Measurements of three flux towers operated during the land atmosphere feedback experiment (LAFE) are used to investigate relationships between surface fluxes and variables of the land–atmosphere system. We study these relations by means of two machine learning (ML) techniques: multilayer perceptrons (MLP) and extreme gradient boosting (XGB). We compare their flux derivation performance with Monin–Obukhov similarity theory (MOST) and a similarity relationship using the bulk Richardson number (BRN). The ML approaches outperform MOST and BRN. Best agreement with the observations is achieved for the friction velocity. For the sensible heat flux and even more so for the latent heat flux, MOST and BRN deviate from the observations while MLP and XGB yield more accurate predictions. Using MOST and BRN for latent heat flux, the root mean square errors (RMSE) are 107 Wm$$^{-2}$$ - 2 and 121 Wm$$^{-2}$$ - 2 , respectively, as well as the intercepts of the regression lines are $$\approx 110$$ ≈ 110  Wm$$^{-2}$$ - 2 . For the ML methods, the RMSEs reduce to 31 Wm$$^{-2}$$ - 2 for MLP and 33 Wm$$^{-2}$$ - 2 for XGB as well as the intercepts to just 4 Wm$$^{-2}$$ - 2 for MLP and $$-1$$ - 1  Wm$$^{-2}$$ - 2 for XGB with slopes of the regression lines close to 1, respectively. These results indicate significant deficiencies of MOST and BRN, particularly for the derivation of the latent heat flux. In fact, in contrast to the established theories, feature importance weighting demonstrates that the ML methods base their improved derivations on net radiation, the incoming and outgoing shortwave radiations, the air temperature gradient, and the available water contents, but not on the water vapor gradient. The results imply that further studies of surface fluxes and other turbulent variables with ML techniques provide great promise for deriving advanced flux parameterizations and their implementation in land–atmosphere system models.

Abstract. Important topics in land–atmosphere (L–A) feedback research are water and energy balances and heterogeneities of fluxes at the land surface and in the atmospheric boundary layer (ABL). To target these questions, the Land–Atmosphere Feedback Observatory (LAFO) has been installed in southwestern Germany. The instrumentation allows comprehensive and high-resolution measurements from the bedrock to the lower free troposphere. Grouped into three components, atmosphere, soil and land surface, and vegetation, the LAFO observation strategy aims for simultaneous measurements in all three compartments. For this purpose the LAFO sensor synergy contains lidar systems to measure the atmospheric key variables of humidity, temperature and wind. At the land surface, eddy covariance stations are operated to record the energy distribution of radiation, sensible, latent and ground heat fluxes. Together with a water and temperature sensor network, the soil water content and temperature are monitored in the agricultural investigation area. As for vegetation, crop height, leaf area index and phenological growth stage values are registered. The observations in LAFO are organized into operational measurements and intensive observation periods (IOPs). Operational measurements aim for long time series datasets to investigate statistics, and we present as an example the correlation between mixing layer height and surface fluxes. The potential of IOPs is demonstrated with a 24 h case study using dynamic and thermodynamic profiles with lidar and a surface layer observation that uses the scanning differential absorption lidar to relate atmospheric humidity patterns to soil water structures. Both IOPs and long-term observations will provide new insight into exchange processes and their statistics for improving the representation of L–A feedbacks in climate and numerical weather prediction models. The lidar component in particular will support the investigation of coupling to the atmosphere.

Hydrology is a major environmental factor determining plant fitness, and hydrological niche segregation (HNS) has been widely used to explain species coexistence.Nevertheless, the distribution of plant species along hydrological gradients does not only depend on their hydrological niches but also depend on their seed dispersal, with dispersal either weakening or reinforcing the effects of HNS on coexistence. However, it is poorly understood how seed dispersal responds to hydrological conditions. To close this gap, we conducted a common-garden experiment exposing five wind-dispersed plant species (Bellis perennis, Chenopodium album, Crepis sancta, Hypochaeris glabra, and Hypochaeris radicata) to different hydrological conditions. We quantified the effects of hydrological conditions on seed production and dispersal traits, and simulated seed dispersal distances with a mechanistic dispersal model. We found species-specific responses of seed production, seed dispersal traits, and predicted dispersal distances to hydrological conditions. Despite these species-specific responses, there was a general positive relationship between seed production and dispersal distance: Plants growing in favorable hydrological conditions not only produce more seeds but also disperse them over longer distances. This arises mostly because plants growing in favorable environments grow taller and thus disperse their seeds over longer distances. We postulate that the positive relationship between seed production and dispersal may reduce the concentration of each species to the environments favorable for it, thus counteracting species coexistence. Moreover, the resulting asymmetrical gene flow from favorable to stressful habitats may slow down the microevolution of hydrological niches, causing evolutionary niche conservatism.Accounting for context-dependent seed dispersal should thus improve ecological and evolutionary models for the spatial dynamics of plant populations and communities.

Abstract. Important topics in Land-Atmosphere (L-A) feedback research are water and energy balances and heterogeneities of fluxes at the land-surface and in the atmospheric booundary layer. To target these questions, the Land-Atmosphere Feedback Observatory (LAFO) has been installed in Southwest Germany. The instrumentation allows comprehensive and high-resolution measurements from the bedrock to the lower free troposphere. Grouped in three components: atmosphere, soil and land-surface and vegetation, the LAFO observation strategy aims for simultaneous measurements in all three compartments. For that the LAFO sensor synergy contains lidar systems to measure the atmospheric key variables humidity, temperature and wind. At the land-surface eddy covariance stations operated to record the energy distribution of radiation, sensible, latent and ground heat fluxes. With a water and temperature sensor network the soil water and temperature is monitored in the agricultural investigation area. The observations in LAFO are organized in operational measurements and intensive observation periods (IOPs). Operational measurements aim for long timeseries dataset to investigate statistics as we present as example the correlation between mixing layer height and surface fluxes. The potential of IOPs is demonstrated with a 24 hour case study with dynamic and thermodynamic profiles with lidar as well as a surface layer observation with the scanning differential absorption lidar to relate atmospheric humidity patterns to soil water structures. Both long-term observations and IOPs are important for improving the representation of L-A feedbacks in climate and numerical weather prediction models.

<p>Soil, land cover and the lower atmosphere form the land atmosphere system. The feedbacks in this system are nonlinear, because of two-way coupling between single variables such as soil moisture and precipitation. A detailed characterization of fluxes and feedbacks within and between the different compartments is essential to improve our understanding of the land atmosphere system. To investigate these fluxes and feedbacks the Land Atmosphere Feedback Observatory (LAFO) was established at the experimental station “Heidfeldhof” of the University of Hohenheim in 2018. LAFO applies a novel synergy of eddy covariance stations, vegetation measurements, and innovative scanning lidar systems. The measurements are comprehensive, highly resolved and very precise, so that new parameterizations of land-atmosphere exchange processes between soil/vegetation and the lower troposphere for model systems can be developed, implemented, and tested.</p><p>LAFO is situated in heterogeneous cropland with nearby urban areas and forests. The heterogeneous terrain forms a complex land atmosphere system. A footprint analysis of eddy covariance data from 2019 enables us to better characterize our site.  Here we present LAFO, its concept and first results on a footprint analysis of the eddy covariance data.</p>