Methods and tools to measure ecosystem metabolism

# Methods and tools to measure ecosystem metabolism
### Dr. Marcus Beck • mbeck@tbep.org • @fawda123
### Program Scientist, Tampa Bay Estuary Program
### Aug. 21, 2020

---

# A bit of history

---

### System-Wide Monitoring Program Resources for the Analysis of Time Series

#### [SWMPrats.net](SWMPrats.net)

* Workshop training materials and cookbooks

* Discussion forum

* Interactive dashboards

]

---

## SWMPrats.net: The SWMPr package

[SWMPr](https://github.com/fawda123/SWMPr) is an open-source R package for working with SWMP data

* Dealing with "bad" data

* Subsetting by date ranges, parameters

* Combining data from different sites

* Standardizing time steps

*[Beck 2016](https://journal.r-project.org/archive/2016/RJ-2016-015/index.html), The R Journal, 8(1)

---

## The current state of affairs

* SWMPrats is no more

* SWMPr is still maintained but not under active development

* The SWMPrExtension package provides a suite of tools that build on SWMPr, with a leaning towards automated reporting

*<https://github.com/NOAA-OCM/SWMPrExtension>

---

# What about ecosystem metabolism?

---

## Buried in the SWMPr package...

---

## In just a few short steps, estimate metabolism

```r
library(SWMPr)

# import and clean data
wq <- qaqc(apadbwq)
met <- qaqc(apaebmet)

# combine wq and met
dat <- comb(wq, met)

# estimate metabolism
res <- ecometab(dat)
```

*Methods from [Murrell et al. 2017](https://link.springer.com/article/10.1007/s12237-017-0328-9), Estuaries & Coasts 41(3)

---

## In just a few short steps, estimate metabolism

```r
plot_metab(res)
```

---

# Metabolic estimates have baggage

---

## Methodological assumptions

#### The Odum open-water method can provide an estimate of whole-system metabolic rates

$$
\frac{\delta DO}{\delta t} = P - R + D
$$
--

* Requires an accurate and continuous DO time series of the diel cycle

* Requires accurate estimate of air-sea gas exchange

* Assumes water column is well-mixed

* Assumes sensor is sampling the same water mass over time

*Odum 1956, Limnology and Oceanography 1

---

## Tidal advection is a nuisance

---

## Leads to "anomalous" estimates

---

# We can get fancy

---

## Detiding the DO signal

*[WtRegDO](https://github.com/fawda123/WtRegDO) R package, [Beck et al. 2015](https://doi.org/10.1002/lom3.10062), L & O Methods, 13(12)

---

## Leads to "better" estimates

---

## Why should you care?

#### You have some pre-existing tools at your disposal

* SWMPr R package: [github.com/fawda123/SWMPr](https://github.com/fawda123/SWMPr)
* WtRegDO R package: [github.com/fawda123/WtRegDO](https://github.com/fawda123/WtRegDO)

#### They can be improved to better satisfy requirements of the data
   
* Better constrain metabolism assumptions
* Field data to support stats models

#### They can be improved to suit your needs!

* What types of products would you like to see?

#### Metabolism as a standard unit of comparison, measure of ecosystem health

* Measure process rates, ripe for cross-reserve comparison

---

# Survey responses

---

# Survey goals

* Help us better understand your needs to develop applied research tools

* Gauge interest in participation

* Facilitate discussion

---

#### Q1: What are some historical or current management issues at your reserve that may be of interest for applying new methods to help understand ecosystem processes?

---

#### Q1: What are some historical or current management issues at your reserve that may be of interest for applying new methods to help understand ecosystem processes?

1. As a multi-component reserve, we have two tidal fresh, and one tidal brackish system, each with their own unique needs and challenges (eg urban vs rural inputs, SLR, saltwater intrusion, etc).
1. climate change
1. Linking more accurately river flow to estuary/bay metabolism and equating it to how the health of the system is changing. Connecting metabolism to ocean acidification and calculating it. Finally, better understanding how changes in river flow, metabolism, and ocean acidification are fluctuating and what it means for the health of the bay.
1. Effects of nutrient enrichment; effects of extreme events / river flow
1. how watershed alterations (hydrological manipulations and increases in nutrient load) affect ecosystem dynamics; potential for NEM be used as an indicator of coastal condition for regular status reports
1. Phosphate inputs; HABs; general increase in chlorophyll

---

#### Q2: What questions would you like to see addressed for cross-reserve comparisons to understand ecosystem processes?

1. Questions relative to climate change and episodic weather events
1. How fluctuations in metabolism correlate to observed biotic changes within the reserves.
1. what drives changes in ecosystem metabolism?; temporal and spatial scales of variability; can we relate water column metabolism estimates to biological observations?; does NEM reflect ecosystem health/condition?
1. Are there differences in primary drivers of metabolism for each reserve, if so, what are they?
1. Possibility the relationship between respiration, nutrients, and pH; carbon cycling/budget implications; any larger scale (teleconnection) patterns affecting variability at multiple reserves or is it mostly driven by local processes
1. what factors (that we can manipulate) contribute to optimal ecosystem status (e.g. carbon sequestration, eelgrass, little/no hypoxia, fisheries production, etc.)
---

#### Q3: Please describe any applications where you have used ecosystem metabolism. What questions were you addressing? What methods did you use to estimate metabolism?

1. for a Bays & Bayous meeting a few years ago, we used SWMPr to estimate ecosystem metabolism from our datasondes, and compared it to the BZI method (using monthly chl a, secchi depth, and PAR). Suggested our system may be switching from benthic primary producer dominance to pelagic; wasn't a super long time period but we have more data now and it might be fun to re-run this comparison.
1. No direct work but interest from carbon cycle and acidification perspective 
1. I have not been directly involved in any metabolism projects at our reserve. I know that there was metabolism work done by Neubauer, Anderson, Neikirk et al. in the late 90's through early 00's at our reserve sites. 
1. assisted a grad student studying impacts of tropical storms on nutrient and carbon loading - Marcus' SWMPr
1. Followed Caffrey paper, Odum open water methods  to understand long-term pH increases at South Slough SWMP stations from 1995 -2010, (has been decreasing pH trend since then) and drivers (ie if overall positive NEM due to eelgrass beds)
1. Estuarine carbon budgets, hypoxia. Summation of bottle / chamber measurements

---

#### Q4: Do you or any of your staff have experience using open-source software (e.g., R or SWMPr) to analyze SWMP data?

1. Not me personally (has been on the to-do list for a while.  New RC has much experience in R.)
1. Shannon Dunnigan (our SWMP Manager) has a lot of experience
1. Yes, Research Associate Lindsay Scheef
1. Yep
1. I use R frequently to analyze SWMP data
1. Some, but not much.

---

#### Q5: Anything else you'd like us to know?

1. I'm looking forward to being involved in this project!
1. We are also looking at how phytoplankton biomass and zooplankton populations respond to freshwater inflows, gross primary production and net ecosystem metabolism
1. have participated in past discussions related to issues with open water method for tidal systems for a short weekend course at Hatfield Marine Science Center, Newport, OR
1. I'm excited to learn and help contribute our long term SWMP data to this effort, but our current staffing capacity is very low.
1. I'm excited to participate in this!
1. very interested in action oriented scenarios- if we reduce nitrogen by X and temps stay at Y, we can expect Z outcome.....