Evaluation of the Change in Some Meteorological Variables Measured with the Automatic Station at the Yabu Meteorological Station, Cuba

Introduction

Meteorological stations change location over time, and it becomes necessary to evaluate measurements under new conditions. The Yabu meteorological station ceased its usual operations in 2023, month 8, day 31, and from then on, data collection began using an automatic station. Therefore, it is necessary to evaluate the record- ings obtained with this new technique. Therefore, the objective of our work will be to assess this change in the main variables, such as extreme temperatures (maximum and minimum), as well as the maximum rainfall within 24 hours.

Materials and Methods

For this work, the Regressive Objective Regression Methodolo- gy (ROR) was used to model the three meteorological variables, namely, extreme temperatures, maximum, minimum, and maxi- mum rainfall in 24 hours [1-3]. For this purpose, a step variable was designed, which takes the value of zero before the change, that is, before 2023, month 8, day 31, and takes the value of 1 after this date, which corresponds to the new automatic station. It should be noted that gaps are observed in the automatic data collection due to the lack of electricity at some stages, which complicates the study. However, the results obtained were good.

Results and Discussion

The model for minimum temperature explains 99.3% of the vari- ability with an error of 2.3547. Fisher's F value is 92,717, signifi- cant at 100%. The model is as follows the Table 1 below. As can be seen, the model depends on temperatures regressed over 11 years.

In particular, Step.2023.08.31 represents a drop of 1.39°C with the use of the automatic station from that date onwards. It can be seen, for example, that the variable step4097 represents a drop of 4.4°C and is a case that has occurred in the station's history. The trend is positive but very small and highly significant at 100%.

Table 1: Co-efficients^{a, b}

In the case of the Maximum Temperature, Table 2., the explained variance is 99.7 with an error of 2.37, Fisher's F is 210068 signif- icant at 100%. The trend is positive but very small and highly sig- nificant at 100%, as in the Minimum Temperature. The new station represents a drop of 1.48 ºC, then it is corroborated that with the automatic station both the maximums and minimums are below what were measured with the previous station where meteorolo- gists were used to measure.

Table 2: Coeficientes^{a, b}

For the 24-hour maximum rainfall (Table 3), a model was obtained that explains 50.7% of the rainfall with an error of 9.7 mm. Fisher's F is 143, significant at 100%. The trend is positive but very small and highly significant. The new weather station reports a 2.9 mm decrease in rainfall.

Table 3: Coeficientes^{a, b}

The model for these variables is very long-term (11 years in ad- vance). It was analyzed that the short-term model results in small- er errors, and the impact of using the automatic station results in small, non-significant parameters. Therefore, it can be concluded that the mean values from the Yabu station can be used, combined with the new data, at least in the short term.

Conclusions

• As can be seen, the model depends on temperatures and precipitation returned over 11 years. In particular, the new automatic station represents a 1.39°C drop in minimum temperature. The trend is positive but very small and highly significant at 100%.

• In the case of maximum temperature, the trend is positive but very small and highly significant at 100%, as is the case with minimum temperature. The new station represents a drop of 1.48°C. It is then confirmed that with the automatic station, both the maximum and minimum temperatures are below what was measured with the previous station used by meteorologists.

• For the maximum rainfall over 24 hours, the trend is positive but very small and highly significant. The new meteorological station reports a 2.9 mm decrease in rainfall.

• In the short term, the errors are smaller, and the impact of using the automatic station results in small and non-significant parameters. Therefore, it can be concluded that the average values from the Yabu station can be used, combined with the new data, at least in the short term.

References

1. González, F. M. W. (2022). Methodology of The Objective Regressive Regression In Function of The Prognosis For Deaths, Critical, Severe, Confirmed And New Cases ofCovid-19 In Santa Clara Municipality and Cuba. Research Review, 3(01), 604-612.
2. Osés, R.R., Fimia, D.R., Osés, L.C., y Jerez, P.L.E. (2022b).Forecasts for deaths, critical cases, serious, confirmed and new cases of COVID-19 in the municipality of Santa Clara and Cuba using the Regressive Objective Regression methodology. UO Medical Affairs. 1(1), 28-39.
3. Rodríguez, R. O., Fimia-Duarte, R., del Valle Laveaga, D., Martin, M. O., Cabrera, N. R., Ferrer, Y. Z., ... & González, F. M. W. (2022). Mathematical Modeling and Its Applicability from Natural Disasters to Human Health.