USGS ScienceSearch

Geology topics

Jerome Fiechter

Publications and source records attributed to Jerome Fiechter.

3 recordsLinked to original sources

A comparison of deep-sea coral and sponge species distribution models and the impact of ocean currents in the Southern California Bight

Deep-sea corals and sponges (DSCS) are signature taxa of deep-water habitats, but ecological mechanisms that drive their geographic distributions can be difficult to uncover due to the challenges of surveying deep-water ecosystems and limited oceanographic data. A comparison of species distribution models was made for three DSCS ( Antipathes dendochristos, Plumarella longispina, and an unidentified sponge morphotype known as Porifera #2), two of which are associated with young-of-the-year and adult rockfish taxa ( Sebastes spp.). Models were built using generalized additive models (GAMs) that accounted for spatial autocorrelation. These models were compared to Maxent models, a commonly used method. All models utilized bathymetrically derived variables (including depth, slope, and Bathymetric Position Index) and output from a coupled physical-biogeochemical ocean model (including current direction and magnitude, food abundance, and dissolved oxygen). Direction of benthic ocean currents played an important role in predictions, pointing to large-scale ecological processes that may determine presence of DSCS. GAMs outperformed Maxent models and predicted more ecologically interpretable patterns. Additionally we quantified the predicted amount of suitable habitat that is currently located within some protected areas of the Southern California Bight and show that many hot spots occur outside protected areas. Given the importance of DSCS for multiple demersal fish species of commercial or conservation concerns, this research provides valuable information that natural resource managers can use as a tool in the Southern California Bight.

California

Spatial variability in ocean-mediated growth potential is linked to Chinook salmon survival

Early ocean survival of Chinook salmon, Oncorhynchus tshawytscha, varies greatly inter-annually and may be the period during which later spawning abundance and fishery recruitment are set. Therefore, identifying environmental drivers related to early survival may inform better models for management and sustainability of salmon in a variable environment. With this in mind, our main objectives were to (a) identify regions of high temporal variability in growth potential over a 23-year time series, (b) determine whether the spatial distribution of growth potential was correlated with observed oceanographic conditions, and (c) determine whether these spatial patterns in growth potential could be used to estimate juvenile salmon survival. We applied this method to the fall run of the Central Valley Chinook salmon population, focusing on the spring and summer period after emigration into central California coastal waters. For the period from 1988 to 2010, juvenile salmon growth potential on the central California continental shelf was described by three spatial patterns. These three patterns were most correlated with upwelling, detrended sea level anomalies, and the strength of onshore/offshore currents, respectively. Using the annual strength of these three patterns, as well as the overall growth potential throughout central California coastal waters, in a generalized linear model we explained 82% of the variation in juvenile salmon survival estimates. We attributed the relationship between growth potential and survival to variability in environmental conditions experienced by juvenile salmon during their first year at sea, as well as potential shifts in predation pressure following out-migration into coastal waters.

Fisheries Oceanography

Assessing first-order emulator inference for physical parameters in nonlinear mechanistic models

We present an approach for estimating physical parameters in nonlinear models that relies on an approximation to the mechanistic model itself for computational efficiency. The proposed methodology is validated and applied in two different modeling scenarios: (a) Simulation and (b) lower trophic level ocean ecosystem model. The approach we develop relies on the ability to predict right singular vectors (resulting from a decomposition of computer model experimental output) based on the computer model input and an experimental set of parameters. Critically, we model the right singular vectors in terms of the model parameters via a nonlinear statistical model. Specifically, we focus our attention on first-order models of these right singular vectors rather than the second-order (covariance) structure.

Journal of Agricultural, Biological, and Environme