!DOCTYPE html> SPE - Calgary Section
Webinar Download "Three Common Statistical Missteps We Make In Reservoir Characterization"

Webinar Download "Three Common Statistical Missteps We Make In Reservoir Characterization"

Data Analytics
←  SPE - Calgary Section

Data Analytics SIG:
Three Common Statistical Missteps We Make In Reservoir Characterization

Date: Thursday Oct 1, 2020

Speaker: Dr. Jerry Jensen, Research Engineer, Bureau of Economic Geology, University of Texas at Austin

SPE Members: Free, Non Members: $10

Recorded Webinar Access Instructions:
Once you purchase the member or non member rate you will receive a confirmation email, inside this email is a link to the pre-recorded session. If you do not receive this email within an hour of your purchase, please email calgarysection@spe.org.


Reservoir characterization analysis resulting from incorrect applications of statistics can be found in the literature, particularly in applications where integration of various disciplines is needed. Here, we look at three misapplications of ordinary least squares linear regression (LSLR) and show how they can lead to poor results and offer better alternatives, where available.  The issues are

  1. Application of algebra to an LSLR-derived model to reverse the roles of explanatory and response variables that may give biased predictions. In particular, we examine pore throat size equations (e.g., Winland’s and Pittman’s equations) and find that claims of over-predicted permeability may in part be due to statistical mistakes.
  2. Using a log-transformed variable in an LSLR model, de-transforming without accounting for the role of noise gives an equation which under-predicts the mean value. Several approaches exist to address this problem.
  3. Mis-application of R2 in three cases that lead to misleading results. For example, model fitting in decline curve analysis gives optimistic R2 values, as is also the case where a multimodal explanatory variable is present.

Using actual and synthetic datasets, we illustrate the effects that these errors have on analysis and some implications for using machine learning results.

Speaker Bio:

Dr. Jerry Jensen is a part-time research engineer at the Bureau of Economic Geology, University of Texas at Austin. From 2007 to 2018, he held the Schulich Chair in Geostatistics at the University of Calgary’s Department of Chemical and Petroleum Engineering.  Prior to 2007, Dr. Jensen held faculty positions at Texas A&M (1998-2007) and Heriot-Watt (1985-1997) Universities and worked as a field engineer for Services Techniques Schlumberger (1973-1977) and Gearhart Industries (1977-1983).

Dr. Jensen received a BSc in electrical engineering from the U. of Birmingham (UK) in 1973 and a PhD degree in petroleum engineering from the U. of Texas at Austin in 1986.  He is author or co-author of over 100 publications, including the books “Statistics for Petroleum Engineers and Geoscientists” (2000) and “Applied Reservoir Characterization (2014), both by Elsevier.  He has research and teaching interests in inter-well connectivity, petrophysical analysis of unconventional reservoirs, and strategic sampling for reservoir analysis and modeling. He was also an SPE distinguished lecturer in 2011-2012 on the topic of inter-well connectivity.

Don't miss out and sign up today!

Is your Membership up-to-date? Are you looking to become an SPE Member? Learn more here!