Homework 6 BMElib track

Advanced mapping estimation

Date given: 10/26

Due: 11/9 noon

 

 Part 1

 

Before you start, review the mappingHardDataST.m program providing a tutorial on using BMElib to obtain BME mean estimates across space using a dataset of space/time hard data. 

 

Using the ENVR468hwk6part1.txt dataset use BMElib to obtain a map of the BME mean estimate and a map of the associated estimation error variance at time t=12 day.  Assume that the space/time covariance model is the following

 

 

where r is the spatial lag (Km), t is the temporal lag (day), and the covariance parameters are c01=1.5 ppm2, ar1=7 Km and at1=20 days, c02=0.4 ppm2, ar1=1 Km and at1=2 days

 

Provide a good quality write-up describing your analysis and the results you obtained.  This write-up should provide you a template for the mapping analysis section of your project report.

 

Part 2

 

Do the same analysis as in Part 1, but this time using the dataset for your class project.  Provide a write up describing your analysis and the results you obtain, including a map of the BME mean estimate for a time that you select, as well as the associated map of mapping error variance.  This write up should serve as the starting point for the mapping analysis section of your project report.

 

Following is a hypothetical example that you can use to inspire your own analysis.  The variable is a hypothetical compound named agentX (measured in ppm).  The dataset you have created in space/time vector format is in the file ENVR468hwk6part2.txt.  The steps of analysis of this dataset includes (1) an exploratory analysis, (2) the modeling of the space/time covariance, and (3) obtaining maps of estimated agentX for selected times.  This analysis is performed in the ENVR468hwk6part2solution.m program.  Download both of these files in a same directory, run the analysis, and see the output (figures) created at each step of the analysis.  The following figure shows some brief results for each of these steps.  Use this as an example that you will expand for the analysis of your own dataset.

 

                                               (a)                                                                                         (b)

datacov

                                                                                            (c)

estmated

Figure: (a) Data of log-agentX (log-ppm) collected at time 2000.096 year, (b) space/time covariance model of log-agentX, and (c) map of the BME mean estimate of log-agentX at time 2000.096 year.