The University of British Columbia

My research deals with the development and application of machine learning and statistical models to climate and weather prediction. Some current projects involve:

• conditional density estimation networks for multi-site climate downscaling
• automated methods for synoptic map-pattern classification and weather typing
• assessing predictive uncertainty in weather and climate forecasting
• stochastic simulation methods for weather elements at multiple stations
• nonstationary extreme value analysis in hydroclimatology
• assessing impacts of climate variability and change on water resources
• climate prediction on seasonal to decadal time scales
  

 

---

Committees

See the AMS Committee on Artificial Intelligence Applications to Environmental Science page for general information on the use of machine learning methods in environmental prediction.

I am on the Editorial Boards of a few journals. Consider submitting an article to Computers & Geosciences, PLoS ONE, or ISRN Meteorology.

---

Software

• R package for the quantile regression neural network (qrnn)
• R package for Conditional Density Estimation Network Creation & Evaluation (CaDENCE) [*test version*]
• R package for the monotone multi-layer perceptron neural network (monmlp)
• R package for the Generalized Extreme Value conditional density estimation network (GEVcdn)
• Robust nonlinear canonical correlation analysis (R-NLCCA) has been incorporated into Prof. Hsieh's NeuMATSA MATLAB Toolbox

---

Journal Publications

In press

• Bürger, G., T.Q. Murdock, A.T. Werner, S.R. Sobie, and A.J. Cannon, in press. Downscaling extremes - an intercomparison of multiple statistical methods for present climate. Journal of Climate. doi:10.1175/JCLI-D-11-00408.1
• Cannon, A.J., D. Neilsen, and W.G. Taylor, in press. Lapse rate adjustments of gridded surface temperature normals in an area of complex terrain: atmospheric reanalysis versus statistical up-sampling. Atmosphere-Ocean. doi:10.1080/07055900.2011.649035
• Cannon, A.J., in press. Neural networks for probabilistic environmental prediction: Conditional Density Estimation Network Creation & Evaluation (CaDENCE) in R. Computers & Geosciences. doi:10.1016/j.cageo.2011.08.023
• Cannon, A.J., in press. Semi-supervised multivariate regression trees: putting the "circulation" back into a "circulation-to-environment" synoptic classifier. International Journal of Climatology. doi:10.1002/joc.2417
• Pellatt, M.G., S. Goring, K.M. Bodtker, and A.J. Cannon, in press. Using a down-scaled bioclimate envelope model to determine long-term temporal connectivity of Garry oak (Quercus garryana) habitat in western North America: implications for protected area planning . Environmental Management.

• Cannon, A.J., 2012. Regression-guided clustering: a semisupervised method for circulation-to-environment synoptic classification. Journal of Applied Meteorology and Climatology, 51(2): 185-190 doi:10.1175/JAMC-D-11-0155.1 
• Cannon, A.J., 2012. Köppen versus the computer: comparing Köppen-Geiger and multivariate regression tree climate classifications in terms of climate homogeneity, Hydrology and Earth System Sciences, 16: 217-229. doi:10.5194/hess-16-217-2012
• Cohen, S., S. Sheppard, A. Shaw, D. Flanders, S. Burch, B. Taylor, D. Hutchinson, A.J. Cannon, S. Hamilton, B. Burton, and J. Carmichael, 2012. Downscaling and visioning of mountain snow packs and other climate change implications in North Vancouver, British Columbia. Mitigation and Adaptation Strategies for Global Change, 17(1): 25-49. doi:10.1007/s11027-011-9307-9
  
• Rasouli, K., W.W. Hsieh, and A.J. Cannon, 2012. Daily streamflow forecasting by machine learning methods with weather and climate inputs. Journal of Hydrology, 414-415: 284-293. doi:10.1016/j.jhydrol.2011.10.039

• Jenkner, J., W.W. Hsieh, and A.J. Cannon, 2011. Seasonal modulations of the active MJO cycle characterized by nonlinear principal component analysis. Monthly Weather Review, 139(7): 2259-2275. doi:10.1175/2010MWR3562.1
• Cannon, A.J., 2011. Quantile regression neural networks: implementation in R and application to precipitation downscaling. Computers & Geosciences, 37: 1277-1284, doi:10.1016/j.cageo.2010.07.005
• Cannon, A.J., 2011. GEVcdn: an R package for nonstationary extreme value analysis by generalized extreme value conditional density estimation network. Computers & Geosciences, 37:1532-1533. doi:10.1016/j.cageo.2011.03.005

• Allen, D.M., A.J. Cannon, M.W. Toews, and J. Scibek, 2010. Variability in simulated recharge using different GCMs, Water Resources Research, 46: W00F03, doi:10.1029/2009WR008932
• Cannon, A.J., 2010. A flexible nonlinear modelling framework for nonstationary generalized extreme value analysis in hydroclimatology. Hydrological Processes, 24: 673-685.
• Quamme, H.A., A.J. Cannon, D. Neilsen, J.M. Caprio, and W.G. Taylor, 2010. The potential impact of climate change on the occurrence of winter freeze events in six fruit crops grown in the Okanagan Valley. Canadian Journal of Plant Science, 90(1): 85-93.

• Cannon, A.J., 2009. Negative ridge regression parameters for improving the covariance structure of multivariate linear downscaling models. International Journal of Climatology, 29:761-769.
• Hao, P., A.J. Cannon, P.H.  Whitfield, and H. Lu, 2009. Pentad average temperature changes of Inner Mongolia during recent 40 years. Journal of Applied Meteorological Science, 20(4): 443-450.

• Cannon, A.J., 2008. Probabilistic multi-site precipitation downscaling by an expanded Bernoulli-gamma density network. Journal of Hydrometeorology, 9(6):1284-1300.  [revised version with errata fixed]
• Cannon, A.J. and W.W. Hsieh, 2008. Robust nonlinear canonical correlation analysis: Application to seasonal climate forecasting. Nonlinear Processes in Geophysics, 15: 221-232.
• Stahl, K., R.D. Moore, J.M. Shea, D. Hutchinson, and A.J. Cannon, 2008. Coupled modelling of glacier and streamflow response to future climate scenarios. Water Resources Research, 44: W02422, doi:10.1029/2007WR005956
• Hsieh, W.W. and A.J. Cannon, 2008. Towards robust nonlinear multivariate analysis by neural network methods. Lecture Notes in Earth Sciences, 12:97-124. doi:10.1007/978-3-540-78938-3_6

• Cannon, A.J., 2007. Nonlinear analog predictor analysis: a coupled neural network/analog model for climate downscaling. Neural Networks, 20(4): 444-453.
• Scibek, J., D.M. Allen, A.J. Cannon, and P.H. Whitfield, 2007. Groundwater-surface water interaction under scenarios of climate change using a high-resolution transient groundwater model. Journal of Hydrology, 333: 165-181.
• Song, L., A.J. Cannon, and P.H. Whitfield, 2007. Changes in seasonal patterns of temperature and precipitation in China during 1971-2000. Advances in Atmospheric Science, 24(3): 459-473.

• Cannon, A.J., 2006. Nonlinear principal predictor analysis: application to the Lorenz system. Journal of Climate, 19(4): 579-589.
• Wang, J.Y., P.H. Whitfield, and A.J. Cannon, 2006. Influence of Pacific climate patterns on low-flows in British Columbia and Yukon, Canada. Canadian Water Resources Journal, 31(1): 25-40.
• Hall, A.W., P.H. Whitfield, and A.J. Cannon, 2006. Recent variations in temperature, precipitation, and streamflow in the Rio Grande and Pecos River Basins of New Mexico and Colorado. Reviews in Fisheries Science, 14(1-2): 51-78.

• Cannon, A.J., 2005. Defining climatological seasons using radially constrained clustering. Geophysical Research Letters, 32: L14706, doi:10.1029/2005GL023410

• Whitfield, P.H., A.W. Hall, and A.J. Cannon, 2004. Changes in the seasonal cycle in the circumpolar Arctic, 1976-1995: Temperature and precipitation. Arctic, 57(1): 80-93.

• Whitfield, P.H., J.Y. Wang, and A.J. Cannon, 2003. Modelling future streamflow extremes - Floods and low flows in Georgia Basin, British Columbia. Canadian Water Resources Journal, 28(4):633-656.
• Whitfield, P.H., A.J. Cannon, J.Y. Wang, and C.J. Reynolds, 2003. Modelling streamflows in present and future climates - Examples from rainfall/snowmelt streams in coastal British Columbia. Hydrological Science & Technology, 19(1-4): 41-56.

• Whitfield, P.H., C.J. Reynolds, and A.J. Cannon, 2002. Modelling streamflow in present and future climates - Examples from the Georgia Basin, British Columbia. Canadian Water Resources Journal, 27(4): 427-456.
• Cannon, A.J., P.H. Whitfield, and E.R. Lord, 2002. Synoptic map-pattern classification using recursive partitioning and principal component analysis. Monthly Weather Review, 130(5): 1187-1206.
• Cannon, A.J. and P.H. Whitfield, 2002. Downscaling recent streamflow conditions in British Columbia, Canada using ensemble neural network models. Journal of Hydrology, 259: 136-151.
• Cannon, A.J. and I.G. McKendry, 2002. A graphical sensitivity analysis for interpreting statistical climate models: Application to Indian monsoon rainfall prediction by artificial neural networks and multiple linear regression models. International Journal of Climatology, 22:1687-1708.
• Whitfield, P.H., K. Bodtker, and A.J. Cannon, 2002. Recent variations in seasonality of temperature and precipitation in Canada - 1976-1995. International Journal of Climatology, 22: 1617-1644.

• Cannon, A.J. and P.H. Whitfield, 2001. Modeling transient pH depressions in coastal streams of British Columbia using neural networks. Journal of the American Water Resources Association, 37(1): 73-89.

• Whitfield, P.H. and A.J. Cannon, 2000. Recent variations in climate and hydrology in Canada. Canadian Water Resources Journal, 25(1): 19-65.
• Whitfield, P.H. and A.J. Cannon, 2000. Polar plotting of seasonal hydrologic and climatic data. Northwest Science, 74(1):76-80.
• Cannon, A.J. and E.R. Lord, 2000. Forecasting summertime surface level ozone concentrations in the lower Fraser Valley of British Columbia: An ensemble neural network approach. Journal of the Air & Waste Management Association, 50: 322-339.

• Cannon, A.J. and I.G. McKendry, 1999. Forecasting all-India summer monsoon rainfall using regional circulation principal components: A comparison between neural network and multiple regression models. International Journal of Climatology, 19(14): 1561-1578.

 

---

Miscellaneous

View a movie comparing the application of nonlinear and linear variants of principal predictor analysis to climate variability in the tropical Pacific Ocean. The spatial asymmetry between warm phase and cold phase sea surface temperature anomalies is represented more realistically by the nonlinear model.