You are here

10 Questions for a Computational Scientist: Kerstin Kleese-Van Dam

June 9, 2011 - 4:35pm

Addthis

Kerstin Kleese-Van Dam

Kerstin Kleese-Van Dam

Meet Kerstin Kleese-Van Dam. At Pacific Northwest National Lab, she’s a master of computers and data – covering a wide span of projects from genomic sciences and climate change to nanometer-scale imaging and power grids. She recently spent some time to give us the download on her many initiatives and why her interests in Information Technology (IT) eventually led her to a scientific career.

Question: What sparked your interest to pursue a career in science?

Kerstin Kleese-Van Dam: I have always had an interest in IT, but wanted to do something that could have a real positive impact on people’s lives, and science gave me the opportunity to contribute to addressing some of society’s big challenges – climate change, environmental remediation, sustainable clean energy and secure power. It also provides for a vibrant and exciting working environment.

Q: At PNNL you wear two hats as the associate director for the Computational Sciences and Mathematical Division and the technical lead for the Scientific Data Management Group. What led you to these fields?

KK: For many years now it has been my passion to make the scientific data we produce at our computational, observational and experimental facilities more accessible to the scientific community beyond the original data creator and to speed up scientific progress through more complete information, helping to bridge the barriers between different levels of theory and different scientific domains. My work as lead for scientific data management groups has allowed me to develop the necessary methodologies and tools to bring us closer to this goal. Moving into a more senior leadership position, such as the associate director position, has provided me with a much better overview of the landscape in which our work and contributions will have to fit. More importantly, it has given me an opportunity to support the development of our future strategies and their implementation, to help to make our combined visions reality.

Q: What projects are you working on right now? What do you hope they will lead to?

KK: We have just successfully finished the design and a prototype implementation of our concept for the Department of Energy Systems Biology Knowledgebase (Kbase). As we envision it, Kbase will greatly accelerate the systems biology research underlying the Department of Energy’s Genomic Science Program by providing a “frictionless scientific ecosystem” for systems biology, supporting broad community collaboration, accelerating innovation through dissemination of new approaches and enhancing data integration across currently unconnected data repositories. Community-wide shared tools and workflows for integrating, analyzing and visualizing the multiple types of data will be part of the core capability of Kbase as we envisage it. Central to our vision is the application of semantic technologies in this environment to ease discovery, access and integration.

Here at PNNL we have developed together with our collaborators a vision for the future of data intensive science, which we see as a key enabler for faster scientific progress. The Department of Energy Systems Biology Knowledgebase and many of my other projects will bring us a little closer to this frictionless scientific ecosystem for all sciences.

Q: Do you have any advice for students interested in becoming scientists?

KK: If you want a versatile and exciting working environment that you can help shape, science is certainly the area for you. Do get involved as early as possible. Look for internships - PNNL has a comprehensive program of opportunities  – they are a great way to find out what goes on at the leading edge of research, what you like and even more importantly which field might not interest you as much after all. I made heavy use of internships during my college and university years and found them particularly useful to see what was used in practice and how. They gave much more meaning and interest to my studies, seeing how sometimes rather dry and theoretical concepts were of great use in the scientific world. A nice bonus is that you can try out employers and see which one suits you best, when you get your first job, you might already have friends and friendly faces around you.

Q: Your expertise in computing and scientific data management has led you to positions in a number of different countries and across private and public sectors. How do international and cross-sector collaborations impact your field?

KK: At the most fundamental level the broader your base of collaborators, the wider your pool of knowledge to learn about possible new ideas and trends, vetted by colleagues you trust. Early knowledge of such developments is what keeps us competitive and relevant, so I am always actively working with my network through means such as LinkedIn, Facebook, joint conference participation and regular email contact.  Furthermore the knowledge of new practices in other domains, correlated with your own work, can lead to new hypothesis and drive your own research into new exciting areas. On a more practical level, ideas, methods and technologies transfer often very well from one science domain to another, allowing you to leverage your existing work to enable progress in many different communities.

Q: The breadth of projects you’re working on is incredible. Can you tell us a little bit about integrated Regional Earth Systems Modeling and the Chemical Imaging Initiative?

KK: The integrated Regional Earth Systems Modeling Initiative (iRESM) is one of my most exciting projects. It will be developing a software capability to help decision and policy makers understand how society and the environment might respond to the impacts of climate change over time. Policy and decision making, in response to climate change, will require both economic and environmental tradeoffs. Decisions about allocating scarce water across competing municipal, agricultural, and ecosystem demands is just one of the challenges ahead, along with decisions regarding competing land use priorities such as biofuels, food, and species habitat.

The fundamental goal of iRESM is the critical analyses of the tradeoffs and consequences of decision and policy making on integrated human and environmental systems, combining the different scientific processes, bridging different temporal and geographic scales and resolving the semantic differences between them. iRESM will work with the climate, socio-economic, crop and energy modeling communities to develop an initial modeling system. The computational challenges faced by the project go hand in hand with the scientific ones – as researchers from different domains start to bridge the gaps, an adaptive computational framework needs to be established that facilitates integration and transition across different geographical, time, thematic and semantic divides.

The Chemical Imaging Initiative (CII) is another very interesting project. CII will develop multi-modal in situ chemical imaging and analysis capabilities, delivering a suite of new state-of-the-art tools that will address a multitude of scientific and technical challenges with nanometer resolution and element specificity. This capability will allow researchers to apply the molecular-level chemical and structural information afforded by these tools to large-scale scientific challenges.

To achieve the initiative goal, CII will develop next-generation coupled optical, electron, mass spectrometric, and scanning probe capabilities and suitably integrate these imaging technologies with light-source-based methods to understand chemical, material, and biological transformations. To support these technical developments the IT group is developing a framework that will not only allow the analysis and integration of results across different imaging technologies, but also will aim to achieve real time or near real time analysis capabilities to allow researchers to influence their experiments as they are progressing.

Q: You also work on power grid data models. What will this project achieve?

KK: The future power grid will require bidirectional, real-time data flow to identify and respond to changes in demand, including quick reaction to extreme events, operational monitoring, daily and medium term operational planning and long term facility planning. Unlike the current grid, with a relatively small number of power sources, the future grid will have thousands of potential power sources including hybrids, wind sources and solar panels, in addition to traditional sub stations and power plants. Each of these sources will operate independently and thus respond differently to fluctuations in price, resources and operational circumstances. Consumption devices will also be smarter, using real-time pricing to make decisions about their level and timing of consumption – for example, an air conditioner that turns on depending not only on temperature, but also the current price of electricity. Each of these production and consumption units, as well as the connecting power infrastructure will be equipped with smart sensors that report on status and deliver operational information and are also receptors of information and instructions (i.e. reduce demand/production).

Within this initiative we are focusing on developing a multi resolution data model and directed real time data reduction, reconstruction and aggregation capabilities that will enable real-time situational awareness of the future power grid throughout its various levels. Key in this work is identifying critical events and notifying all relevant players on the grid about such events.

Q: What can you never start a day at the lab without?

KK: My tea, a habit I picked up in Britain.

What is your favorite tool in the lab?

KK: CAT – our Collaborative Analytical Toolbox.

Q: What do you enjoy doing in your free time?

KK: Hiking with my family, reading good mystery books.

Addthis