**Geographic Setting.**

La Jolla Pier is on the Pacific coast about 21 km NNW of San Diego, California, USA. The concrete Pier extends 332m out from Scripps Beach which is adjacent to the UCSD Scripps Institution of Oceanography.

The climate for San Diego is classified as Mediterranean with statistics as follows:

. Jan Feb Mar Apr May Jun

Average high in °C: 18.4 18.3 18.7 19.7 20.3 21.6

Average low in °C: 9.4 10.4 11.8 13.3 15.2 16.7

Av. precipitation in mm: 50 58 46 20 3 2

Days with precipitation: 7 7 7 5 2 1

Hours of sunshine: 216 212 262 242 261 253

. Jul Aug Sep Oct Nov Dec

Average high in °C: 23.7 24.7 24.4 22.7 20.6 18.2

Average low in °C: 18.6 19.3 18.4 15.9 12.0 9.1

Av. precipitation in mm: 1 1 4 14 26 39

Days with precipitation: 0 0 1 3 4 6

Hours of sunshine: 293 277 255 234 236 217

San Diego weather averages:

Annual high temperature: 20.9°C

Annual low temperature: 14.2°C

Average temperature: 17.55°C

Average annual precipitation – rainfall: 264 mm

Days per year with precipitation – rainfall: 43 days

Annual hours of sunshine: 2958 hours

**Daily flask CO2 data analysis.**

Measurements of the CO2 concentration have been made at La Jolla Pier as part of the Scripps Institute atmospheric CO2 sampling programme. Flask samples were collected from the station which was located at latitude 32.9°N, longitude 117.3°W, elevation 10 metres. A data file of the measurements, daily_flask_co2_ljo.csv, was downloaded from the Internet at:

https://scrippsco2.ucsd.edu/data/atmospheric_co2/ljo.html

It covered the period 23 October 1968 to 17 October 2019. The time series is show below in Figure 1. Due to the incidence of noise spikes and large sampling intervals, this analysis was confined to data from 31 March 1972 onwards. That data was reduced to 1223 values by removing values flagged as rejected by the provider for a variety of reasons listed in the introduction to the file.

The resulting data file had sample intervals ranging from 0.6 days to 113.86 days with a mean interval of 14.21 days and a median of 9.95 days. Linear interpolation was applied to the time series to create data at a fixed interval of 10.146 days, being one thirty-sixth of an average year of 365.25636 days. The final, fixed interval data series of 1712 values is shown in Figure 2. Calculation of the linear trend of the data gave a rate of increase of 1.77 ppm per annum for the irregularly spaced data compared to 1.73 ppm per annum for the fixed interval data string.

The seasonal variation in the CO2 concentration is clearly evident on both Figures 1 and 2. The range was estimated to be from 7.7 ppm to 14.7 ppm in amplitude with the lessor value probably compromised by the irregular sampling interval in the original time series. That is, the seasonal variation is of the order of 6 or 7 times the annual rate of increase indicating that biological sources were a major component of the CO2 in the atmosphere.

The annual rate of change of the CO2 concentration was calculated by taking the difference in values 36 intervals apart. A linear trend was fitted to the resulting data series and removed to give the detrended, annual rate of change of the CO2. This was shown to not have a Normal statistical distribution by the Cramér-von Mises test with a statistic of 0.72 and probability of Normality of 0.0115. An Autocorrelation test gave zero probability of it being a random series with a Ljung-Box statistic of 2577.5 so standard Normal Distribution statistical methods were not applicable.

Visual inspection of the original time series gave the average date for the annual concentration minimum to be 27 August and maximum 02 May. The sharp decline in the concentration commenced in Spring time at the end of the winter rainfall as the temperature was rising and finished on average 118 days later at the peak of the monthly temperatures about two months prior to the start of the rainy season. This is the direct opposite of the UN IPCC claim that increasing CO2 concentration causes an increasing temperature. A major contributing cause may be the annual bloom of phytoplankton in the shallow waters along the seashore together with other seasonal biological causes.

A low-pass filter, cutoff frequency 1.25664, that is Pi/2.5, equivalent to a wavelength of 12.75 days, was applied to the detrended annual rate of change of the CO2, a Hamming Window was applied across the whole of the time series, the series was padded with zeros at both ends to give a length of 2048 values and the Fourier spectrum was calculated. The resulting spectrum is shown in Figure 3.

The significant periods were:

Days Amplitude Source

1298.69, 4.6243, 3.6 year El Nino average period,

364.544, 8.2419, Earth orbital period 365.25636 days,

230.878, 2.1914, twice Mercury synodic period being 231.8 days,

182.272, 6.8073, half Earth orbital period being 182.628 days,

149.489, 1.3274,

107.108, 1.7202, Mercury synodic period of 115.9 days,

81.168, 1.4105, 3 x the Moon’s draconic period being 81.9 days,

66.175, 1.7373,

57.880, 1.1134, twice the Moon’s draconic period being 54.6 days,

51.180, 0.8628,

43.471, 0.4975,

39.579, 0.5324,

36.519, 0.3535,

32.723, 0.3403,

29.898, 0.3136, Moon synodic period 29.53059 days,

27.779, 0.2605, Moon anomalistic period 27.55455 days.

25.877, 0.1841,

24.303, 0.1324,

22.884, 0.0740,

21.735, 0.1072,

20.492, 0.0849.

The appearance of periodicities relevant to the Solar System in the time series for the annual rate of change of the CO2 concentration implies that the Sun is a major factor in the generation of the atmospheric CO2 via photosynthesis and temperature change. This supports the earlier thesis, from the evidence in the seasonal variation, that a major portion of the atmospheric CO2 must be of biological origin.