Web landshape.org

First Time At Niche Modeling?

This is a blog on the power of numeracy. My first book — Niche Modeling — is now in print.

The first six chapters are tutorial topics in R programming and theoretical topics in niche modeling: functions, data, spatial, topology, environmental data collections, and examples. The last six chapters are about using niche modeling to detect errors: bias, autocorrelation, non-linearity, long term persistence, circularity and fraud - useful information for any biological modeler.

February 29, 2008

Results Management

Filed under: Uncategorized, Statistics, Finance, Climate Change — admin @ 12:58 am

Results management can be defined as interventions used to change the actual or interpretation of a result. The most recent example in the financial area is Enron, where revenue numbers were subject to upwards revision via dubious accounting interventions.

Results management is distinct from ‘results-based management’ – a legitimate management approach focusing on achieving outcomes, implementing performance measurement, learning and changing, and reporting performance.

Luboš Motl of The Reference Frame identified a possible example of results management in Science in his article Borehole climate reconstructions & hockey stick revolution in 1998. By way of background, Boreholes were identified as decoding past temperatures by examining the temperatures profiles down the hole. Using this method the authors published the success of this result, and summarized the scientific finding that temperatures were higher than the present during a number of periods in the past. (more…)

February 27, 2008

Global Temperatures 2008

Filed under: Uncategorized, Science, Climate Change — admin @ 8:50 pm

Predicting global temperatures seems to be entering general awareness as a worthwhile exercise. As I have published about recently, I think climate models are inadequately validated, confidence in the skill of models to forecast global warming is vastly exaggerated, and current skill is not enough to serve useful purposes. I thought I would tabulate some of the various predictions as I come across them. This is a fair test, as the future is unknown, and at the end of the year we can see whose is most accurate.

SourcePredictionObservedQuotes
UK Met Office and Uni. of Anglia 0.37 °C above the long-term (1961-1990) average of 14.0 °C
UK Met Office and Uni. of Anglia Global temperature for 2007 is expected to be 0.54 °C above the long-term (1961-1990) average of 14.0 °C 0.41 °C (2007) is likely to be the warmest year on record globally, beating the current record set in 1998, say climate-change experts at the Met Office.

(2008) This (cooling) was due to a much quicker than expected decline of a moderate El Niño that warms the climate, followed by the development of the strong cooling influence of the current La Niña.

lucia… the average temperature anomaly for 2008, as reported by GISS Land/Ocean measurements will be 0.70 ± 0.11 C.… the curve shown assumes that the forcing will increase at a constant rate of 0.48 W/m2 after 2003.
Niche ModelingO.2 (CRU)Future temperatures predicted using a random fractional differencing algorithm that generates realistic LTP behavior.

February 24, 2008

Surface Temperatures - How significant is the January 2008 fall?

Filed under: Uncategorized, Climate Change — admin @ 9:08 pm

As in the previous post about recent plummeting global temperatures, I want to look at the statistics of the drop, and determine its significance. The sort of questions of interest are, how improbable is a fall in temperatures of that magnitude of a 12 month period? After all, it is irresponsible to report alarming results without demonstrating the statistical significance. Unfortunately it is a common practice, for example, see record high temperatures from NASA.

The statistical setup for answering the question is encoded in the question. As we are only looking at falls in temperature, this should be a one-tailed test. The data we need are the twelve-month changes in global temperature anomalies, of which there are twelve every year to compare against the previous year. We then need the area of the distribution curve for these results, up to and including the value in question, -0.5906 in the case of the HadCRU data.