Historical Seawater Data - Moss Landing Marine Labs
Doument updated 06/26/24

--Data Disclaimer-- 
Moss Landing Marine Laboratories (MLML) provides these data "as is", 
with no warranty, expressed or implied, of the data quality or consistency.
Historical Seawater Data - Moss Landing Marine Labs

--Data Disclaimer-- 
Moss Landing Marine Laboratories (MLML) provides these data "as is", 
with no warranty, expressed or implied, of the data quality or consistency.
It is provided without support and without obligation on the part of MLML to assist in its use, correction, 
modification, or enhancement. For use in publication, authors should obtain written permission from the director of MLML, 
and acknowledge MLML as the data source in those publications.

--Contacts--
Data Manager: Steven Cunningham (Steven.cunningham@sjsu.edu)

--Data Citation--
[MLML] Moss Landing Marine Labs. 2014. Historical Seawater Data (insert date range) [Internet]. [cited (insert date of download)]. Available from: 

--Brief Narrative--
Seawater data observations are collected from raw seawater drawn through an intake pipe. Prior to October 31, 2013, the intake opening
was set at a depth of ~18.3 m (60ft) below mean lower low water (MLLW). Due to sediment build-up around the pipe, the intake opening was raised to ~16.6m (54.4ft) below MLLW on October 31, 2013.
The intake opening is located at 36.8025N and 121.7915W
The seawater sensors are cleaned of biofouling agents on a weekly/twice-weekly interval, though some data drift can be observed in transmission, beam attenution, and fluorescence.

Full maintenance Log available at: https://docs.google.com/spreadsheets/d/1KSEeCiINy9HNK2eE12wzfdXaGwKs50c7/edit?usp=sharing&ouid=115377605326757725344&rtpof=true&sd=true

Sensors and Calibrations/Redeployments
Seabird SBE19 CTD: First deployed Sept 3, 2010; swapped with calibrated CTD on Aug 26, 2022; 
AADI Oxygen Optode: First deployed Sept 3, 2010; redeployed following calibration on Aug 13, 2012; 
C-Star Transmissometer (10cm): First deployed Sept 3, 2010; calibrated on January 25, 2012; 
WetStar Fluorometer: First deployed Sept 3, 2010: calibrated on Nov 13, 2023; 
Honeywell Durafet III: First deployed May 4, 2012; replaced 2022; Last calibrated Oct 12, 2023
Turner C-Sense: First deployed November 2015. calibrated in Mar 2017.

--QAQC--
Quality assurance and quality control analysis based on standards set forth by Integrated Ocean Observing Systems' (IOOS) Quality Assurance for Real-Time Oceanographic Data (QARTOD): http://www.ioos.noaa.gov/qartod/welcome.html
Flags as follows---
 1 = Pass; Data have passed critical real-time quality control tests and are deemed adequate for use as preliminary data
 2 = Not Evaluated; Data have not been QC-tested, or the information on the quality is not available
 3 = Suspect or of High Interest; Data are considered to be either suspect or of high interest to data providers and users. They are flagged suspect to draw further attention to them by operators
 4 = Fail; Data are considered to have failed one or more critical real-time QC checks. If they are deciminated at all, it should be readily apparent that they are not of acceptable quality.
 9 = Missing data; Data are missing/NULL/NAN

--Quality Checking: Data were checked against the following thresholds--

Temp/Optode Temp: C
 Sensor range test -5<temp<35 
 User range test 4<temp<22
 Spike test: value is more than threshold values from the mean of the previous 3 values: low threshold = .4, high threshold = 0.8
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 

Conductivity : (conductivityRatio)
 Sensor range test 0<cond<7
 User range test 0.75<cond<1.1
 Spike test: value is more than threshold values from the mean of the previous 3 values: low threshold = .05, high threshold = 0.1
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 



Salinity: (PSU)
 Sensor range test 20<sal<50
 User range test 31<sal<35
 Spike test: value is more than threshold values from the mean of the previous 3 values: low threshold = 0.2, high threshold = 0.7
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 



Fluorescence: (ug/L)
 Sensor range test 0.03<Fluor<75
 User range test 0.03< Fluor <20
 Spike test: value is more than threshold values from the mean of the previous 3 values: low threshold =0 .2, high threshold = 0.5
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 



Beam Attenuation: (m?¹)
 Sensor range test 0<beam<100
 User range test 0<beam<50
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 



Transmission: (%)
 Sensor range test 0<trans<100
 User range test 0<trans<100


Dissolved Oxygen: (uM)
 Sensor range test 0<DO<500
 User range test 0<DO<400
 Spike test: value is more than threshold values from the mean of the previous 3 values: low threshold =50, high threshold = 100
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 



Dissolved Oxygen Saturation:
 Sensor range test 0<sat<120
 User range test 0<sat<120
 Spike test: value is more than threshold values from the mean of the previous 3 values: low threshold =10, high threshold =20
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 



pH:
 Sensor range test 0<cond<14
 User range test 7<cond<10
 Spike test: value is more than threshold values from the mean of the previous 3 values: low threshold =0 .2, high threshold = 0.5
 Flatline test: 3 consecutive values without change
 Data gap: If there is more than 900 seconds between the current and previous point, the data will receive a flag of 2 as spikes can’t be properly evaluated. 



--Data Units--
Temp/Optode temp: degrees Celcius
Conductivity: conductivity ratio
Salinity: PSU-78
Fluorescence: micrograms per liter
Beam Attenuation: meters^-1
Transmission: percent (%)
Dissolved Oxygen: micromole per liter
Dissolved Oxygen Saturation: percent (%)
pH: pH units
Tide: feet relative to mean lower low water (obtained from http://tidesonline.nos.noaa.gov/data_read.shtml?station_info=9413450+Monterey,+CA)
co2: microatmospheres (ppm)
Nitrate: micromoles per liter
Nitrogen: miligrams per liter

Functions:

The functions are called to return a flag value. The flags are appended to a list for each variable. The maximum flag value is applied to the datapoint.


def sensorRange( val , t_sensor_min , t_sensor_max ):
    # val = current data point
    # t_sensor_min = minimum scaled sensor value
    # t_sensor_max = maximum scaled sensor value
    if val < t_sensor_min or val > t_sensor_max:
        flag = 4
    elif math.isnan(val) == True:
        flag = 9
    else:
        flag = 1
    return flag

def userRange( val , t_user_min , t_user_max ):
    # val = current data point
    # t_user_min = minimum scaled sensor value
    # t_user_max = maximum scaled sensor value
    if val < t_user_min or val > t_user_max:
        flag = 3
    elif math.isnan(val) == True:
        flag = 9
    else:
        flag = 1
    return flag

def flatLine( val , comp , eps , rep_cnt_fail=6 , rep_cnt_suspect=3 ):
    # val = current datapoint
    # comp = comparison values (n). Must be n>rep_cnt_fail
    # eps = tolerance value
    # rep_cnt_fail = number of repeated equal values to fail (default=6)
    # rep_cnt_suspect = number of repeated equal values to suspect (default=3)
    fail=rep_cnt_fail
    suspect=rep_cnt_suspect
    eps=eps
    if len(comp) < fail:
        raise ValueError("Not enough values to compare. Number of values in comp must be greater than rep_cnt_fail and rep_cnt_suspect.")
    

    count = 0
    flag=[]
    for i in np.arange(len(comp)):
        test = abs(val - comp[i])
        if test<=eps:
            count+= 1
    if count >= suspect:
        flag.append(3)

    count = 0
    for i in np.arange(len(comp)):
        test = abs(val - comp[i])
        if test<=eps:
            count+= 1
    if count >= fail:
        flag.append(4)

    else:
        flag.append(1)

def spike_ref(current_temp, current_time, compList, compTimes, thrshld_low, thrshld_high, time_gap_threshold):
    # Ensure compList has sufficient length
    if len(compList) < 2 or len(compTimes) < 2:
        return 2  # Not enough data

    # Calculate the time gap
    last_time = compTimes[0]
    time_gap = current_time - last_time

    # Check if time gap exceeds the threshold
    if time_gap > time_gap_threshold:
        return 2  # Low flag due to time gap

    # Calculate the baseline mean from historical data
    mean = np.nanmean(compList[:3])

    # Calculate the spike reference value (difference between current temp and baseline mean)
    spk_ref = abs(current_temp - mean)

    # Determine the flag based on thresholds
    if spk_ref >= thrshld_high:
        return 4  # Spike detected
    elif spk_ref >= thrshld_low:
        return 3  # Near spike
    else:
        return 1  # No spike