State of the Ecosystem Product Development Workflow

class: right, middle, my-title, title-slide

# State of the Ecosystem Product Development Workflow
## Openscapes 15 October 2021
### Kim Bastille <a href="mailto:kimberly.bastille@noaa.gov" class="email">kimberly.bastille@noaa.gov</a> Github: @kimberly-bastille Twitter: @swimbastille

---

class: top, left

# Presentation Overview
.pull-left[

### State of the Ecosystem Overview

### Workflow Structure

### Cool technique for propagating plots throughout workflow

]

.pull-right[
![:img SOE-logo, 80%](EDAB_images/SOE_logo.png)
]

---

# SOE Overview

## State of the Ecosystem Reports
.pull-left[
* 2 Annual reports that give ecosystem context to fisheries managers.
* 2 councils, 3 Ecological Production Units
* ~50 Indicator datasets, Over 60 data contributors
* Evolving since 2016
* Supporting products
  * [Technical Documentation](https://github.com/NOAA-EDAB/tech-doc) 
  * [Response Memo](https://github.com/NOAA-EDAB/memos/tree/master/SOE-response-memo)
  * [Presentations](https://github.com/NOAA-EDAB/presentations), etc.

.contrib[

[State of the Ecosystem 2021, Mid-Atlantic](https://apps-nefsc.fisheries.noaa.gov/rcb/publications/SOE-MAFMC-2021-508-Final.pdf)

[State of the Ecosystem 2021, New England](https://apps-nefsc.fisheries.noaa.gov/rcb/publications/SOE-NAFMC-2021-508-Final.pdf)
]  
]

.pull-right[
![:img soe-map, 80%](EDAB_images/epu_map.png)
]

---

# SOE Workflow Structure

## State of the Ecosystem workflow tree

![:img soe-tree, 80%](EDAB_images/soe_workflowtree.png)
---
# ecodata - R package

.pull-left[
![:img ecodata_logo, 80%](EDAB_images/ecodata_logo.png)

.contrib[

https://github.com/NOAA-EDAB/ecodata
]
]

.pull-right[
## ecodata package purpose

* Data Processing
* Data Sharing
* Data Visualization

### Data Processing
* Rscipts that format contributor data

### Data Sharing
* Data saved in `data` folder
* Can be lazyloaded with package
  * mtcars
]

---
## Data Visualization and Propagation

1) Plots built in Visualization.Rmd

.contrib[

https://noaa-edab.github.io/ecodata/macrofauna_MAB#Forage_Anomaly
]

---
## Data Visualization and Propagation

2) Rmd Code Chunks converted to Individual R scripts

3) Rscripts stored on github

![:img ecodata_snap, 70%](EDAB_images/ecodata_screenshot.png)

.contrib[

* https://github.com/NOAA-EDAB/ecodata/blob/master/data-raw/code_chunks_to_scripts.R
* https://github.com/NOAA-EDAB/ecodata/tree/master/chunk-scripts
]

---
## Data Visualization and Propagation

4) use `code = readLines` in Rmd

![:img report-rmarkdown, 80%](EDAB_images/report_rmarkdown.png)
![:img forage_anom, 80%](EDAB_images/forage_anomaly.png)
---
# Benefits

.pull-left[
![:img soe-tree, 90%](EDAB_images/soe_blanktree.png)
]

.pull-right[
## Cool Visualization technique
* No copy/paste errors
* All associated products get most up-to-date figure
* Clearly defined tasks for individuals
* Easier to maintain 100+ plots
* Easy to set up and update

## Workflow as a whole
* Keeps code and data all in one place
* Processed data available to public
* Can be maintained by anyone
]
---
# Questions??

.center[
![:img SOE_feature, 70%](EDAB_images/SOE_feature_image.png)

]