Ecological forecasts for marine resource management during climate extremes

Brodie, Stephanie1 ; Pozo Buil, Mercedes1; Welch, Heather1; Bograd, Steven2; Hazen, Elliott2; Santora, Jarrod2; Seary, Rachel1; Schroeder, Isaac1; Jacox, Michael2

Published Nov 12, 2023 on Dryad. https://doi.org/10.5061/dryad.z08kprrjr

Abstract

Forecasting weather has become commonplace, but as society faces novel and uncertain environmental conditions there is a critical need to forecast ecology. Forewarning of ecosystem conditions during climate extremes can support proactive decision-making, yet applications of ecological forecasts are still limited. We showcase the capacity for existing marine management tools to transition to a forecasting configuration and provide skilful ecological forecasts up to 12 months in advance. The management tools use ocean temperature anomalies to help mitigate whale entanglements and sea turtle bycatch, and we show that forecasts can forewarn of human-wildlife interactions caused by unprecedented climate extremes. We further show that regionally downscaled forecasts are not a necessity for ecological forecasting and can be less skilful than global forecasts if they have fewer ensemble members. Our results highlight capacity for ecological forecasts to be explored for regions without the infrastructure or capacity to regionally downscale, ultimately helping to improve marine resource management and climate adaptation globally.

README

This README file was generated on 2023-10-17 by Stephanie Brodie

GENERAL INFORMATION

1.Title of Dataset: Ecological forecasts for marine resource management during climate extremes
2.Lead author Information: Stephanie Brodie, University of California Santa Cruz, sbrodie@ucsc.edu
3.Study area: California Current System, USA
4.Funding: Funding was provided by the NOAA Climate Program Office Modeling, Analysis, Prediction, and Projections program (NA17OAR4310108), the California Current Integrated Ecosystem Assessment (no grant number), and the California Ocean Protection Council (2021-242-UCSC).

SHARING/ACCESS INFORMATION
1. Data used and generated are from this publication: Brodie et al.\, (in press). Ecological forecasts for marine resource management during climate extremes. Nature Communications.
2. Links to ancillary data sets:

Global SST forecasts can be accessed at the North American Multi-Model Ensemble (http://iridl.ldeo.columbia.edu/SOURCES/.Models/.NMME/).
Output from the UCSC CCS reanalysis is available from https://oceanmodeling.ucsc.edu.
The real-time HCI (https://oceanview.pfeg.noaa.gov/whale_indices/) and TOTAL (https://coastwatc h.pfeg.noaa.gov/loggerheads/index.html) management tools are publicly accessible.
Corresponding code used to generate forecasts is on GitHub: https://github.com/stephbrodie1/Ecological_Forecast. Cited as: Brodie, Stephanie, Pozo Buil, Mercedes, Welch, Heather, Bograd, Steven, Hazen, Elliott, Santora, Jarrod, Seary, Rachel, Schroeder, Isaac, and Jacox, Michael (2023). Ecological forecasts for marine resource management during climate extremes. https://doi.org/10.5281/zenodo.8429074.
Additional details on downscaling methods are described in: Jacox, M., Pozo Buil, M., Brodie, S., Alexander, M., Amaya, D., Bograd, S., Edwards, C.A., Fiechter, J., Hazen, E., Hervieux, G., et al. (in press). Downscaled seasonal forecasts for the California Current System: Skill assessment and prospects for living marine resource applications. PlosOne. 10.1371/journal.pclm.0000245

3. Recommended citation for this dataset:
Brodie, Stephanie, Pozo Buil, M., Welch, Heather, Bograd, Steven, Hazen, Elliott, Santora, Jarrod, Seary, Rachel, Schroeder, Isaac, and Jacox, Michael (2023). Ecological forecasts for marine resource management during climate extremes. doi:10.5061/dryad.z08kprrjr.

DATA AND FILE OVERVIEW
1. Folder: HCI

Global_hci_forecast folder: forecast HCI for each global model 'forecast_allmembers_HCI_1981-2021.rds'. The ensemble mean of all global models and their ensembles ('forecast_ensmean_HCI_1981-2021.rds') and the ensemble of only 3 ensemble members from CanCM4 ('forecast_ensmean_only3_HCI_1981-2010.rds') are provided as separate files.
Downscaled_hci_forecast folder: forecast HCI for the downscaled models at each initialisation time ('downscaled_forecast_allmembers_HCI_1981-2010.rds'). The ensemble mean is provided as a separate file ('downscaled_forecast_ensmean_HCI_1981-2010.rds').
Column name description for these files:
- glob_model: the global model (see below for full names fro acronyms)
- ensemble: numeric ensemble number
- init_month: initialisation month
- lead_month: forecast lead month
- forecast_date: forecast month
- sst_area: number of grid cells used in the calculation of the habitat compression index
- area: the habitat compression index value

2. Folder: TOTAL

Global_total_forecast_1981-2020 folder: forecast TOTAL for each global model ('total_metric_globalforecasts_1980-2020_bestcase.rds') (1981-2020) and the ensemble mean ('total_metric_globalforecasts_1980-2020_bestcase_ensemble.rds') (1981-2020)
Global_total_forecast_1981-2010 folder: forecast TOTAL for each global model ('total_metric_allforecasts_1980-2010_apples.rds') (1981-2010) and the ensemble mean of TOTAL forecast ('total_metric_globalforecasts_1980-2010_apples_ensemble.rds') (1981-2010)
Global_total_forecast_3ens_1981-2010: ensemble mean of TOTAL forecast (1981-2010) for the 3 ensemble members of CanCm4 ('total_metric_globalforecasts_1980-2010_only3_ensemble.rds')
Downscaled_total_forecast_1981-2010: downscaled forecast TOTAL ('total_metric_globalforecasts_1980-2010_apples_ensemble.rds') (1981-2010)
Column name description for these files:
- glob_model: the global model (see below for full names fro acronyms)
- ensemble: numeric ensemble number
- lead year: year of forecast
- june_fcast: TOTAL forecast for the june closure month
- june_obs: TOTAL observed for the june closure month
- jul_fcast: TOTAL forecast for the july closure month
- jul_obs: TOTAL observed for the july closure month
- aug_fcast: TOTAL forecast for the august closure month
- aug_obs: TOTAL observed for the august closure month

Global Model Acronyms:

CanCM4i: Canadian Center for Climate Modeling Analysis
GEM_NEMO: Global Environmental Multiscale - Nucleus for 10 European Modelling of the Ocean
GFDL-SPEAR: Geophysical Fluid Dynamics Laboratory- Seamless System for Prediction and Earth System Research
NASA-GEOSS2S: National Aeronautics and Space Administration - Goddard Earth Observing System Subseasonal to Seasonal
COLA-RSMAS-CCSM4: Center for Ocean Land Atmosphere Research - Rosenstiel School of Marine, Atmospheric, and Earth Science - Community Climate System Model version 4.0
NCEP-CFSv2: National Centers for Environmental Prediction - Climate Forecast System Version 2

3. Folder: Source_data
File name correspond to figures in the published manuscript.
Column name description for these files:

fig1_bottomleft_ts.csv

year: year
month: month
area: observed habitat compression index
date: dated to be the first day of the month-year for plotting

fig1_bottomright_ts.csv

year: year
month: month
day: first day of the month
date: dated to be the first day of the month-year for plotting
sst: sea surface temperature
sst_clim: sst climatology
obs_anom: sst anomaly
roll_obs_anom: 6 month rolling mean of the sst anomaly

fig1_topleft_map.csv

x: longitude in 0-360 degrees
y: latitude
value: sea surface temperature
lon: longitude in -180-180 degrees

fig1_topright_map.csv

x: longitude
y: latitude
value: sea surface temperature

fig2_bottom.csv

month: month
lead: forecast lead month
cor: correlation value of the retrospective forecasts
cor_sig: binary value of whether correlation is significant (1) or not (0)
acc: accuracy of the retrospective forecasts
acc_sig: binary value of whether accuracy is significant (1) or not (0)
sedi: Symmetric Extremal Dependence Index (sedi) of the retrospective forecasts
sedi_sig: binary value of whether sedi is significant (1) or not (0)

fig2_top.csv

forecast_date: habitat compression index date
type: observed or forecast
area: habitat compression index value
lead_month: lead months for forecast
mean: threshold used to define habitat compression
group: a binary value indicating high compression (1) or not (0)

fig3_bottom.csv

lead: forecast lead month
correlation_months: TOTAL closure month corresponding to correlation_value
correlation_values: correlation value of the retrospective forecasts
cor_sig_months: TOTAL closure month corresponding to cor_sig_value
cor_sig_values: binary value of whether correlation is significant (1) or not (0)
acc_months: TOTAL closure month corresponding to acc_value
acc_values: accuracy value of the retrospective forecasts
acc_sig_months: TOTAL closure month corresponding to acc_sig_value
acc_sig_values: binary value of whether accuracy is significant (1) or not (0)
sedi_months: TOTAL closure month corresponding to sedi_value
sedi_values: sedi value of the retrospective forecasts
sedi_sig_months: TOTAL closure month corresponding to sedi_sig_value
sedi_sig_values: binary value of whether accuracy is significant (1) or not (0)

fig3_top.csv

forecast_date: forecast date
type: observed or forecast
sst_anom_rolling: sea surface temperature anomaly 6 month rolling mean
lead: lead month
ssta_quantile: threshold used to identify potential closures

fig4.csv

month: forecast month
lead: lead time
model: type of restrospective forecast test time period used
values: habitat compression index
tool: habitat compression tool
metric: skill metric (correlation, accuracy, and sedi)

4. NETCDF FILES
The regionally downscaled sea surface temperature files are provided here. File name structure: sst_forecset_ens_initialisation_monthavg_1982_2010.nc, where:
ens: indicates the ensemble member (either 2, 8, 10, or the ensemble mean)
initialisation: indicates whether forecasts were initialised in January (init1) or July (init7).
Files contain sea surface temperature (sst) for every grid cell (lat and lon) for each lead time, ensemble member, and initialisation time.

5. HOW TO OPEN FILES
RDS files can be opened in R, such as:
readRDS(filename.rds)

Netcdf files can be viewed through Panoply, or opened in R. For example:
install.packages("ncdf4")
library(ncdf4)
data <- nc_open(‘FILENAME.nc’)
print(data)

<br>

Methods

Summary

We configure two existing resource management tools, originally configured to use observed (historical) ocean temperatures, to a forecasting system and conduct a retrospective forecast to test their skill. We first conducted a retrospective forecast using global forecasts (73 ensemble members) across the full historically available period (1981-2020) – termed the ‘Global’ model. Global forecasts of monthly sea surface temperature were obtained from the North American Multimodel Ensemble (NMME; Table S1; https://www.cpc.ncep.noaa.gov/products/NMME/).

We then compared the performance of three forecast configurations: First, we used global forecasts (73 ensemble members) across a reduced historical period (1981-2010) - termed the ‘Global Full Ensemble’. Second, we used forecasts regionally downscaled (3 ensemble members) to the CCE for the same reduced historical period (1981-2010) - termed the ‘Downscaled Ensemble’. Third, we used a reduced subset of the global forecasts (3 ensemble members) for the same reduced historical period (1981-2010) - termed the ‘Global Reduced Ensemble’.

All forecasts are compared to SST observations, extracted from a CCE regional reanalysis. This reanalysis is based on the Regional Ocean Modeling System (ROMS) and covers the west coast of the U.S. (30-48˚N, 134-115.5˚W) with 0.1 degree (~10 km) horizontal resolution and 42 terrain-following vertical levels.

Case Study 1: Habitat Compression Index

The Habitat Compression Index (HCI) is a regionally resolved measure of cool thermal habitat along the U.S. West Coast; the index presented here monitors surface water conditions off California (35-40°N). The HCI is used to assess the degree to which upwelling habitat (indicated by cool water) is compressed against the coast, as nutrient-rich upwelled waters attract whales seeking enhanced foraging opportunities. The HCI was calculated as the number of grid cells with SST lower than a monthly SST threshold within 150 km of the coastline. The HCI was normalized by the total number of grid cells of the 150 km domain to scale values from 0 to 1. Monthly SST thresholds are the mean monthly SST from 1981-2010 from the coast to 75 km offshore. Low HCI values represent high compression, or reduction of cool thermal habitat, and are the primary interest to resource managers tasked with mitigating whale entanglement risk. The long-term mean of the HCI is used to identify a high compression event (i.e. values below the mean.

Case Study 2: TOTAL Tool

The Temperature Observations to Avoid Loggerheads (TOTAL) tool monitors anomalously high SST in the Southern California Bight (31-34°N, 120-116°W) as an indicator of turtle bycatch risk and to recommend potential implementation of a fishery closure. TOTAL was calculated as the six-month rolling mean of SST anomalies in the Southern California Bight domain. The spatial closure is potentially enacted during three months of the year (June, July, August) based on SSTA of the preceding six months. If SSTA exceeds a threshold, calculated as the minimum monthly anomaly value preceding three historical closure periods (Aug 2014, Jun-Aug 2015, & Jun-Aug 2016), a closure is recommended.

Skill assessment

Forecast skill of each management tool was assessed by comparing observed and forecast values using three metrics: (1) correlation coefficient, which provides a statistical measure of the strength of a linear relationship between observed and forecast values; (2) forecast accuracy, which indicates the fraction of correct forecasts; and (3) the Symmetric Extremal Dependence Index (SEDI) which has several properties that make it well suited to quantifying skill for rare events. Details and equations for metrics are described in the manuscript.