A long-standing challenge in studying the global carbon cycle has been understanding the factors controlling inter–annual variation (IAV) of carbon fluxes related to vegetation photosynthesis and respiration, and improving their representations in existing biogeochemical models. Here, we compared an optimality-based mechanistic model and a semi-empirical light use efficiency model to understand how current models can be improved to simulate IAV of gross primary production (GPP). Both models simulated hourly GPP and were parameterized for (1) each site–year, (2) each site with an additional constraint on IAV (CostIAV), (3) each site, (4) each plant–functional type, and (5) globally. This was followed by forward runs using calibrated parameters, and model evaluations at different temporal scales across 198 eddy covariance sites. Both models performed better on hourly scale than annual scale for most sites. Specifically, the mechanistic model substantially improved when drought stress was explicitly included. Most of the variability in model performances was due to model types and parameterization strategies. The semi-empirical model produced statistically better hourly simulations than the mechanistic model, and site–year parameterization yielded better annual performance for both models. Annual model performance did not improve even when parameterized using CostIAV. Furthermore, both models underestimated the peaks of diurnal GPP in each site–year, suggesting that improving predictions of peaks could produce a comparatively better annual model performance. GPP of forests were better simulated than grassland or savanna sites by both models. Our findings reveal current model deficiencies in representing IAV of carbon fluxes and guide improvements in further model development.

Land surface temperature (LST) is crucial for understanding earth system processes. We expanded the Advanced Baseline Imager Live Imaging of Vegetated Ecosystems (ALIVE) framework to estimate LST in near-real-time for both cloudy and clear sky conditions at a 5-minute resolution. We compared two machine learning models, Long Short-Term Memory (LSTM) networks and Gradient Boosting Regressor (GBR), using top-of-atmosphere (TOA) observations from the Advanced Baseline Imager (ABI) on the GOES-16 satellite against observations from hundreds of measurement locations for a 5-year period. LSTM outperformed, especially at coarser resolutions and under challenging conditions, with a clear sky R² of 0.96 (RMSE 2.31 K) and a cloudy sky R² of 0.83 (RMSE 4.10 K) across CONUS, based on 10-repeat Leave-One-Out Cross-Validation (LOOCV). GBR maintained high accuracy (R² > 0.90) and ran 5.3 times faster, with only a 0.01-0.02 R² drop. Feature importance revealed infrared bands were key in both models, with LSTM adapting dynamically to atmospheric changes, while GBR utilized time information in cloudy conditions. A comparative analysis against the physically based ABILST product showed strong agreement in winter, particularly under clear sky conditions, while also highlighting the challenges of summer LST estimation due to increased thermal variability. This study underscores the strengths and limitations of data-driven models for LST estimation and suggests potential pathways for integrating these approaches to enhance the accuracy and coverage of LST products.

Dry deposition is the second-largest tropospheric ozone (O3) sink and occurs through stomatal and nonstomatal pathways. Current O3 uptake predictions are limited by the simplistic big-leaf schemes commonly used in chemical transport models (CTMs) to parameterize deposition. Such schemes fail to reproduce observed O3 fluxes over terrestrial ecosystems, highlighting the need for more realistic treatment of surface-atmosphere exchange in CTMs. We address this need by linking a resolved canopy model (1D Multi-Layer Canopy CHemistry and Exchange Model, MLC-CHEM) to the GEOS-Chem CTM, and use this new framework to simulate O3 fluxes over three north temperate forests. We compare results with in-situ measurements from four field studies and with standalone, observationally-constrained MLC-CHEM runs to test current knowledge of O3 deposition and its drivers. We show that GEOS-Chem overpredicts observed O3 fluxes across all four studies by up to 2×, whereas the resolved-canopy models capture observed diel profiles of O3 deposition and in-canopy concentrations to within 10%. Relative humidity and solar irradiance are strong O3 flux drivers over these forests, and uncertainties in those fields provide the largest remaining source of model deposition biases. Flux partitioning analysis shows that: 1) nonstomatal loss accounts for 60% of O3 deposition on average; 2) in-canopy chemistry makes only a small contribution to total O3 fluxes; and 3) the CTM big-leaf treatment overestimates O3-driven stomatal loss and plant phytotoxicity in these temperate forests by up to 7×. Results motivate the application of fully-online, vertically explicit canopy schemes in CTMs for improved O3 predictions.

The spatiotemporal variability of latent heat flux (LE) and water vapor mixing ratio (rv) variability are not well understood due to the scale-dependent and nonlinear atmospheric energy balance responses to land surface heterogeneity. Airborne in situ and profiling Raman lidar measurements with the wavelet technique are utilized to investigate scale-dependent relationships among LE, vertical velocity (w) variance (s2w), and rv variance (s2wv) over a heterogeneous surface in the Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors 2019 (CHEESEHEAD19) field campaign. Our findings reveal distinct scale distributions of LE, s2w, and s2wv at 100 m height, with a majority scale range of 120m-4km in LE, 32m-2km in s2w, and 200 m – 8 km in s2wv. The scales are classified into three scale ranges, the turbulent scale (8m–200m), large-eddy scale (200m–2km), and mesoscale (2 km–8km) to evaluate scale-resolved LE contributed by s2w and s2wv. In the large-eddy scale in Planetary Boundary Layer (PBL), 69-75% of total LE comes from 31-51% of the total sw and 39-59% of the total s2wv. Variations exist in LE, s2w, and s2wv, with a range of 1.7-11.1% of total values in monthly-mean variation, and 0.6–7.8% of total values in flight legs from July to September. These results confirm the dominant role of the large-eddy scale in the PBL in the vertical moisture transport from the surface to the PBL. This analysis complements published scale-dependent LE variations, which lack detailed scale-dependent vertical velocity and moisture information.

The drag coefficient (CDN), Stanton number (CHN) and Dalton number (CEN) are of particular importance for the bulk estimation of the surface turbulent fluxes of momentum, heat and water vapor at water surfaces. Although these bulk transfer coefficients have been extensively studied over the past several decades mainly in marine and large-lake environments, there are no studies focusing on their synthesis for many lakes. Here, we evaluated these coefficients through directly measured surface fluxes using the eddy-covariance technique over more than 30 lakes and reservoirs of different sizes and depths. Our analysis showed that generally CDN, CHN, CEN (adjusted to neutral atmospheric stability) were within the range reported in previous studies for large lakes and oceans. CHN was found to be on average a factor of 1.4 higher than CEN for all wind speeds, therefore, likely affecting the Bowen ratio method used for lake evaporation measurements. All bulk transfer coefficients exhibit substantial increase at low wind speeds (< 3 m s-1), which could not be explained by any of the existing physical approaches. However, the wind gustiness could partially explain this increase. At high wind speeds CDN, CHN, CEN remained relatively constant at values of 2 10-3, 1.5 10-3, 1.1 10 -3, respectively. We found that the variability of the transfer coefficients among the lakes could be associated with lake surface area or wind fetch. The empirical formula C=b1[1+b2exp(b3 U10)] described the dependence of CDN, CHN, CEN on wind speed well and it could be beneficial for modeling when coupling atmosphere and lakes.

The observing system design of multi-disciplinary field measurements involves a variety of considerations on logistics, safety, and science objectives. Typically, this is done based on investigator intuition and designs of prior field measurements. However, there is potential for considerable increase in efficiency, safety, and scientific success by integrating numerical simulations in the design process. Here, we present a novel approach to observing system simulation experiments that aids surface-atmosphere synthesis at the interface of meso- and microscale meteorology. We used this approach to optimize the Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors 2019 (CHEESEHEAD19). During pre-field simulation experiments, we considered the placement of 20 eddy-covariance flux towers, operations for 72 hours of low-altitude flux aircraft measurements, and integration of various remote sensing data products. High-resolution Large Eddy Simulations generated a super-sample of virtual ground, airborne, and satellite observations to explore two specific design hypotheses. We then analyzed these virtual observations through Environmental Response Functions to yield an optimal aircraft flight strategy for augmenting a stratified random flux tower network in combination with satellite retrievals. We demonstrate how this novel approach doubled CHEESEHEAD19’s ability to explore energy balance closure and spatial patterning science objectives while substantially simplifying logistics. Owing to its extensibility, the approach lends itself to optimize observing system designs also for natural climate solutions, emission inventory validation, urban air quality, industry leak detection and multi-species applications, among other use cases.

The Earth’s surface is heterogeneous at multiple scales owing to spatial variability in various properties. The atmospheric responses to these heterogeneities through fluxes of energy, water, carbon and other scalars are scale-dependent and non-linear. Although these exchanges can be measured using the eddy covariance technique, widely used tower-based measurement approaches suffer from spectral losses in lower frequencies when using typical averaging times. However, spatially resolved measurements such as airborne eddy covariance measurements can detect such larger scale (meso-{$\beta$}, $\gamma$) transport. To evaluate the prevalence and magnitude of these flux contributions we applied wavelet analysis to airborne flux measurements over a heterogeneous mid-latitude forested landscape, interspersed with open water bodies and wetlands. The measurements were made during the Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors (CHEESEHEAD19) intensive field campaign. We ask, how do spatial scales of surface-atmosphere fluxes vary over heterogeneous surfaces across the day and across seasons? Measured fluxes were separated into smaller-scale turbulent and larger-scale mesoscale contributions. We found significant mesoscale contributions to H and LE fluxes through summer to autumn which wouldn’t be resolved in single point tower measurements through traditional time-domain half-hourly Reynolds decomposition. We report scale-resolved flux transitions associated with seasonal and diurnal changes of the heterogeneous study domain. This study adds to our understanding of surface atmospheric interactions over unstructured heterogeneities and can help inform multi-scale model-data integration of weather and climate models at a sub-grid scale.

The Chequamegon Heterogeneous Ecosystem Energy-balance Study Enabled by a High-density Extensive Array of Detectors 2019 (CHEESEHEAD19) is an ongoing National Science Foundation project based on an intensive field campaign that occurred from June-October 2019. The purpose of the study is to examine how the atmospheric boundary layer responds to spatial heterogeneity in surface energy fluxes. One of the main objectives is to test whether lack of energy balance closure measured by eddy covariance (EC) towers is related to mesoscale atmospheric processes. Finally, the project evaluates data-driven methods for scaling surface energy fluxes, with the aim to improve model-data comparison and integration. To address these questions, an extensive suite of ground, tower, profiling, and airborne instrumentation was deployed over a 10×10 km domain of a heterogeneous forest ecosystem in the Chequamegon-Nicolet National Forest in northern Wisconsin USA, centered on the existing Park Falls 447-m tower that anchors an Ameriflux/NOAA supersite (US-PFa / WLEF). The project deployed one of the world’s highest-density networks of above-canopy EC measurements of surface energy fluxes. This tower EC network was coupled with spatial measurements of EC fluxes from aircraft, maps of leaf and canopy properties derived from airborne spectroscopy, ground-based measurements of plant productivity, phenology, and physiology, and atmospheric profiles of wind, water vapor, and temperature using radar, sodar, lidar, microwave radiometers, infrared interferometers, and radiosondes. These observations are being used with large eddy simulation and scaling experiments to better understand sub-mesoscale processes and improve formulations of sub-grid scale processes in numerical weather and climate models.

Groundwater depletion in Central Wisconsin, due in part to agricultural high-capacity wells, has sparked an interest in precision irrigation to reduce groundwater pumping without a significant reduction in yield. A key challenge for bridging precision irrigation research and application is how best to monitor water stress in real-time. Aerial and satellite imagery are potential solutions. Drawbacks of these methods include cost, spatiotemporal resolution, and cloud interference, especially in humid regions. Recent advancements in remotely piloted aircrafts (RPAs) have made frequent, low-flying imagery collection more economical and feasible than ever before. We partnered with the Wisconsin Potato and Vegetable Grower Association to generate high-resolution maps of crop water stress using remotely sensed thermal and multi-spectral RPA imagery. Data were collected at a commercially irrigated potato field in the Central Sands region of Wisconsin from June to August 2019. Missions were flown weekly using a quadcopter RPA system instrumented with a newly released, combined multispectral/thermal camera developed for agricultural applications. Each mission included flights at 30, 60, and 90 m above ground level to assess tradeoffs between resolution, area, and flight time. We used biophysical data from an eddy covariance system installed within the flight domain to validate crop water stress maps generated from the remotely sensed RPA data. Ground measurements of surface temperature and soil moisture were collected throughout the domain within fifteen minutes of each mission. Ongoing results will be used to develop best practices for integrating RPAs into precision irrigation programs.

The surface is the interface through which the bedrock interacts with the boundary-layer. Here, living organisms of various shapes, sizes, and function intermingle with the mineral soil, organic residue, and canopy air space that reside there. Together, they breath, absorb momentum, exchange water and gases, and bask in the heat from the sun and clouds above and thermal reservoirs below. While many of these functions are well understood, we knew less about how those functions operate and behave at different spatial and time scales. More intriguing, surface variance with scale influences scales of motion in the atmosphere. Here, I present a generalized look at how land and atmosphere scales interact, focusing on the lens of the surface energy budget. These processes are investigated through intensive measurements and high-resolution models conducted at the CHEESEHEAD19 field experiment in Wisconsin. A combination of airborne and tower eddy covariance networks, drone and airborne canopy imaging, and turbulence resolving simulations reveal persistent mesoscale contributions in the atmosphere enabled by surface heterogeneity.

Lakes emit globally significant amounts of carbon dioxide (CO) to the atmosphere, but quantifying these rates for individual lakes is extremely challenging. The exchange of CO across the air-water interface is driven by physical, chemical, and biological processes in both the lake and the atmosphere that vary at multiple spatial and temporal scales. None of the methods we use to estimate CO flux fully capture this heterogeneous process. Here, we compared concurrent CO flux estimates from a single lake based on commonly used methods. These include floating chambers (FC), eddy covariance (EC), and two concentration gradient based methods labelled fixed (F-CO₂) and spatial (S-CO₂). At the end of summer, cumulative carbon fluxes were similar between EC, F-CO₂ and S-CO₂ methods (-4, -4 and -9.5 gC), while methods diverged in directionality of fluxes during the fall turnover period (-50, 43 and 38 gC). Collectively these results highlight the discrepancies among methods and the need to acknowledge the uncertainty when using any of them to approximate this heterogeneous process.

Extratropical cyclones are major contributors to consequential weather in the mid-latitudes and tend to develop in regions of enhanced cyclogenesis and progress along climatological storm tracks. Numerous studies have noted the influence that terrestrial snow cover exerts on atmospheric baroclinicity which is critical to the formation and trajectories of such cyclones. Fewer studies have examined the explicit role which continental snow cover extent has in determining cyclones intensities, trajectories, and precipitation characteristics. While several examinations of climate model projections have generally shown a poleward shift in storm tracks by the late 21st century, none have determined the degree to which the coincident poleward shift in snow extent is responsible. A method of imposing 10th , 50th , and 90th percentile values of snow retreat between the late 20th and 21st centuries as projected by 14 Coupled Model Intercomparison Project Phase Five (CMIP5) models is used to alter 20 historical cold season cyclones which tracked over or adjacent to the North American Great Plains. Simulations by the Advanced Research version of the Weather Research and Forecast Model (WRF-ARW) are initialized at 0 to 4 days prior to cyclogenesis. Cyclone trajectories and their central sea level pressure did not change substantially, but followed consistent spatial trends. Near-surface wind speed generally increased, as did precipitation with preferred phase change from solid to liquid state. Cyclone-associated precipitation often shifted poleward as snow was removed. Variable responses were dependent on the month in which cyclones occurred, with stronger responses in the midwinter than the shoulder months.