Workshop excercises

From Intamap

Contents

[edit]

Introduction

Detecting and monitoring of nuclear releases is an important issue. Many European countries have installed monitoring networks to perform this task. In the Netherlands the National Radioactivity Monitoring network was established. Real-time automatically generated maps from the network would benefit the interpretation of the data. A prototype for such a system was designed and implemented between September 2006 and June 2007 in a BSIK/RGI project (RGI-182, which is continued until December 2008 in RGI-302)

The excercises in this workshop consist of a number of multiple choice questions divided into several topics. The links needed for the excercises can be found here.

[edit]

Excercise 1 - Normal conditions

Open the client application using the links. It shows the most recent prediction map by default.

Question 1

Zoom in to the Southern province of Limburg (don't forget the Refresh button). Why is the gamma dose rate in Limburg higher than in the rest of the country?

  • a) This is the highest part of the Netherlands and higher elevation means a higher dose rate
  • b) The soil in the area (Loess) contains more radioactive particles
  • c) The area is home to a nuclear waste processing facility
  • d) This is a coincidence, natural variability.

Question 2

There exist different types of interpolation. What kind of interpolation do we use in this prediction map?

  • a) Ordinary Kriging
  • b) Universal Kriging (also known as kriging with external drift, or regression kriging)
  • c) Simple Kriging
  • d) Cokriging.

Question 3

Pan the map to the province of Utrecht, in the center of the Netherlands. There are a lot of white pixels visible in this province. Why is there no prediction available for these pixels?

  • a) There are no monitoring stations in the vicinity of these pixels
  • b) There is no information on the trend variable(s) available for these pixels
  • c) The variogram is undefined for these pixels
  • d) These pixels were covered by clouds

Question 4

Switch from Prediction view to Standard error view, and turn on (toggle display of) the layer containing the monitoring stations. Which of the following statements is true?

  • a) The uncertainty of predictions is lower around the monitoring stations
  • b) Uncertainty of predictions is higher where variability of observations is larger
  • c) Uncertainty of predictions is higher where the gamma dose rate are higher.
  • d) Uncertainty of predictions varies with the trend.

Question 5

Switch the Standard error view to Prediction interval. What is the interpretation of grid cells classified as Not distinguishable?

  • a) The threshold value is below the lower bound of the interval
  • b) The threshold value is above the upperr bound of the interval
  • c) The threshold value is within the prediction interval so we are not confident whether the threshold is exceeded, due to limited information
  • d) For these cells, the maps is not distinguishable from the time step before the current one

Question 6

Which of the following statements are true about not distinguishable cells

  • a) The spatial variability is so large that no indication of threshold exceedance can be given.
  • b) The uncertainty is so high that no indication of threshold exceedance can be given.
  • c) Not enough monitoring stations are nearby to say whether the threshold is exceeded.
  • d) They would vanish (become lower or higher) when the data did not contain measurement error.
[edit]

Excercise 2 - Rainstorm passing over the Netherlands

The washout of radon by precipitation increases the gamma dose rate. Watch the animation of the rainfall radar together with an interpolated map of the NRM. If you are finished, change the date of the WMS application to the 20th of July 2007, 07:00 hours. Advance time by using "next" to see the rainstorm passing over the Netherlands. Try different types of maps (prediction, uncertainty, prediction interval).

Question 7

Switch the view to the prediction map. From 10:50 to 11:00 the rain storm seems to vanish from the prediction map and instead the whole of the Netherlands suddenly has an elevated gamma dose rate. How can this be explained.

  • a) Rainfall intensity becomes more constant over the Netherlands.
  • b) A burst of radiation from outer space causes an increase overr the Netherlands.
  • c) The higher values in the South-West influence the values for the trend in the remainder of the Netherlands.

Question 8

The interpolations shown in the animation do not contain sharp boundaries. What kind of interpolation did we use for the animation?

  • a) Ordinary Kriging
  • b) Universal Kriging (also known as kriging with external drift, or regression kriging)
  • c) Simple Kriging
  • d) Cokriging.
[edit]

Excercise 3 - A recently observed event

Change the date of the application to the 28th of September 2007, 00:00 hours and switch the view to Deviation from long term mean. Advance through time to 07:00 using the next button.

Question 9

What would be a plausible explanation for this phenomenon?

  • a) Malfunctioning monitoring equiptment.
  • b) The arrival of air from Germany containing a higher concentrations of radon in combination with heavy rainfall during the night
  • c) Small release from the Dutch nuclear powerplant at Borsselle (Zeeland).
  • d) Climate change.
Retrieved from "http://wiki.intamap.org/index.php/Workshop_excercises"