingly, we show that refinement of the performance of land surface models in these boundary conditions, consisting of only 12 % to 31 % of time steps, would see large improvements (22 % to 114 %) in an aggregated performance metric. Precise targeting of model development towards these meteorological boundary conditions offers a fruitful avenue to focus model development, ensuring future improvements have the greatest impact.