Paper_notes.Rmd

---
title: "Paper notes"
author: "Adam"
date: "11/29/2021"
output: html_document
---

```{r setup, include=FALSE}
knitr::opts_chunk$set(echo = TRUE)
```

# Key Messages

1.  Route trends are useful for covariates and for exploring spatial patterns in population change

2.  For short periods of time, a log-linear slope is a reasonable estimate of the long-term trend

3.  spatial hierarchical slopes and intercepts provide a reasonable way to explore the spatial-temporal patterns of variation in population size. Assumption that abundance and trend are spatially dependent is extremely reasonable.

4.  leave future out, cross validation of the spatial vs non-spatial version. Consider also a comparison of the simple spatial version without the added random trend component.

5.  covariates on the slopes and intercepts can be easily added

6.  metric of the fit of a linear change - autocorrelation of the errors

7.  archived estimates of 15-year trends, and code to generate estimates for any other time-period