Ever increasing global heat has necessitated more robust urban shade analysis in order to keep, among other applications, pedestrians safe and comfortable. Specifically, the impacts of shade-trees on the urban shade landscape is a topic of growing interest and research, and new tools and workflows are being developed in order to allow urban planners to have the clearest understanding of how to best provide shade to citizens. Geometric shade calculation techniques have been used for some time, but they are slow to run at scale and can involve convoluted workflows that slow down analysis. Thus, this thesis investigates the applicability of deep learning to generate shade rasters. The goal is to see if deep learning can allow for faster shade map derivations and a smoother workflow, without sacrificing accuracy. Furthermore, it is the first research of its kind to consider a purely deep learning based approach in deriving tree-aware urban shade maps which capture the contributions of shade trees.

This thesis presents a Conditional Generative Adversarial Network based off of the pix2pix framework, and explores four different combinations of input data and how they affect the final model performance. The model can produce shade rasters ~25 - 45.5% faster than the geometric approach, depending on the parameters used when generating outputs. It also does this using less data and less preprocessing than the geometric alternative. However, while it produces very convincing rasters in some cases, the outputs have insufficient temporal coherence to serve as a reliable alternative to the geometric approach. Thus, while it serves as an interesting foundation for future research, the model as presented in this report is not sufficient to replace geometric shade derivation techniques. ... Read more

To help prevent flooding of rivers and cities, Dutch maritime contractor Van Oord regularly dredged 52 side channels as part of the Dutch Department of Waterways and Public Works' (Rijkswaterstaat) "Room for Rivers" strategy. Side channels make rivers more resilient to flooding by providing increased flow capacity, buffer space, and a secondary path downstream for water. Van Oord wishes to know how they can better leverage their growing historical data collection to enable predictive maintenance of side channels in the form of dredging.
Instead of developing a complex hydrological model, which would require deep knowledge of river morphology. We, as Geomatics students, extracted insights directly from the available geospatial data. For our 10-week MSc Geomatics Synthesis Project, our main research question is as follows: "How can the features of a side channel be identified and extracted to enable predictive maintenance?"
In order to answer this question for our client Van Oord, we performed a literature review and interviewed domain experts to identify relevant characteristics of side channels. Then, we explored the available geo-spatial data to determine which characteristics can be modeled as features, before processing the data in an FME pipeline to calculate these feature values in an automated, extendible, and understandable way. These features were then stored in a geo-spatial database. Reading from this database, we created a prototype machine learning model that takes the features as input. The model enables analysis of the side channels to derive insights into the sedimentation of side channels, reaching 84% accuracy within a 5cm error for the Bakenhof channel.

The result is a robust FME-based data processing pipeline, a geo-spatial database with 19 unique features for 26 suitable side channels, and a prototype neural network showing significant predictive ability. The product enables the client to better estimate side channel behavior, enabling informed predictive maintenance, as well as allowing the client to better decide moments when expensive channel measurements can be skipped. ... Read more