Weather Modeling: Parallelized and Pedagogical

Skylar Thompson

Table of Contents

1  Introduction

2  Weather Modeling

• Pre-computers: looking west
• Need more sources here

2.1  Computational

2.2  Parallel Aspects of Weather Modeling

• History back to 1960s on SOLOMONs
• Mesh
• Data type

3  What is Cligen?

• Model vs. Forecast
  • Short-term vs. long-term
  • Human vs. computer
• Data types
• Compare with MM5

3.1  Parallelization points

• Core statistical equations (marked with PARALLELIZE)
  • Calculation of the distance from the mean of precipitation, with units being std.dev.
  • Smoothing of monthly rainfall data
  • Computation of a, the ratio of half-hourly rainfall to total rainfall
  • Computation of c², or how well the data fit to a theoretical data
  • Computation of Taylor series. I don't know what it is or what it is used for, but I could parallelize it.
• Efficiency calculations
• Speedup calculations

4  User interface

• Intuitive
• Easy to make changes
• Fast results
• Explanation?
• Ability to combine multiple stations?
• How fine-grained should the changes be?

This document was translated from L^AT_EX by H^EV^EA.