\chapter{Introduction}
\labelchapter{intro}

Aerosols and aerosol-cloud interactions remain a great source of uncertainty
in global climate models. Haero, the modal aerosol model described in this
document, attempts to help researchers better understand the relevant physical
processes and their contributions to the global climate.

Any model must define abstractions and make assumptions in order to produce
quantitative answers to scientific questions. The abstractions and assumptions
for Haero live in this document. We also describe Haero's software
components---its data structures and programming interface---which are designed
to match these abstractions. In particular, we aim to provide an interface that
allows a scientist to write code that resembles statements made in technical
conversations. We hope that this close correspondence between code and
the related scientific discourse can allow people with different areas of
expertise to contribute to the development of Haero.

In \refchapter{physics}, we very briefly outline the governing equations of
aerosol dynamics from a mathematical perspective. We begin from basic transport
equations formulated in terms of aerosol size distribution functions. We also
describe the assumptions that enter the model at the level of continuous
mathematics (whose approximation errors are present even in analytic
solutions!).

\refchapter{library} introduces the Haero library and its various abstractions.
Because the coupling of aerosol-related processes is still an active area of
research, Haero's software interface focuses on defining specific physical and
mathematical entities, providing a set of elementary building blocks from which
more elaborate models can be constructed.

Next, in \refchapter{testing}, we describe procedures for testing aerosol processes
individually, to make sure they work properly. This type of testing is sometimes
called ``unit testing'' by software people, and ``verification testing'' by
engineers. Either way, it's essential to test your aerosol processes in an
isolated environment with simple inputs that lead to outputs with predictable
characteristics. If you don't test your process this way, {\em you have no idea
what it actually does!}

In \refchapter{processes}, we describe the various processes provided by Haero
that model the stages of the aerosol lifecycle. Here we explain the basic
physical assumptions embedded in each process, along with the details of each
available implementation.

In \refchapter{driver}, we introduce Haero's stand-alone driver program, which
provides a way to run simple aerosol simulations and test parametrizations.
The driver provides a dynamics model to allow researchers to study how transport
and radiative processes affect aerosols.

A comprehensive description of the API appears in \refappendix{api}. The input
specification for the driver is explained in detail in \refappendix{driver_input}.
A glossary of terms and definitions is also provided in \refappendix{glossary}.