This is a past event. Registration is closed. View other South African Statistical Association events.
Save the date
Save the dateApril 22, 202213:00 - 14:00 GMT+2
Organizer
Organizer
Contact Person: SU Statistics Department
Email: krugere@sun.ac.za
Speaker
Speaker
Prof. Gerard Heuvelink
Professor Gerard Heuvelink holds a dual appointment as senior researcher at ISRIC – World Soil Information and as associate professor at Wageningen University. As of 1 June 2017, Professor Heuvelink was appointed as special professor in Pedometrics and Digital Soil Mapping, funded by ISRIC - World Soil Information and positioned within the Soil Geography and Landscape chair group of Wageningen University. The professorship focuses on the development and application of mathematical and statistical techniques to analyse and model spatial and temporal variability in soil. This is achieved by combining these techniques with knowledge of soil processes and patterns. Professor Heuvelink is the recipient of the Richard Webster Medal from the Pedometrics Commission of the International Union of Soil Science and of the Peter Burrough medal of the International Spatial Accuracy Research Organisation. He is also the deputy editor of the European Journal of Soil Science and editorial board member of five more international scientific journals.
Abstract
Abstract
Spatially resolved estimates of change in soil organic carbon (SOC) stocks are necessary for supporting national and international policies aimed at achieving land degradation neutrality and climate change mitigation. In this work we report on the development, implementation and application of a data-driven, statistical method for mapping SOC stocks in space and time, using Argentina as a pilot. We used Quantile Regression Forest machine-learning to predict annual SOC stock at 0-30 cm depth at 250 m resolution for Argentina between 1982 and 2017. The model was calibrated using over 5,000 SOC stock values from the 36-year time period and 35 environmental covariates. We pre-processed NDVI dynamic covariates using a temporal low-pass filter to allow the SOC stock for a given year to depend on the NDVI of the current as well as preceding years. Predictions had modest temporal variation with an average decrease for the entire country from 2.55 kg C m‑2 to 2.48 kg C m‑2 over the 36-year period (equivalent to a decline of 211 Tg C, 3.0% of the total 0‑30 cm SOC stock in Argentina). The Pampa region had a larger estimated SOC stock decrease from 4.62 kg C m‑2 to 4.34 kg C m‑2 (5.9%) during the same period. For the 2001-2015 period, predicted temporal variation was 7-fold larger than that obtained using the Tier 1 approach of the IPCC and UNCCD. Prediction uncertainties turned out to be substantial, mainly due to the limited number and poor spatial and temporal distribution of the calibration data, and the limited explanatory power of the covariates. Cross-validation confirmed that SOC stock prediction accuracy was limited, with a Mean Error of 0.03 kg C m-2 and a Root Mean Squared Error of 2.04 kg C m-2. In spite of the large uncertainties, this work showed that machine learning methods can be used for space-time SOC mapping and may yield valuable information to land managers and policy makers, provided that SOC observation density in space and time is sufficiently large.
Tickets
Tickets
Spatial Statistics interest group hybrid event hosted by the Department of Statistics and Actuarial Science at Stellenbosch University.
Due to limited seating, in-person attendance should be confirmed by Wednesday, 13 April (17:00). Send an email to krugere@sun.ac.za to confirm your seat. In-person venue: Corner of Victoria and Bosman Streets, Van der Sterr Building, Room 2048. Refreshments to be served for in-person attendees after the seminar.
Online Attendance
By registering here, you will receive a link to join the seminar via MS Teams.
If you would like to attend in-person, please RSVP to the Department of Statistics (krugere@sun.ac.za) by Wednesday, 13 April 2022.
Member Price Complimentary
Sponsors and Partners
Sponsors and Partners
