Past Meetup

Modeling the landscape: Investigating soil genesis and geography with R

This Meetup is past

68 people went


NOTE: RSVPs required by 4:30PM Friday so we can send the attendee list to our hosts, Google. Make sure your RSVP includes your full name to ensure admission. Please be sure to arrive between 6:30 and 7:00 to be escorted through Security. Thanks!

Our February meeting will feature a talk from Dylan Beaudette with a perspective on using R for soil science.


6:30 - 7:00 Networking and refreshments (with thanks to Google)

7:00 - 7:10 Announcements, etc.

7:10 - 8:00 Dylan Beaudette, Modeling the landscape

8:00 - 8:30 Discussion, wrap-up.

8:30: Close

Here are the details of Dylan's talk:

Modeling the landscape: How R has given soil scientists a powerful new framework for the investigation of soil genesis and geography.

Soils support a wide range of natural ecosystems, agricultural production, industrial processes, and the largest terrestrial carbon pool. The misuse of the soil resource is one of the strongest predictors of past civilizations' demise. Informed land-use decisions depend on a spatially detailed inventory of the soil resource. A staggering quantity of soils information has been collected to support soil survey operations, natural resource inventories, and research over the last 100 years. Integrated analysis of large soil profile collections, however, is often hampered by the high dimensionality of these data, changes in soil classification over time, and difficulties associated with processing horizon data that vary widely in depth and thickness. New users of soil survey information (hydrologic, ecologic, and atmospheric modelers) require soil property information at several levels of generalization, from landscape to regional scales. Clearly new quantitative approaches for processing, aggregating, and classifying large collections of soil information are needed in order to provide a more robust description of soil properties and their variability-- across multiple scales. R provides an ideal framework for implementing this goal: from data collection, management, and aggregation; to model building, prediction, and mapping. Once mastered, the R language (coupled with numerous user-contributed packages) equips the soil scientist with a vocabulary for quantitatively describing, testing, and visualizing postulated soil-landscape relationships. This presentation will demonstrate several case studies in which R has been used within the field of soil science, and what the future might hold for adoption of R within the USDA-NRCS.

About the speaker:

Dylan Beaudette ( is a finishing a Ph.D. in Soils and Biogeochemistry at UC Davis. He is interested in applying numerical methods to soil survey, soil classification, and towards extending our understanding of soil genesis. Dylan is currently working on several studies related to scaling soil survey information down to the landscape-scale, and up to the regional scale. These experiments involve (sensor-calibrated) quantification of soil climate, modeling of geomorphic surfaces, and estimation of within map unit variability. Open source software is used in all of his research, data management, analysis, and presentation tasks.