Our research group, Hydro AI group, led by Dr. Hyunglok Kim, explores how land surface and the terrestrial water cycle interact at regional to global scales and their response to future climate change. We aim to understand how water movement through the land surface translates to larger-scale behavior and how we can use remote sensing, land surface models, and machine learning to improve natural disasters and water resource prediction accuracy. Our research focuses on developing new remote sensing datasets, surface models, and validating them, particularly in the context of microwave proxies for terrestrial water content. We're particularly interested in exploring the impact of these interactions on the Earth's water balance under future climate scenarios.
Have a look at the process of general research steps.
Obtaining or simulating data from satellites and land surface models involves the acquisition of remote sensing observations and the use of numerical simulations to represent the Earth's surface processes. Satellite data provides extensive spatial coverage and continuous measurements of key hydrological and Earth science variables, while land surface models capture the complex interactions between the atmosphere, vegetation, and soil. Combining these data sources enables researchers to better understand natural processes, monitor large-scale environmental changes, and improve the accuracy of forecasts and predictions in hydrology and Earth science.
Data wrangling, also known as data munging or preprocessing, is the process of transforming and mapping raw data into a more structured, usable format for analysis. This often involves cleaning, aggregating, and reformatting data to address inconsistencies, missing values, and other issues that may impact the quality of insights derived from the data. Data mining, on the other hand, is the practice of discovering patterns, trends, and relationships in large datasets using various analytical techniques, such as machine learning, statistical analysis, and database querying. The goal of data mining is to extract valuable, previously unknown information from the data and use it to inform decision-making, predict outcomes, or gain a deeper understanding of a hydrological domain.
Modeling and analyzing data with machine learning methods involve using algorithms to learn from the input data, identify patterns, and make predictions or decisions without being explicitly programmed. In the context of data analysis, machine learning techniques help to uncover hidden patterns and relationships in the data that might be difficult to discover using traditional statistical methods, to build predictive models that can generalize to new, unseen data, enabling accurate forecasts or classifications.Handle large, complex, and high-dimensional datasets, which are common in the hydrological research field. This process typically involves selecting an appropriate machine learning algorithm (e.g., linear regression, decision trees, or neural networks), training the model using a subset of the data, and evaluating its performance on a separate test set. The ultimate goal is to create a model that can provide accurate and actionable insights, enhancing understanding, and informing decision-making in hydrological research domains.
Attending international conferences and presenting research to others is important because it facilitates knowledge sharing and collaboration among researchers, advancing the field by exposing attendees to new ideas and diverse perspectives. Furthermore, presenting research helps us refine our communication skills, enabling us to effectively convey complex ideas to both technical and non-technical audiences. Overall, participation in conferences fosters personal and professional growth, while contributing to the broader scientific community's progress.
Sharing knowledge through peer-reviewed journals and data is essential in science as it fosters the open exchange of ideas, ensures the rigor and credibility of research findings through critical evaluation, and facilitates the reproducibility and validation of scientific results, ultimately advancing our collective understanding and driving innovation in the field.
Deploying machine learning models on various platforms, such as cloud computing services (e.g., AWS, Google Cloud) is important in science because it facilitates scalable, efficient data processing and real-time decision-making, ultimately enhancing the accessibility and impact of scientific insights across diverse applications and environments.
We use GPU servers and High-Performance Computing (HPC) systems to enable faster processing and analysis of large, complex datasets, significantly reducing computation time and facilitating advanced research in various fields.
If you have a keen interest in the intersection of climate change and its impact on hydrological research fields, I encourage you to consider pursuing a Master's, PhD, or postdoctoral position. By delving deeper into this critical area of study, you can play an essential role in addressing the world's most pressing environmental challenges and help safeguard our water resources, ecosystems, and communities. Your dedication and expertise can significantly contribute to the development of sustainable solutions and innovative approaches to hydrological research. Embark on this exciting journey and become part of the passionate community of scientists working towards a more resilient and environmentally responsible future.