This dataset is a subset of data from the "Luminoxyscape environmental" dataset (BCO-DMO dataset 860397). In this "Luminoxyscape maximum depth" dataset, we provide the rows of information for the maximum depths only. The "Depth" column refers to the maximum depth of either oxygen limitation for vision (V50), oxygen limitation for metabolism (Pcrit), or irradiance limitation (Light_Lim). All species and limits are presented in one, combined data frame.

All corresponding environmental hydrographic cast data were collected by the California Cooperative Oceanic Fisheries Investigations (CalCOFI) long-term monitoring system. All information regarding the sampling program can be found on https://calcofi.org/about/. Briefly, CalCOFI and the Southern California Coastal Ocean Observing System (SCCOOS) samples a random array within a gridded station map during quarterly cruises and conducts vertical hydrographic casts to ~500-meters (m) depth (or 10 m above bottom where bottom depth < 500 m). More information can be found at: https://calcofi.org/index.php. All data used can be found in their original form at: https://calcofi.org/data/oceanographic-data/ctd-cast-files/.

In this analysis, we used data from CalCOFI stations restricted to an area from La Jolla, California to Point Conception (Line 80.0- Line 93.3; 32.95-34.46 °N at shoreline) and 215 kilometers maximum offshore (Station 60; 119.57-121.150 °W). We analyzed daytime casts (09:00-16:00) of both discrete bottle data and continuous CTD casts. This dataset has combined bottle and CTD casts to represent the date range 1984-2019. We used the oxygen and irradiance measurements to determine the visual luminoxyscape for each of the larval species. This range was bounded by the oxygen (partial pressure) where the pO2 would permit 50% minimum retinal function (V50; 13, 7.2, 10.2, and 6.8 kPa for larvae of Doryteuthis opalescens, Octopus bimaculatus, Metacarcinus gracilis, and Pleuroncodes planipes, respectively), and where there is sufficient irradiance for a visual response (0.0311 µmol photons m-2 s-1) for each species.

The "limits" are calculating depths where those oxygen and irradiance requirements would no longer be met. This would be the 'visual luminoxyscape depth' (VLD); these depths are the maximum depths presented in this metadata file. Oxygen values for oxygen limits for visual physiology were taken from McCormick et al., 2019.

Additionally, oxygen limits for metabolism were used to determine the depth of occurrence of the Pcrit (the oxygen below which the animal cannot maintain a constant metabolic rate). This was calculated in McCormick, 2019. The depths of occurrence for metabolic limits were determined for larvae of D. opalescens and O. bimaculatus.

Instruments:
Full details of the instrumentation on the CalCOFI CTD can be found here: https://calcofi.org/sampling-info/methods/bottle-sampling-methods/. Wherever possible, we used the cruise-corrected data in the "final" cast files provided by CalCOFI, indicating the files already QC by CalCOFI data managers.

For this analysis, we primarily used data from oxygen optodes [SBE 13 (1998-2003) and SBE 43 (2003-2019); SeaBird Scientific] and PAR irradiance sensors [QSP-200L (1998-2010); QSP-2300 (2003-2019); Biospherical Instruments, Inc.].

Code:
There are 3 R files associated with this dataset: Luminox_May21_Part1_ann.R, Luminox_May21_Part2_ann.R, and Luminox_May21_Part3_ann.R. These are further described in the attached file "Luminoxyscape_Code_Description_of_Files.txt". In general:
The code (separated into 3 parts for simplicity and organization) is the entire code dataset from the Luminoxyscape project. The code covers everything including the quality-control for data downloaded directly from CalCOFI (https://calcofi.org/data/oceanographic-data/ctd-cast-files/) in Part 1, the analysis of maximum depths in Part 2 (associated with data "Luminoxyscape maximum depth", and the final analysis for the paper (McCormick et al., Limnology and Oceanography Letters, accepted) including code for all statistics in the analysis and graphs used in the figures. The code is clean and works, but is designed to show the process of the analysis for the project, rather than direct reproducibility. Contact information is included for those who wish to ask questions. All code was run using R (v3.6.3) run through RStudio (v. 2021.09.0).