Methodology:
High-intertidal pools are common in all three regions, allowing us to work at two spatially separated sites in each region. At each site, we identified 18 tide pools, marking each pool and quantifying physical attributes, including surface area, volume, and height on the shore. We also anchored TidbiT temperature datalogger (Onset; Bourne, Massachusetts, USA) in most pools. In the fall of 2017, we identified tide pools and conducted initial surveys and measurements. We repeated surveys every three months until immediately prior to the establishment of grazing experiments at each site in the summer of 2018. These surveys provided insights into the natural temporal variability in community and ecosystem metrics and provided baseline information on relationships between grazer abundances and producer biomass.

Sampling and analytical procedures:
During the quarterly surveys, we quantified consumer abundances, nutrient fluxes, oxygen fluxes, and photosynthetic biomass in each tide pool. Organism abundances were measured by pumping the water from each pool into a bucket, spreading a flexible mesh quadrat over the bottom of the pool, and censusing the algae and invertebrates present in each pool. Nutrient and oxygen fluxes were measured during whole-pool incubations in the dark and in the light.

Productivity:
These productivity data include information on pool attributes by date as well as measurements of oxygen concentrations and fluorescence. Incubations were ideally conducted on sunny days when underwater photon flux measurements in tide pools were > 500 micromoles of photons per square meter per second (umol photons/m2/s). During each incubation, we made initial measurements of the oxygen concentration in each pool and collected a 50 milliliter (ml) water sample from each pool and from the adjacent ocean for nutrient analysis. Then we covered each pool with an opaque plastic sheet for 30 minutes. We repeated our sample collection, then let each pool incubate in the light for another 30 minutes before collecting a third set of samples (Altieri et al. 2009, Sorte and Bracken 2015). Nutrient samples were analyzed for concentrations of NO3-, NO2-, NH4+, and PO43- using standard spectrophotometric and fluorometric methods (Wood et al. 1967, Hansen and Koroleff 1999, Holmes et al. 1999) and used to quantify fluxes of nutrients (micromoles per liter per hour (umol L-1 h-1)) as a function of consumer abundances. Quantifying NO3- and NO2- in addition to NH4+ is important because previous work has highlighted nitrification as an important process during tide pool emersion (Bracken and Nielsen 2004, Pfister 2007). Changes in oxygen concentrations (milligrams per liter (mg L-1), YSI ProODO optical oxygen meter and probe) in the dark and light will be used to calculate rates of community respiration and net and gross community production (Altieri et al. 2009, Noël et al. 2010, O’Connor et al. 2015, Sorte and Bracken 2015).

We used pulse amplitude modulated (PAM) fluorometry to quantify the minimal dark-adapted fluorescence (F0) values in each tide pool (DIVING-PAM, Heinz Walz GmbH). PAM fluorometry provides rapid, nondestructive estimates of photosynthetic biomass (Serôdio et al. 1997, Honeywill et al. 2002, Maggi et al. 2013, LaScala-Gruenewald et al. 2016), and dark-adapted fluorescence values are closely correlated with benthic chlorophyll a concentrations (Honeywill et al. 2002, LaScala-Gruenewald et al. 2016; M. Bracken, personal observation). We cross-calibrated PAM units by relating measured F0 values to extracted chlorophyll a values across multiple instrument gain settings at each site, converting all values to chlorophyll a per unit area.

This work was conducted at sites located in three regions along the California (USA) coast: (1) Bodega Head, Sonoma County (38.31°N, 123.07°W); (2) Kenneth Norris Rancho Marino Reserve and Hazards Canyon Reef, San Luis Obispo County (35.54°N, 121.09°W and 35.29°N, 128.88°W, respectively); and (3) Corona del Mar State Beach, Orange County (33.59°N, 117.87°W).

Known problems/issues:
Some environmentally-related (e.g., tides, darkness) issues caused gaps in the data. These are indicated by "nd".

seasonal_productivity.csv