On February 9, 2023, two significant earthquakes of magnitudes Mw 7.8 and Mw 7.7 struck southern Turkey, devastating the Kahramanmaraş region, causing nearly 50,000 fatalities, and displacing millions of people (Zhang et al., 2023). The first earthquake ruptured the Nurdağı–Pazarcık Fault, a subsidiary of the East Anatolian Fault Zone (EAFZ), while the second occurred about 90 km away on the Çardak–Sürgü Fault. This event later became known as the Kahramanmaraş earthquake doublet. With a combined magnitude of Mw 8, it represents the largest recorded continental strike-slip earthquake sequence.
The main strand of the EAFZ is a well-known seismically active zone, with many seismic records and studies conducted on the area. In contrast, the earthquake on the Çardak–Sürgü Fault was unexpected, as it had remained quiet for several centuries. This has left the Çardak–Sürgü Fault relatively understudied compared to the main strand of the EAFZ, particularly in terms of its kinematics and seismic risk.
In this study, we estimate the slip deficit along the main strand of the East Anatolian Fault Zone (EAFZ) and the Çardak–Sürgü Fault prior to the Kahramanmaraş earthquake doublet, using elastic strike-slip and dip-slip models combined with data assimilation. We find that the slip-deficit rates for the EAFZ are estimated at 2–8 mm/yr in the Anatolia-fixed frame and 0–11 mm/yr in the Arabia-fixed frame. For the Çardak–Sürgü Fault, slip-deficit rates are 1–4 mm/yr in the Anatolia-fixed frame and 2–11 mm/yr in the Arabia-fixed frame.
Dip-slip rates show wider variability. Along the EAFZ, they range from 0 to –12 mm/yr in the Anatolia-fixed frame and from –12 to 2 mm/yr in the Arabia-fixed frame. For the Çardak–Sürgü Fault, dip-slip rates range from –4 to 5.5 mm/yr in the Anatolia-fixed frame and from –3 to 10 mm/yr in the Arabia-fixed frame.
To further evaluate our estimates, we compared the interseismic slip deficit with coseismic slip distributions from the 2023 Kahramanmaraş earthquakes. The results indicate that the estimated slip deficit is broadly consistent with the observed coseismic slip for the Amanos and Çardak segments, but underestimates slip for the Pazarcık, Erkenek, and Göksun segments. A key limitation of the approach is that the locking depth remains poorly constrained in the joint inversion, particularly when using GNSS velocities in the Arabia-fixed frame rather than the Anatolia-fixed frame. In addition, the choice of kinematic representation for the joint fault mechanism exerts a strong influence on the results, leading to notable differences in the estimated slip deficit.

Composite materials are crucial for advancing sustainable engineering practices as they offer high strength-to-weight ratios, making them ideal for applications in aerospace, automotive and civil engineering. Meeting global sustainability goals demands a shift towards efficient materials, resulting in composites becoming increasingly significant. Besides difficulties in producing highly complex composites there are also challenges in computational methods to accurately predict the behavior of structures made from composite materials within a reasonable computational time frame. In most civil engineering practices, the Finite Element Analysis (FEA) is a useful method to derive stress-strain paths. However, these methods often rely on homogenized material assumptions which fail to capture complex microstructural stresses in high-performance composites. A more suitable method to predict the behavior of these materials under complex loading conditions is available using multi-scale modeling techniques. This approach is essential but computationally expensive, especially when high accuracy at the microscale is needed. Surrogate models based on a machine learning framework have been investigated in prior research. These models typically make predictions on homogenized average stress at macroscale rather than the maximum stress at microscale, which identifies as the research gap. The maximum microscopic stress has been recently used as a failure indicator. The main objective of this research is to design a surrogate model, based on the physically recurrent neural network (PRNN) designed in previous research, that can accurately predict the local maximum stress at the microscale in heterogeneous composite materials. Additionally, this research explores how knowledge from existing models that predict average stresses can be leveraged to enhance new models that predict local quantities at the microscale through transfer learning and multi-task learning techniques. Among the three developed models, two models using an encoder-decoder architecture performed poorly because the decoder averages the results instead of being able to find the limits. The decoder was removed in the third model and this yielded promising results. This model maintains high fidelity in stress predictions with lower computational costs because the number of trainable parameters in this model is limited. Transfer learning yielded faster convergence but did overall not improve significantly in terms of error. The decrease in computational time is very limited when transfer learning is applied over direct training as the models used in this thesis are small and do not need extensive datasets or epochs to converge. The multi-task approach achieved very promising results as it enables accurate predictions for average and maximum stresses in a single model. Cross validation showed strong model robustness and flexibility across unseen loading scenarios. Finally, a triple-task model showed that the PRNN is able to act as a modular framework as also the minimum microscopic stress can be accurately predicted without increasing the model and or training set size. The developed surrogate model shows that the architecture of the PRNN developed in previous research can be effectively modified to make predictions on other tasks. It provides a viable tool for simulating composite materials behavior at microscale.