Skip to main content

Keynote speaker

Keynote speaker

Dr. Pierre Goovaerts

Chief Scientist BioMedware
Founder PGeostat LLC

Dr. Pierre Goovaerts is an internationally recognized geostatistician with over 30 years of experience developing advanced spatial and space-time statistical methods for environmental science, public health, and natural resource management. At BioMedware, he leads research and software development efforts, including VESTA, the company's next-generation platform integrating geostatistics, interactive visualization, and AI-assisted data analysis.

He is also the founder of PGeostat LLC, a consulting firm specializing in geostatistics, environmental risk assessment, and spatial data analytics. Through PGeostat, he has provided scientific expertise to government agencies, universities, law firms, and private companies on projects involving environmental contamination, groundwater and sediment remediation, exposure assessment, and uncertainty quantification.

Dr. Goovaerts has served as principal investigator or key contributor on numerous federally funded research projects, including NIH, NSF, and EPA initiatives. He is the author of the influential textbook Geostatistics for Natural Resources Evaluation, with over 26,000 citations and a Google Scholar h-index of 61.

His contributions have been recognized through the International Association for Mathematical Geosciences Distinguished Lecturer Award and the ISARA Founder's Award. He also serves as an editor of Mathematical Geosciences.

Session details: Vibe analytics: The future of geostatistics or a dangerous shortcut?

Recent advances in large language models have enabled vibe analytics, an emerging approach to data analysis through natural-language interactions. While promising greater accessibility, this paradigm raises important concerns about scientific rigor, reproducibility, and domain expertise in geostatistics.

Using daily ozone concentrations from monitoring stations in the Los Angeles region, this presentation evaluates the capabilities and limitations of ChatGPT for exploring space-time variability, identifying temporal patterns, investigating geographical influences, and modeling spatial and temporal dependence. The findings highlight opportunities for exploratory analysis and hypothesis generation, alongside challenges involving inconsistent results, unreliable geographical information, limited spatial interactivity, and insufficient methodological transparency.

The presentation introduces VESTA (Visualization and Exploratory Space-Time data Analysis), a geospatial analytical workbench integrating interactive visualization, statistical modeling, and an LLM-powered advisor. By guiding users through scientifically defensible analytical workflows, VESTA illustrates how conversational interfaces can complement specialized geostatistical tools rather than replace domain expertise.

The central argument is that the future of vibe analytics in geostatistics depends on combining conversational accessibility with rigorous methods, explicit uncertainty assessment, and transparent, reproducible analytical workflows.

Our site uses cookies

We use these to improve your browser experience. By continuing to use the website you agree to the use of cookies.