My spatial problem

• A description of the research question that you are exploring.

My research aims to determine the current abundance and spatial distribution of native Olympia oysters in Yaquina Bay, Oregon. This oyster species has experienced massive decline in population due to overharvest during European settlement of the western United States. Yet its value to the ecosystem, its cultural importance, and its tastiness have made the Olympia oyster a current priority for population enhancement. For my research, I will be focusing on a local population of Olympia oysters in the Yaquina estuary. The goal of my project is to gather baseline information about their current abundance and spatial distribution, then develop a repeatable biological monitoring protocol for assessing this population in the future. Using spatial technology, I will first assess whether the distribution of Olympia oysters can be predicted using three habitat parameters: salinity, substrate availability, and elevation. In collaboration with the Oregon Department of Fish and Wildlife (ODFW), I will use the results of this spatial analysis and field surveys to determine ‘index sites’, which are specific locations within the estuary that are indicative of the larger population. These index sites will be revisited in the future by ODFW’s Shellfish and Estuarine Assessment of Coastal Oregon (SEACOR) team to assess changes in population size and spread over time. If predictions of Olympia oyster distribution are accurate based on the habitat parameters I’ve identified, then I’d also like to analyze potential species distribution under future environmental conditions and under different management scenarios, including habitat restoration and population enhancement.

For this course, I will be exploring this main research question:

How is the spatial pattern of Olympia oysters in the Yaquina estuary [A] related to (caused by) the spatial pattern of three habitat parameters (salinity, substrate, elevation) [B]?

• A description of the dataset you will be analyzing, including the spatial and temporal resolution and extent.

I will be using three spatial datasets, representing each of the habitat parameters, overlaid on one another to rank most to least likely locations for Olympia oyster presence. The salinity dataset is based on historical measurements (1960-2006) and represents a gradient from highest salinity (~32psu) at the mouth of the estuary to fresher water up stream (<16psu). Elevation is represented through a bathymetric dataset from 2001-2002, sourced from the Environmental Protection Agency office in Newport, OR. The substrate data comes from the Oregon ShoreZone mapping effort in 2014, which is managed and updated by the Oregon Coastal Management Program. There’s a couple different ways this data can be used, either as a substrate layer that characterizes substrate type broadly (low resolution) or through vector data with associated data tables that describe the substrate within a tidal zone along the shoreline (higher resolution, but spatial extent is limited).

The images here show the three habitat parameter spatial datasets:

Yaquina Bay bathymetry derived from subtidal soundings in 1953, 1999, 1998, and 2000 by U.S. Army Corps of Engineers.
Data from EPA.

Salinity figure digitized from Lewis et al. (2019) based on Oregon’s wet-season salinity measurements (average salinity November-April).
Lewis, N. S., E. W. Fox, and T. H. DeWitt. 2019. Estimating the distribution of harvested
estuarine bivalves with natural history-based habitat suitability models. Estuarine, Coastal and Shelf Science, 219: 453-472.

Substrate component classes of Yaquina Bay based on data classifications from Coastal and Marine Ecological Classification Standard (CMECS) ‘Estuarine Substrate Component’ layer.
Data from Oregon Coastal Management Program.

• Predict the kinds of patterns you expect to see in your data, and the processes that produce or respond to these patterns.

I am hypothesizing that the distribution/spread of Olympia oysters in Yaquina Bay is influenced by availability of appropriate habitat parameters; where these parameters align within the appropriate range will determine where the oysters can be found. However, I think that I will find that not all of the parameters equally influence oyster distribution. For example, Olympia oysters have been observed to tolerate a broad salinity range, but are absolutely not present without suitable substrate. I am expecting to see that the influence of a particular habitat parameter changes depending on where the oysters are located within the estuary. I’m curious to see, if possible, which parameter will be most important at what life stage and what may drive changes in population per specific site in the estuary.

I do expect that I will be able to make a prediction about where the oysters will be located based on the habitat parameters, though I am uncertain that the resolution of the spatial data is sophisticated enough to capture nuances in distribution. For example, Olympia oysters are known to be opportunistic in finding suitable substrate and will settle on a wide variety of hard surfaces, including derelict boating equipment, discarded shopping carts, and pilings.

• Describe the kinds of analyses you ideally would like to undertake and learn about this term, using your data.

I want to be able to produce a model that can predict where the oysters are located based not just on the three habitat parameters of interest, but under various environmental conditions and different management scenarios. For example, where might the oysters settle in a given year if rainfall is substantially higher, or if adult oysters spawn earlier, or if oyster growers create Olympia oyster beds for harvest, or if a new habitat restoration site is established, etc.

I’m not especially handy at statistical analysis, so I would like to gain a better understanding of statistics through spatial data. I know that I will need to use statistics to determine how successfully the prediction of Olympia oysters aligns with actual observations in the field, but currently unsure how to do that. A recent study in Yaquina estuary was just released using a similar approach for predicting the distribution of five other bivalve species. This study used R to generate a logistic regression model to determine the probability of each species presence within a given area. I would like to do something similar for my analysis, but need some help.

• What do you want to produce — maps? statistical relationships?

The desired products of this research are habitat suitability maps of the current and future (pending the success of the initial effort) distribution of native Olympia oysters for use by ODFW. As a part of this effort, I will create a map of index site locations to be used in future species monitoring. I would also like to generate a predictive model that can determine distribution of oysters based on annual changes in the local environment (El Nino conditions, heavy rainfall, restoration efforts, introduction of invaders, etc.). While salinity, substrate, and elevation seem to be the main factors influencing oyster distribution, there are a number of other factors that can have effects, including temperature, proximity to the mouth of the estuary, and tidal retention.

• How is your spatial problem important to science? To resource managers?

ODFW currently does not have reliable baseline information on the distribution of Olympia oysters in Oregon. As an ecological engineer, the species provides a number of important benefits to the ecosystem, including water filtration and habitat for other marine creatures. It is culturally significant to local tribes, including the Confederated Tribes of Siletz. This species is not currently listed as threatened or endangered, but if it becomes listed one day, then that designation will trigger a number of mitigation and conservation measures that will be difficult and expensive for agencies and private landowners. Additionally, there’s been some exploration that if the population can become robust again, there is potential to grow and harvest this species as a specialty food product. Given the current slow food movement and interest in local products, Olympia oysters could fit well in this niche.

• How much experience do you have with:

(a) Arc-Info – Little experience, used a bit with older versions of ArcMap.

(b) Modelbuilder and/or GIS programming in Python – I am comfortable with ModelBuilder, but have no experience with Python.

(c) R – Some experience; I took Stats 511 where we used R heavily in a series of lab exercises. I have not applied my own data in R.

(d) Image processing – I have used a variety of Adobe products for graphic design, including Photoshop and InDesign.