Streamflow and Drought Predictions over Ungaged Regions using Deep and Transfer Learning Approaches

January 31, 2024

Leveraging remote sensing data and advanced machine learning techniques significantly improves streamflow and flash drought predictions in ungaged regions by capturing complex hydrological processes and environmental factors.

Remotely sensed data can be highly beneficial in predicting streamflow and flash droughts over ungaged regions by providing spatially extensive and continuous observations of key hydrological variables, such as precipitation, soil moisture, land surface temperature, and vegetation indices. When combined with deep and transfer learning approaches, these data sources enable the development of robust predictive models that can capture complex relationships between hydrological processes and environmental factors, ultimately improving our ability to forecast streamflow and flash drought events in regions lacking traditional ground-based monitoring networks.

For example, transfer learning is especially useful because it allows models trained on data-rich regions to be adapted for ungaged areas, effectively utilizing existing knowledge and reducing the need for extensive new data collection, ultimately enhancing the accuracy and efficiency of hydrological predictions in data-scarce locations.

The concept of the transfer learning. blue: data rich regions / red: data-scarce regions

What we aim to achieve

Through this research, we aim to develop more accurate and efficient streamflow and flash drought prediction models for ungaged regions by utilizing remotely sensed data and advanced machine learning techniques, such as deep and transfer learning. By achieving this goal, we hope to improve water resource management, inform disaster mitigation strategies, and enhance our understanding of hydrological processes in data-scarce areas.n lead to a deeper understanding of hydrological processes, helping researchers better comprehend the complex interactions between precipitation, evaporation, and runoff.

Data and analytic skills we use for this project

For this research, the following data analytic skills are essential:

  1. Remote sensing expertise: Knowledge of various remote sensing platforms, data types, and processing techniques is crucial for obtaining and interpreting the relevant hydrological variables from satellite imagery.
  2. Machine learning proficiency: Familiarity with deep learning algorithms, such as convolutional neural networks (CNNs) and recurrent neural networks (RNNs), and transfer learning techniques is necessary for developing robust predictive models.
  3. Hydrological domain knowledge: A strong understanding of hydrological processes, including streamflow dynamics and drought mechanisms, is essential for interpreting results and connecting data insights to real-world applications.
  4. Programming and software skills: Proficiency in programming languages, such as Python or R, and familiarity with relevant machine learning libraries, like TensorFlow or PyTorch, is necessary for implementing and automating various stages of the analytical process.

Our supportive academic environment is here to help you learn and develop the necessary skills along the way.

If you are interested in applying and learning AI approaches in Earth science, please contact me. The following are a few example research topics we can explore together.
  1. Estimating future agricultural water demand under climate change: Use deep learning and transfer learning approaches to integrate remotely sensed data, such as evapotranspiration rates and vegetation indices, with climate projections to estimate future water demand for agriculture and inform adaptive water management strategies.
  2. Predicting coastal vulnerability to sea-level rise and storm surges: Employ deep learning and transfer learning methods to analyze remote sensing data, including shoreline changes, coastal topography, and land use, to assess the vulnerability of coastal regions to the impacts of climate change, such as sea-level rise and increased storm surges.
  3. Monitoring and predicting ecosystem shifts under climate change: Utilize deep learning and transfer learning techniques to combine remotely sensed data, such as land cover and temperature, with climate change projections to predict shifts in ecosystem boundaries and assess potential impacts on biodiversity and ecosystem services.
  4. Assessing the risk of wildfires under climate change: Apply deep learning and transfer learning approaches to integrate remote sensing data, including vegetation moisture content and fire history, with climate projections to identify regions at increased risk of wildfires due to changing climate conditions and inform fire management strategies.

Read other projects

Harnessing Deep Learning to Predict and Decode the Mysteries of Flash Droughts (GAN/SHAP/3D-CNN with Transfer Learning)

The application of deep learning in predicting flash droughts offers a transformative approach to understanding and anticipating these rapid-onset events, significantly enhancing preparedness and response strategies. By unraveling the complex mechanisms behind flash droughts, this project aims to provide precise, timely forecasts, thereby mitigating the severe agricultural, ecological, and socioeconomic impacts associated with these phenomena.

Read this project
Streamflow and Drought Predictions over Ungaged Regions using Deep and Transfer Learning Approaches

Streamflow and flash drought predictions are essential for managing water resources and mitigating potential disasters in ungaged regions. With remotely-sensed data, deep and transfer learning approaches provide powerful tools to analyze complex hydrological data, enabling more accurate predictions and better decision-making in these areas.

Read this project
Applications of Bayesian Machine Learning in Big Data in Earth Science

Bayesian methods help us improve our guesses by using new information. In Earth science, these methods are applied to big data to better understand our planet. This approach is useful for predicting things like natural disaster patterns and climate changes. By continuously updating our knowledge with new data, we can make more accurate predictions and decisions in Earth science.

Read this project
Water Balance Budgeting with Bayesian Machine Learning

The water balance equation in Earth science, P = E + R + etc, describes the relationship between precipitation (P), evaporation (E), runoff (R), and etc (e.g., soil moisture, ground water) in a given area. Bayesian inference can be applied to solve this equation by incorporating prior knowledge and updating the probability distributions of the variables based on new data, ultimately improving water resource management and prediction.

Read this project
Integrating Earth Science and Engineering for Climate Resilience: Innovative Approaches to Infrastructure and Societal Justice

Earth science informs infrastructure development by providing insights into site suitability, resource management, and sustainable design, enhancing the resilience and long-term viability of projects. It also plays a crucial role in addressing societal justice related to climate change by helping identify vulnerable communities and develop mitigation strategies, ensuring equitable access to resources and protection from environmental hazards.

Read this project
Enhancing Earth Science Predictions through Advanced Data Assimilation Techniques

Data assimilation is vital in earth science as it integrates diverse observations and model simulations, improving the accuracy of forecasts and predictions. This process enhances our understanding of complex Earth systems, enabling better decision-making for environmental management and climate adaptation.

Read this project
Floods and Droughts Predictions using Machine Learning Approaches

Satellite data and machine learning transformed Earth science by predicting and monitoring natural disasters. This combination delivers precise and timely predictions, crucial for mitigating the impacts of events like floods and droughts.

Read this project
Data Error Characterizations

Characterizing the error of satellite data and land surface models is vital in Earth science, as it ensures the accuracy and reliability of information used for monitoring and predicting environmental phenomena. By understanding these errors, scientists can refine data interpretation, enhance models, and ultimately make better-informed decisions about the Earth's complex systems.

Read this project
Developing Algorithms to Improve the Temporal Sampling of Satellite Data

Enhancing the temporal repeat of satellite data for obtaining soil moisture information is a vital research area due to its implications for agriculture, water resource management, climate change research, and ecosystem health. It helps in making informed decisions, increasing productivity, and reducing the impact of natural disasters, as well as contributing to our understanding of the global climate system.

Read this project
Exploring the Impact of Human Activities on the Subdaily Global Terrestrial Water Cycle

Humans have been modifying the Earth's surface for thousands of years, with practices like clearing forests for agriculture and creating uniform land covers. But how do these changes impact the subdaily global terrestrial water cycle? That's the question a project aims to answer.

Read this project
Satellite Image Disaggregation with Machine Learning

Microwave soil moisture data is critical for agriculture, weather, and climate modeling, but has low spatial resolution. Disaggregation via machine learning can improve resolution, offering detailed local soil moisture data. Machine learning can handle complex relationships between microwave signals and soil moisture.

Read this project