Thesis: Peter uses statistical models to help lakes in trouble

2015-11-27

Peter Dimberg defended his thesis "Predictions Within and Across Aquatic Systems using Statistical Methods and Models" the 27:th of November. 

From space, the Earth is blue. More than 70 percent of the planet's surface is covered by oceans. Besides the great oceans, planet earth also have more than 300 million lakes and ponds. But many lakes are in trouble. Peter Dimberg has developed new statistical methods to better investigate and predict the ecological state of lakes.

- Today there are a lot of statistical models to use, says Peter Dimberg when he settles down in my office with his brand new thesis in his hand.

- The problem is that many models are so complicated that only experts can use them. Or they might require large amount of data. Therefore, many lakes do not get the proper predictions, because of too few measurements, he adds. 

Peter wanted to develop simple methods, which can be used for many lakes, and can be used even though few measurements and sparse data. His thesis consists of five seperate studies that Peter hopes will simplify and improve estimates of the ecological status of different lakes. 

As an example, Peter has developed methods to predict how the annual chlorophyll concentration in a lake might vary, based on just few measurements in the summer, combined with the lake's latitude. The method can be useful for those lakes that lack regular recording during the cold months, he explains.

If you want to learn more about how to use statistical models to measure and predict chlorophyll concentration, phosphorus and oxygen content, dont miss Peter Dimberg doctoral thesis "Predictions Within and Across Aquatic Systems using Statistical Methods and Models."

Katarina Sundberg 

Link to thesis

News archive 2015