The adoption of hydrogen as an alternative fuel in aircraft has the potential to reduce the climate effect of aviation significantly. However, hydrogen leakage during production, storage, or use can offset these benefits by altering atmospheric chemistry and composition, particularly through interactions with methane, ozone, and stratospheric water vapour. This study employs surrogate models based on recurrent and convolutional neural networks to simulate the climate effects of hydrogen leaks, achieving rapid projections 30,000 times faster than conventional climate models, with an error margin of less than 5%. This efficiency enables the quantification of uncertainties related to hydrogen leakage rates and atmospheric chemistry through Monte Carlo simulations, allowing for an assessment of their contributions to radiative forcing under various Shared Socioeconomic Pathway (SSP) climate scenarios. By 2100, the radiative forcing from aviation-induced hydrogen leaks is projected to reach 43.9 ± 21.2 mW m-2 (±1σ) under the stringent climate change mitigation projection SSP1-2.6, accounting for 35% of aviation’s total radiative forcing. Under the more conservative scenario SSP3-7.0, higher methane levels reduce the oxidative capacity of the atmosphere, lowering the projection to 17.7 ± 6.9 mW m-2 (±1σ) which corresponds to approximately 5% of the aviation’s total radiative forcing for this scenario. These findings demonstrate that hydrogen leaks have the potential to substantially contribute to aviation’s total radiative forcing, with the magnitude of their impact heavily influenced by background climate conditions. It will be important to minimise hydrogen leakage to fully harness the climate benefits of transitioning to hydrogen as a fuel for aircraft.

Climate change causes an increase of precipitation intensity and frequency, which leads to an increased amount of landslide occurrences. Therefore, there is an urgent need for prevention measures. The risk of landslide occurrence can be mitigated by decreasing the pore water pressure in a hillslope, e.g. by drainage from the subsurface. The implementation of Nature Based Solutions (NBS) is advantageous, as these are not disruptive for the hillslope ecosystem. A Live Pole Drain (LPD) is a type of NBS which is inexpensive and can be installed at remote locations. However, its effect on hillslope hydrology has not been studied before.

In this study, fieldwork at an Open Air Lab, a laboratory experiment with miniature LPDs and the development of a conceptual model were used to characterise the hydrological behaviour of a slope in which a LPD is installed. After comparison of this hydrological behaviour to that of a slope with bare soil, it was found that a LPD facilitates rapid infiltration and drainage of precipitation and subsurface water. Furthermore, runoff volumes decrease after installation of a LPD and this effect is enhanced through plant development. Vegetation also contributes to the decrease of the stored volume of water in the subsurface through evapotranspiration. An increase of the design aspects diameter and macropore fraction of the LPD increases its drainage capacity. This study resulted in a first version of a numerical model to quantify the hydrological processes in a hillslope with LPD and the confirmation of the expected effect of drainage through preferential flow by a LPD.

MOSES DSS web-platform aims to assist stakeholders such as governments and farmers in order to manage water irrigation distribution in a higher efficiency and sustainability. The constructed algorithms are focused on forecast using weather models, data, as well as satellite multispectral observations in such a way that a 7-day ahead crop water requirement estimation is generated. The current drawback of the system in using the available and free satellite products such as Landsat 8 and Sentinel 2, is that it assumes that the crops are under standard conditions, e.g. there is no water stress, diseases etc.. The current work investigates how possible errors due to this assumption can be potentially tackled in the future by comparing Crop Water Requirement (CWR) with S2REP VI and/or with water stress index and see the discriminative power of the latter. Furthermore, a comparison between several discrepancies between S2 and L8 (e.g. AC and co-registration) are studied since it is a crucial issue especially in temporal applications such as MOSES. On the one hand, the results showed that a harmonization of the two products is certainly needed. On the other hand, it seems that S2REP is capable of revealing crop stress information based on the methodologies of this work, thus it could potentially give more information compared to NDVI which is not sensitive to crop stress.

Glacier fluctuations are regarded as one of the most significant indicators of climate change. The expansion and contraction of glaciers can be observed by outlining glacier boundaries or measuring snow lines with optical Earth observation satellites. New satellites, such as Sentinel-2A/B, provide high spatial resolution images and short revisit times that can be used to make ample measurements to accurately determine glacial variability. Likewise, ever increasing volumes of satellite data make automated boundary and snow line detection a desirable solution for researchers.

Clouds, however, present a challenge to obtaining useful optical image data. Mountainous regions are often surrounded or covered by clouds. And clouds can be a menacing phenomenon in remote sensing because they greatly attenuate and reflect the short wavelengths used by optical Earth observation satellites. Thus, prior to snow cover or glacier analysis, clouds must be separated from cloud-free pixels. Currently, many techniques exist to automatically detect clouds and classify them. However, they are not perfect. Many techniques have encountered difficulties in cases where snowy and icy landscapes share similar properties with clouds. The mountainous topography also adds to the challenge. Steep slopes and topographic shadows have a profound effect on surface reflectances and can lead to misclassifications.

To address these issues, this study presents an assessment of multiple cloud detection techniques. Several sentinel-2A scenes were acquired at various times between 12/07/15 and 31/12/16. The scenes are centered on the Bara Shigri (glacier), a large 11 km long glacier in Himachal Pradesh, India. Eventually, 6 scenes were processed using 6 different cloud detection techniques: Default L1C, Sen2Cor, Fmask, Temporal Averaging, Maximum Likelihood Classification, and a human-based manual mask which was used as a reference mask.

This study found that automatic cloud detection methods over mountainous terrain performed poorly. Results showed that Maximum Likelihood Classification was the most accurate automatic technique, followed by Sen2Cor. Cloud detection methods often misclassified mountain peaks and glaciers as clouds. At lower altitudes, cloud detection often missed marking known clouds by classifying them as cloud-free instead. Furthermore, it was found that a majority of snow and ice pixels are relatively easy to distinguish from clouds using Sentinel-2's band B3 and B11, or NDSI feature. However, a minority of snow and ice pixels, particularly near cloud edges, are almost indistinguishable from clouds. Results also strongly suggest that mountainous topography negatively affects cloud detection performance. Finally, this study came to the conclusion that cloud detection methods have the potential to improve, but currently are not yet beneficial to glacier research.

Glaciers play an important role in sea level predictions and are an important supplier of fresh water. Understanding the physics and dynamics of glaciers is important in local and global climate predictions and predictions of fresh water supplies. The Himalayas are the biggest storage of fresh water outside the polar regions with many different local climates and glaciers. One important glacier parameter used in dynamic studies is the flow velocity, as the flow velocity is used as a boundary condition in mass balance and run-off models. The flow velocity of a glacier is relatively easy to measure on a large scale using satellite missions and the increased coverage of satellite missions provides the means to large scale velocity monitoring. Robust methods for measuring large scale seasonal and long term velocity dynamics of glaciers, however, remain an elusive goal.

The optical Landsat mission has a long history of monitoring and is thus useful for long term flow velocity analysis. Feature tracking algorithms applied on these Landsat images provide means of automatically calculating large scale velocities. Automated approaches for large scale analysis are difficult because of the lack of validation and good filtering techniques to deal with shadows, surface changes and clouds. Large scale temporal analysis is even harder because the errors arising in the flow velocity calculations are often large compared to the velocities due to the short periods between the images from which the velocities are calculated.

This research proposes, implements and tests a new method for automatically creating large scale velocity time series using the optical Landsat database and feature tracking in the Himalayas. Where normally velocity time series consists of consecutive single velocity fields, the novel method uses combinations of velocities to estimate these single velocities. This method is tested against results from single velocity fields for the Everest region and the Karakoram region. A sensitivity and parameter analysis provides the best parameter settings for the new method.

The result is a novel method that provides validation, robustness and acts as a filter for erroneous velocities. When possible, the new method increases or retains the number of results while increasing the precision. Furthermore, when the number of results is lower, it is due to filtering of erroneous velocities. The new method also provides an error indication which could be of use in future research. The main sources of errors: geo-location and precision of the feature tracking algorithm, are shown to have a large effect on the results and should be as small as possible. The magnitude of these errors make it difficult to measure seasonal changes in flow velocity for slow moving glaciers. Furthermore, Landsat 4-5 TM results are shown not to be useful for dense time series as the geo-location accuracy and artifacts from the TM scanner have a large effect on the precision of these results. The Landsat 8 velocity results, however, show good similarity to results from other researches and show promise for future research.

The sensitivity analysis of this research was incomplete as the complete algorithm was not optimized and future developments might focus on improving this. Also, it was found that the proposed method was very time intensive. A correct choice for the method of calculating flow velocities might drastically decrease the computational effort and improve applicability to large area research. In total, the new method is not only useful for use on the Landsat database, but could be used to improve any optical or radar flow velocity time series.