The weather significantly influences daily life, which is predominantly due to short-term weather phenomena occurring in the atmospheric boundary layer (ABL). The HARMONIE-AROME (HARMONIE) model, used by the Royal Netherlands Meteorological Institute (KNMI), simulates the ABL by discretizing the atmosphere into a three-dimensional grid. Processes occurring at scales significantly larger than these grid can be resolved by the model, but processes occurring at scales smaller than the grid (subgrid) are parameterized by theoretical frameworks. At the current horizontal grid resolution of the HARMONIE model, both shallow convection and smaller-scaled diffuse turbulent transport are parameterized by the Eddy-Diffusivity (ED) and the Mass-Flux (MF) scheme, respectively, coupled in the EDMF-framework. In which the MF is described separately for the dry and the moist (cloudy) updraft.
Increasing the model’s resolution promises an increase in atmospheric representation, yet introduces challenges in the so-called Grey zone of turbulence, where the scale of the turbulent motions are in the same order of magnitude as grid size, making
them neither fully resolved nor fully subgrid.
This study aims to investigate the scale-adaptivity of the HARMONIE EDMF-scheme in the Grey zone of turbulence, for the shallow cumulus boundary layer. To this end, high-resolution Large Eddy Simulation (LES) results of two shallow-cumulus cases are coarse-grained to quantify the partitioning of resolved and unresolved turbulence. It is reviewed how well these partitionings scale against the resolution, normalized with height of the dry (h) and the cloudy boundary layer (h+hc), to investigate the potential of scale-adaptivity of the EDMF-scheme with this height. Additionally, the HARMONIE model is run for one of these cases at three EDMF settings: without scale adaptations, with a scale-adaptive scheme based on both h and hc, and with an additional vertical velocity threshold.
In the dry boundary layer, the partitionings of turbulences showed to scale well with the height of the dry boundary layer h, but also implied additional large scaled turbulent transport not carried by strong updrafts. The scaling down of the dry MF in the HARMONIE run showed significant reduction, with increased resolved transport. However, the unresolved partitioning to the total flux still was higher than expected by LES results, which may be explained by these large scaled turbulent transport not accounted for with the mass-flux. In the cloud layer, scaling the resolved and unresolved partitioning of the total turbulence with the height of the cloud h + hc showed not as effective, and indicated that it may not sufficiently represent the strength of convection in the cloud layer. This is supported by the
HARMONIE run, that showed too much decrease of the moist updraft. The addition of the vertical velocity threshold showed a too strong decrease of mass-flux, both in the dry and in the mixed layer. An additional figure from LES results suggest that this threshold was set too low and scale-adaptivity of this threshold may be needed.

Clouds play a crucial role in Earth’s systems, influencing the radiation budget and the hydrological cycle. However, their dynamics are poorly represented in climate models, leading to uncertainties in predicting global temperature changes. To better understand cloud dynamics and improve model parametrizations, the large-scale field campaign EUREC4A focused on studying cloud organization and its potential impact on climate feedback in the North Atlantic trade-wind region, where shallow marine cumulus clouds are prevalent. Within the EUREC4A framework, this study focuses on cloud fields and organization in the HARMONIE weather model to assess the impact of shallow convective parametrizations on its outputs. By comparing three experiments denominated as ’HARMONIE noHGTQS’, HARMONIE noHGTQS noSHAL’ and ’HARMONIE noHGTQS noUVmix’ one can identify discrepancies and determine what the impact of the various shallow convective parametrizations are. Several cloud organization metrics have been computed for the HARMONIE experiments as well as GOES-16 satellite snapshots. The results show that none of the experiments precisely capture the observed diurnal patterns successfully. There exist disparities in cloud cover, cloud quantity, cloud sizes, and the organization index. Removing shallow convective parametrization partly improves the representation of clouds. It exhibits a stronger correlation with the observed cloud cover data from ceilometer measurements and demonstrates a higher capability to replicate the amount of
cloud entities seen in observations in comparison to the other experiments. Furthermore, disabling shallow convection leads to instability and deeper clouds. On the other hand momentum mixing alone has a small impact on clouds, yet it alters wind patterns, resulting in reduced speeds, and little production of precipitation. Overall the model struggles to reproduce detailed cloud patterns such as sugar, mainly due to its coarse resolution. However, it does exhibit a
degree of day-to-day variability in cloud patterns. To gain a deeper understanding on the role of parametrized shallow convection, further investigation into model configurations, momentum mixing dynamics, and hybrid approaches are proposed.

The exchange of energy and mass between the ocean and atmosphere plays a crucial role in shaping oceanic and atmospheric circulation patterns. However, accurately representing these air-sea fluxes remains a challenge for current weather and climate models. Improving the accuracy of bulk flux pa- rameterizations is crucial to improve the quality of weather forecasts and climate predictions, as these parameterizations play a fundamental role in estimating the air-sea fluxes. This study aims to evaluate the performance of the ECUME and ECUME6 parameterizations in simulating air-sea fluxes by utilizing in situ observations obtained from R/V Ron Brown and R/V Meteor, and conducting a comparison with the COARE3.6 parameterization.

To evaluate the ECUME and ECUME6 parameterizations, surface flux diagnostics are established, which illustrate how air-sea fluxes vary with changes in the respective atmospheric variables. By comparing the surface flux diagnostics of the in situ observations with those of the parameterizations, sources of error are identified. The analysis reveals that both ECUME and ECUME6 tend to overestimate the heat fluxes in comparison to EC observations and the COARE3.6 parameterization, with ECUME6 exhibiting a larger overestimation. The degree of overestimation becomes more pronounced as wind speeds increase. Concerning the momentum flux, the parameterizations exhibit an underestimation, with the discrepancy becoming more significant at elevated wind speeds.

By employing an offline model for ECUME and COARE3.6, the iteratively obtained parameters are compared. This analysis demonstrates that the air-sea fluxes derived from the parameterizations strongly depend on the determined neutral transfer coefficients. Addressing these sources of error and refining the parameterization methodology can improve the accuracy of the parameterizations and enhance their applicability for estimating air-sea exchange between the Earth’s surface and the atmosphere.

In the tropics, the character of the trade-winds is decisive for setting convergence and the large-scale circulation. Nevertheless, the vertical transport of momentum by shallow convection (SCMT) and yet its impact on trade- winds has not been investigated in depth yet. With this study, we aim to contribute in understanding the isolated effect SCMT has on the large scale circulations, setting the strength of the circulation and favouring zones of deep convection. Six climate models participated by each conducting three aquaplanet simulations, referring to a simplified climate model in which the entire planet’s surface is covered by ocean, useful for studying fundamental atmosphere-ocean dynamics in the absence of complex land surface interactions. In three distinctive simulations, the models either turn off or exaggerate their SCMT scheme, alternatively by altering an approximate convective momentum scheme. We hypothesized that given the typical tropical wind profile, north-easterly flow at the surface and south-westerly flow aloft, the depth of mixing is determinative for the long-term reaction in setting the circulation. By deeper types if mixing, the scheme will likely bring opposite winds to the surface, affecting both the magnitude and direction of the trade-winds. A first analysis reveals that certain confine the mixing of surface friction to the sub-cloud layer, showing a clear cut-off of induced tendency between the mixed layer and the cloud layer. It is those models that favour a local response only in the shallow overturning circulation, opposite to the models that actively mix momentum between the mix-layer and cloud-layer and show adjustments in the deep overturning circulation. Further building earlier work showing that the horizontal advection of moist static energy (MSE) in the sub-tropics drives tropical convection, we analyzed to what extent enhanced SCMT leads to a change in atmospheric cooling rates by di- or convergence of MSE. Apparent differences by enhanced SCMT are observed in the in cooling rates by advection of MSE between the models that show strong responses in the deep overturning versus the models that do not.

The use of deep learning in global weather forecasting has shown significant promise in improving both forecasting accuracy and speed. Traditional numerical weather prediction models have gradually improved forecasting skills but at the cost of increased computational complexity. In contrast, new deep learning models, trained directly on reanalysis data, have demonstrated significant gains in forecasting accuracy, achieving competitive levels of performance.

However, the potential of deep learning in predicting spatiotemporal chaotic systems, such as weather patterns, remains unexplored. To address this gap, we investigate the efficacy of a data-driven Fourier neural operator Markovian forecaster to replicate the intrinsic predictability and the characteristic Lyapunov spectrum of the Kuramoto-Sivashinsky system.
FNO reproduces intrinsic predictability and precisely estimates the characteristic Lyapunov spectrum, even with a small dataset.
They cannot represent one of the invariant symmetries, a zero characteristic Lyapunov exponent in the spectrum.

Our findings suggest that deep learning can not only enhance the speed and accuracy of traditional numerical forecasting models but also replicate the weather's chaotic nature.
This has significant implications for generating large ensembles and improving overall probabilistic forecasts.

The research is limited to a deterministic system defined on a single process time scale surrogated by FNO.
The future merits a similar study to investigate if FNO models can perform similarly on larger, stochastic, coupled, and/or multiple time-scale spatiotemporal chaotic systems.

One of the areas of major importance when it comes to weather prediction are clouds, due to their significant impact on the climate and because they are major source of the spread in climate sensitivity in climate models. This paper will focus on the main numerical weather prediction model, HARMONIE, used by the Royal Netherlands Meteorological Institute, with the objective to determine whether this model is able to accurately simulate the meso-scale organization and structures of subtropical low marine clouds.
This is done by comparing various HARMONIE simulation outputs with the observed variables obtained from the Barbados Cloud Observatory. To determine how the different processes within HARMONIE are affecting the simulation outputs, several experimental runs are analyzed. Based on the results it can be
seen that HARMONIE cycle 43 does produce a higher amount of low cloud cover compared to HARMONIE cycle 40. This is expected as HARMONIE cycle 43 does contain the additional height variance which was added in order for the model to produce more low clouds. However, it is overestimating the total low
cloud cover, in contrast to the underestimation by HARMONIE cycle 40. It can be seen that when the additional height variance is removed the simulation of the low cloud cover is significantly improved and is comparable to the observational low cloud cover. Furthermore, when the additional height variance and
shallow convection are both removed the amount of low cloud cover produced by HARMONIE cycle 43 also closely resembles the observed low cloud cover amount. When it comes to cloud fraction in the atmosphere, all of the experimental runs are not able to produce clouds that are present above 3000 m. This result was also obtained from simulation outputs of HARMONIE cycle 40. Besides this, also the specific humidity and potential temperature from the simulation outputs of HARMONIE cycle 43 are not accurately modelled.
All of the experiments seem to have a dry BIAS near the surface and a cold BIAS throughout the whole entirety of the atmosphere. Based on these findings it can be concluded that HARMONIE cycle 43 is able to produce meso-scale subtropical marine clouds, however it does not do this accurately but is overestimating
at all times. Moreover, it can be seen that the additional height variance has significant impact on the cloud production and it is recommended it be removed as the simulation outputs more accurately describe the observational data when it is not present. Furthermore there is no major improvement between cycle 40 and 43 in the cloud formation above 3000m. Both cycles are not able to produce any clouds in the simulation outputs. Based on this new understanding of what is the major source affecting the cloud production in HARMONIE, it is possible to further look into them with the aim to improve the overall weather prediction
of HARMONIE.

Extreme precipitation can be characterized by the tail behavior of precipitation probability distributions.The tail contains the most extreme precipitation events and tells something about both the magnitude and frequency of these events. Here the heaviness amplification factor is used to represent the tail behavior. This research determines the most important hydro-climatological controls of the heaviness amplification factor. The study area is the entire world, where Antarctica and the oceans are not taken into account. A pre-defined set of 17 controls divided into four groups is used. The four groups are: general climate condition, geography, land cover and climate variability. The relation between the heaviness and the controls is determined with multiple linear regression, where the standardized controls are used. Since multiple linear regression is applied, multicollinearity might be a problem. This term refers to the occasion where independent variables are correlated to another independent variable or a linear combination of variables. This could lead to erroneous and unreliable results. To deal with this problem, three elimination methods are applied. This first method consists of a correlation analysis, where two controls are highly correlated if their correlation coefficient is higher than 0.9. The control with the lowest regression coefficient is eliminated. This is followed by another regression analysis with all the remaining controls, where controls with a low regression coefficient are also eliminated. The second method applies an iterative regression procedure, where in each iteration the control with the lowest coefficient is eliminated until three controls remain. The last method calculates the variance inflation factor, which is a measure for multicollinearity. It removes the controls with the highest variance inflation factor if higher than 10, until none of the controls has a score above this limit. The described analysis is done for the entire world, but also for 44 IPCC climate reference regions. The conclusion of the analysis is that the most important controls for the heaviness belong to the general climate condition and the climate variability. In total, 13 of the 17 controls occur in a top three for at least one region in one or more of the elimination methods. The most important control is precipitation variability which is in 57% of the regions in the top three based on the results of methods 1, 2 and 3. Each region gives a different top three or at least a different order of controls. The heaviness is calculated based on the outcomes of the multiple linear regression. The R-squared ranges from 0.2 to 0.8, with an average value around 0.3. This suggests that the controls explain about 30% of the variance. The average error for the world between the data and the calculation of the heaviness is 0.11. When the heaviness of a single region is calculated, the absolute error between the data and the calculation of the heaviness reduces from 0.12 to 0.10 when using regional coefficients. This result leads to the conclusion that regional processes do play a role in the behavior of extreme precipitation. This makes it less appropriate to define a global set of most important controls. None of the three elimination methods is able to solve the problem of multicollinearity completely. On average, 35% of the controls have a opposite sign between the correlation and regression coefficient.

Clouds, specifically shallow clouds, are known as a major source of uncertainty in global climate models. Shallow clouds over the global oceans show different spatial patterns and organizations that may be influenced by climate change. Besides, the frequency of these patterns can change the climate feedback of marine clouds in the subtropics. Experts in atmospheric sciences have proposed different kinds of organization metrics. In a recent paper (Janssens et al., 2021), 21 cloud organization metrics were collected and computed over 5000 satellite scenes of shallow clouds located near the east of Barbados. Almost all of the existing metrics of cloud organization are bulk parameters of the cloud field. Accordingly, over the same data as Janssens et al. (2021), this project aims to define a metric that originated from the mutual arrangements of the individual clouds in the field. To this end, network theory as a mathematical tool is employed to define new cloud organization metrics to explore how cloud objects interact with and coordinate concerning each other. It should be noted that this research is the first in which cumulus cloud fields are considered as complex spatial networks. In this regard, an important question is whether a new network metric can distinctly explain a variability which has not been captured by previously defined metrics in cloud organization. To address the research question, we utilize a multivariate regression model to understand whether a linear combination of principal components of the existing metrics can encapsulate a variation in newly proposed network metrics. We found that degree standard deviation and mean of clustering coefficient are two network metrics that can distinctly capture a variability that has not been encapsulated by the existing metrics in cumulus cloud organizations. Degree standard deviation simultaneously measures the homogeneity of the cloud size distribution and the distance between nearest neighboring clouds distribution. A large value of the average clustering coefficient indicates that the network field consists of either large clouds located on corners of triangles or relatively small clouds closely coagulated. Additionally, two sensitivity analyses were performed to understand how the main results are influenced by either center-spacing or edge-spacing distance between clouds and the nodes (clouds) located close to the boundaries of the field. Finally, one should note that all of the results of network theory-based approaches are comprehensively affected by how the network is defined. Defining a network that combines both geometric and physical interaction between cloud objects will be the future work of this research.

In the well-known process of the turbine design, turbulence intensity (TI) plays a vital role in prediction of the power output and loads on the turbine's structure. TI is believed to be an important statistical parameter of the wind speed that can be extracted from the signals recorded by the dedicated sensors in the wind energy area. Despite the limitations of the mast-mounted sensors, they are probably more popular than LIDARs in wind energy applications. Although the sonic anemometers are reference tools in measuring turbulent features of wind, they are expensive instruments to be employed in a large-scale. In this regard, the cup anemometers appear to be the most commonly used instruments in the wind energy community. Accordingly, it would be tremendously advantageous if the 1-Hz cup anemometer data can be employed with the synthetic down-scaling idea to build the turbulence-like velocity signal fields. In this research, small-scale fluctuations are constructed via the Fractal Interpolation (FI) technique. In addition, this study aims to assess the compatibility of the FI technique in enhancing the cup anemometer data. The analysis has been carried out for the data collected in September 2018. Through this analysis, it is deduced that the cup anemometer data can be improved using the FI method. Subsequently, by applying the FI method, in most of the cases, the standard deviation values of the cup anemometer data are increased.

This report proposes improvements of the land surface model (LSM) used in the turbulence-resolving Dutch Atmospheric Large-Eddy Simulation model (DALES). Important changes include the infiltration of precipitation, the parametrization of the soil hydraulic functions and the formulation of the soil water extraction by vegetation roots. The performance of the improved LSM version is validated using offline simulations and observations from the CESAR meteorological observatory in Cabauw (Netherlands), during the year of 2015. An optimal parameter set for this location set is obtained from both satellite retrievals and in-situ observations. In particular, the seasonal dependence of both the Leaf Area Index (LAI) and the roughness lengths for heat and momentum is prescribed to the model.

The offline simulations determined that the improved version of the LSM is able to model the surface fluxes during the year of 2015 with an accuracy of 10 W/m2. This accuracy was found to be strongly dependent on the chosen land surface parameters such as the LAI, the roughness lengths, the soil textures and the vegetation root profiles. Furthermore, the lower boundary for soil water transfer, set to free drainage, was found to cause too much vegetation stress after an extended simulation time.

Finally, the LSM is tested in a fully-coupled mode during three different meteorological events observed at the CESAR observatory. Firstly, stratocumulus simulations confirm the importance of the surface evaporation on the development of the stratocumulus layer, which is considerably reduced compared to simulations that assume the surface to be saturated with water. Secondly, the growth of the clear convective boundary layer was found to be greatly affected by the partitioning of the surface fluxes, as the sensible heat flux plays a key role in entrainment. Although the modelled sensible heat flux was found to be in agreement with the observations, the growth of the boundary layer was still substantially underestimated. Lastly, it was found that it is possible to model the main development stages of radiation fog in DALES, despite the fact that the modelled fog layer appears to be too persistent.

The wind energy industry is growing more than ever before and wind energy as a renewable energy source has shown quite a potential over the years. Unfortunately, the power yield of a wind farm can fluctuate largely over time, which originates from fluctuating wind speed magnitudes. Convection of air, turbulence, humidity and radiation of heat are processes in the atmospheric boundary layer, the lowest region of the troposphere above Earth’s surface, that are responsible for these fluctuations. This region is characterized by a diurnal cycle in which two examinable cases can be found, namely the convective boundary layer (CBL) case during the daytime period and the stable boundary layer (SBL) case during the nighttime period of the diurnal cycle. In the SBL case, the earlier mentioned physical processes result in a cool surface layer and relatively little turbulence is present, whereas in the CBL case a relatively warm surface layer and high turbulence level are present. The Dutch Atmospheric Large Eddy Simulation (DALES) model is used in which wind turbines are implemented with the help of former TU Delft master student P.A. van Dorp. With DALES, a simulation is performed for each case and the output data of the two simulations are analysed and compared to find the optimal case in terms of total power yield of a configuration of two wind turbines with the second fully in the wake produced by the one in front. An arbitrary distance of 600m between the wind turbines is chosen, which corresponds to 7.5 times the diameter length of the wind turbines, and a hub height of 80m is chosen. First, the simulations are validated by examining the turbulent kinetic energy profiles. Then, the wind speed profiles over the domain are analysed, corresponding to the turbulent kinetic energy profiles and showing that the average wind speed at hub height for the SBL case is larger than for the CBL case. Furthermore, the wake profiles behind the turbines are displayed and compared for the two cases, showing little difference. Finally, the total power yield of this specific wind turbine configuration is calculated. With an average power yield of 1.71MW for the SBL case compared to 0.90MW for the CBL case, the SBL case is shown to be the optimal atmospheric boundary layer case in terms of power production for this specific wind turbine configuration.