I am working with a humpback whale dataset collected across the North Pacific from 2004-2006.  Given the large spatial extent, I have selected a subset of data from the Gulf of Alaska (GOA) and would like to look for spatial patterns in the genetic diversity of the whales sighted in the GOA in relation to their environment.  Complicating this problem is the fact that most of the data was collected opportunistically, making the spatial distribution of whale sightings a better reflection of where researchers collected the data and not indicative of whether or not environmental variables influence  humpback whale habitat use.

Figure 1. North Pacific humpback whale sightings from SPLASH.  The data include > 18,000 photo-identification records and 2,700 DNA profiles for 8,000+ unique individuals.

Figure 2. A subset of the SPLASH data for the Northern and Western Gulf of Alaska. The data subset includes 2,622 records (both photo-identification and DNA profiles) for 1,448 unique individuals.

Ultimately, I need to figure out a method that will allow me to get beyond the uneven (non-systematic) sampling effort to determine if there is any sort of spatial pattern in the data based on genetics and environmental features (i.e. depth, slope, etc).  Two (among many) working hypotheses:

1. Humpback whales are found in clusters at a particular depth  or slope range.
2. Humpback whales that share the same haplotype (maternally inherited mitochondrial DNA) cluster together.

In the geological sciences spatial statistical analysis of gas distribution and migration thru subsurface systems has been applied in a limited number of studies across a variety of systems, CO2 storage, hydrocarbon exploration, landfills, and natural gas storage facilities on a fairly limited basis.  My study seeks to evaluate the source and mechanism of potential contaminates in groundwater systems affiliated with engineered-subsurface resource activities (e.g. hydrocarbon development, CO2 storage, EGS stimulation) using currently available datasets.

Specifically, this project seeks to apply geostatistical techniques in combination with spatial analysis of key datasets from a single geologic basin to evaluate the source and mechanism of gas and other potential contaminants in groundwater systems.  This project hypothesizes that larger-scale patterns in shallow methane concentrations in groundwater aquifers can be correlated to both primary migration pathways (such as wellbores or fracture networks) and the underlying volume of in situ hydrocarbon.  The general approach to this study is to identify, standardize, and integrate preexisting data from the study basin for use in geostatistical, relational, and probabilistic evaluation and interpretation.

The box model diagram below conceptually simplifies the primary systems interacting in the subsurface.  Datasets key to characterizing the flux of gas in and out of these systems, i) sources, ii) pathways, and iii) receptors, will require spatial characterization and statistical analysis in order to support predictions of areas of likely high-flux to receptors versus low-flux to receptors in relation to both natural and anthropogenic processes.

The following is the abstract of the paper I presented earlier this month at AAG:

The specific geography of individual wine growing regions has long been understood to be a significant factor in predicting both a region’s success in producing high quality grapes, and the resulting demand for wines produced from that region’s fruit. In the American wine industry, American Viticultural Areas (AVAs) are increasingly being used to designate a uniqueness and specificity of place. This process is often predicated on the argument that these areas represent a certain degree of physiographic uniformity or homogeneity. This is particularly the case with regard to the phenomenon of sub-AVAs, wherein smaller areas within large, spatially heterogeneous AVAs seek to differentiate themselves based on the physiographic features that are purportedly unique to those smaller subregions. In many cases, there is a strong correlation between soil classes and AVA boundaries, whereas in other cases the correlation is not as strong. This suggests that there are factors other than physiographic homogeneity contributing to the designation of these sub-AVAs. This study employs GIS and spatial analysis to examine and potentially correlate the soil classes of Oregon’s northern Willamette Valley with the sub-AVAs in that area. In doing so, this study presents maps and statistical results in order to provide a quantitative summary of the geographic context of vineyards in this region with respect to both the soil classes present and the federally designated AVA boundaries in which they are located.

About my data and my spatial problem:

The data set that I am working with is a legacy National Resources Conservation Service (NRCS) data set detailing soil classes throughout Oregon’s Willamette Valley. Using meets and bounds descriptions provided by the United States Department of the Treasury’s Alcohol and Tobacco Tax and Trade Bureau (TTB), the Federal entity tasked with approving AVA designation petitions, I have generated a series of polygons representing the Willamette Valley AVAs (Willamette Valley and its 6 sub-AVAs: Chehalem Mountains, Ribbon Ridge, Dundee Hills, Yamhill-Carlton, McMinnville, and Eola-Amity Hills). I also have a handful of raster data layers (slope, aspect, landform, lithology, and PRISM) that I am using to calculate zonal statistics. Many spatial statistical methods are designed around the use of point data – this poses a problem for me because all of my data is in either a vector polygon or raster format. I am interested in exploring which methods/tools within the Spatial Statistics toolbox are most appropriate for using with my data. I am also interested in getting feedback from others in this course so as to make my research more robust, defensible, and statistically sound.

-Doug

For my research, I use annual Landsat satellite images to view the Willamette River to examine disturbances and loss in the valley’s wetlands. I also have several other critical data sets including a LiDAR inundation raster based on a 2 year flood return interval and several shapefiles showing the location of mitigation wetlands. One of the spatial problems I’d like to investigate in this class involves relating the data sets to each other; one of the ecological questions I’m asking through my research involves investigating the spatial distribution of wetlands created and restored through mitigation versus those destroyed and disturbed. For example, do the two differ in their proximity to the river and its tributaries? Is one more clumped/distributed than the other?

Utilizing the spatial statistics toolbox, specifically regression and mapping trends/clusters, may help me answer some of these questions.

Some of my annual satellite imagery viewed in Tasseled Cap Index:

My spatial problem:

Under what forest and climatic conditions do endemic mountain pine beetle populations switch to epidemic populations?

The null hypothesis is that conditions are random and alternative hypotheses include: 1) population eruptions are simply cyclic or periodic; 2) there are specific environmental condition triggers; 3) some combination of alternative hypotheses 1 and 2.

Mountain pine beetle survival is dependent on availability of susceptible hosts and suitable temperature range, with the primary limitation being minimum temperature.  In an endemic state, mountain pine beetles may kill several trees in a dispersed pattern, while in an epidemic state, nearly continuous, widespread host tree mortality is observed.  Population eruptions exhibit both temporal and spatial patterns over the landscape making spatial statistics a useful analysis tool.

There are two parts to the study: 1) outbreak detection and monitoring using 40 years of Landsat satellite imagery and 2) analysis of relationships between outbreak initiation and spread and forest and climate conditions

Independent variables include:

Host availability at time of outbreak:

Host density

Host age

Topography:

Elevation

Slope

Aspect

Climate at and prior to time of outbreak:

Min and max temperature for various time periods

Precipitation for various time periods

Forest structure:

Composition

Management

Disturbance history

Each of these variables could potentially be related to outbreak timing and position through geographically weighted regression.

Coastal salt marshes are at great risk from a large number of factors, especially climate change and sea-level rise.  In an effort with the USGS, I’m working to determine how different salt marshes along the Pacific Coast will response to changes in sea level.  Part of our approach is collecting fine-scale, baseline field data in the form of RTK GPS elevation points and vegetation surveys.  Through analysis of data from a range of sites (up to 15 along the coast), I hope to better characterize plant  habitat requirements with an ultimate goal of producing improved community response projections under sea-level rise scenarios.  In this class, and in Jim Graham’s Spatial Modeling/Big Data class, I will be working with the elevation & veg data to characterize spatial relationships of plant species against plot-level factors (inundation frequency, distance to channel, elevation) and site factors (temperature, salinity, tidal range).  I have hundreds of vegetation plots per site, with about 2000 survey plots completed across our PNW sites.

Currently, I’m still in data processing mode, combining databases and gathering environmental data; field data collected wrapped up in January. However, the inundation data needs to be developed, first by kriging the elevation data into DEMs and then using site-specific waterlogger data to determine flooding frequency. The water logger data itself needs to be processed for barometric pressure and elevation. Marsh channels need to be digitized before a distance to channel raster can be created. There’s a lot of work still be done to get the data in shape for analysis, however by focusing on one or two sites, I’ll be able to explore the spatial statistics toolbox and push forward with this project.

My research seeks to quantify and explain patterns of variability as they relate to specific soil properties (such as nutrients, physical structure, ect.) There are patterns in the data itself (distribution shapes such as normal or bimodal, skewness, and variance) and in the spatial distribution of those data values.

I wish to learn more about tools that can characterize these data distributions and spatial patterns (emphasis on spatial for this class). It is especially a challenge because soil variables often don’t follow well-known distribution such as Gaussian or Exponential. This leaves me wary about the use of certain mathematical tools that requires assumptions such as a normal distribution. The central limit theorem does not apply when we move beyond questions about the mean.

At this point I do not have a specific question in mind, and I should also mention I haven’t collected any data for this project yet (I’m just starting lab analysis this spring). I do not have a specific spatial question, rather I’d to learn about various classification and interpolation methods.

Some background info on soil for the curios

A blurb from the Soil Science Society of America on the importance of soil:

Soil provides ecosystem services critical for life; soil acts as a water filter and a growing medium; provides habitat for billions of organisms, contributing to biodiversity; and supplies most of the antibiotics used to fight diseases. Humans use soil as a holding facility for solid waste, filter for wastewater, and foundation for our cities and towns. Finally, soil is the basis of our nation’s agroecosystems which provide us with feed, fiber, food and fuel.

On the source of soil variability:

Soil is HIGHLY heterogeneous. It is a mix of weathered rock minerals, plant organic matter, liquid, and gas. It’s been forming for thousands of years. A multitude of environmental variables affect that formation at spatial scales from nanometer bacterial interaction to varying climate across landscapes. The real challenge is that variability increases as a function of spatial area under consideration. The variability of a 0.5m X 0.5m plot is different than that of a 5m x 5m and is different than a 50m x 50m plot and so on.

A graphical look at shifting soil scale and methods of characterizing variability

My data consists of points derived from a GPS track log, which contains spatial information for GPS points taken at 30-second intervals along with a time stamp for each point. I also have a spreadsheet of field data containing location information for the start and end of an encounter with a species of cetaceans, the time the encounter started and when it ended and other important information such as the species, the number of animals, etc. In order to pair species’ encounters with the GPS tracklog, I use the time information of the encounter and associate those points in the tracklog that correspond with the beginning and ending times.

I am interested in a couple of spatial aspects of this data that are pertinent to this class:

1. Patterns in the environmental and oceanographic characteristics of the encounter locations that may explain melon-headed whale (Peponocephala electra) utilization of these locations.
2. The spatial distribution of melon-headed whales and other small cetaceans and the patterns in the presence or absence of melon-headed whales and the presence or absence of other species.

These areas of interest bring up the following spatial statistics related questions:

1. Do environmental and oceanographic characteristics differ significantly between locations?
2. Which variables are significant predictors of melon-headed whale utilization of these areas?
3. Do encounter locations differ significantly from locations where melon-headed whales were not seen?
4. Is there a relationship between the presence (or absence) of melon-headed whales and the presence of other species of small cetaceans?

I am sure there will be more questions that present themselves once I begin delving into the data.

Today, I solved my first problem (thanks Jen!) and successfully projected my data so that I could begin running spatial statistic analyses. My data went from GCS_WGS_1984 (unprojected) to NAD_1983_StatePlane_Alaska_1_FIPS_5001 which will allow for improved accuracy in spatial calculations for whale sighting data in southeastern Alaska. I ran an Average Nearest Neighbor analysis on humpback whales sightings in southeastern Alaska and found that the observed nearest neighbor distance was significantly smaller than the expected value. This significant difference is most likely due to the complex geography of southeastern Alaska which creates a clustering of individuals. I also learned that results of my spatial statistics analyses will be presented in meters. I look forward to running additional analyses next week!

My spatial problem is that my data were not collected randomly and field efforts were influenced by predicted habitat use or confirmed sightings of whales. Thus, what appear to be hot spots or patterns of habitat use within southeastern Alaska, might actually be areas of increased field effort. This will undoubtedly complicate my analyses and I continue to turn to the Arc Blog (and Dori) for answers.

We have talked about creating a random sample of whales in southeastern Alaska and comparing their patterns of habitat “use” to what we actually have in our data. Stay tuned for more on that…