We found a match
Your institution may have rights to this item. Sign in to continue.
- Title
GAP FILLING PROCEDURES OF CLIMATOLOGICAL SERIES IN THE STATE OF PERNAMBUCO.
- Authors
CAVALCANTE SILVA, LÍVIA MARIA; SIMPLICIO BEZERRA, FABIANO; DE VASCONCELOS, MARIA CATIANA; BARBALHO DA SILVA, MADSON RAFAEL; DAVINO DOS SANTOS, ANA CLÁUDIA; GUEDES CABRAL DE ALMEIDA, CERES DUARTE
- Abstract
This study aimed to compare the applicability of three methods of filling gaps in rainfall and temperature data from thirteen automatic weather stations (AWS) in the state of Pernambuco, from January to December 2019. The methods used were arithmetic mean, regional weighting, and simple linear regression. The data estimated by filling techniques have been subjected to comparison using R² and descriptive statistical analysis. The estimated data of air temperature presented R2 equal or very close to 1 for the three methods. On the other hand, the estimated data of rainfall showed values similar or closer to the real data only to regional weighting (R² = 1) and linear regression (R² = 0.99) methods. The smallest values of standard deviation (1.70) for temperature were obtained with linear regression. The regional weighting method and unfilled data showed greater uniformity for precipitation. The analyzed methods to estimate the climatic variables, air temperature, and precipitation, on a monthly scale, were efficient to fill in missing data in the evaluated AWS. The simple linear regression method is more efficient and adequate, followed by regional weighting, to fill in missing data in climate databases.
- Subjects
PERNAMBUCO (Brazil); ATMOSPHERIC temperature; AUTOMATIC meteorological stations; ARITHMETIC mean; MISSING data (Statistics); ESTIMATION theory
- Publication
Revista IRRIGA - Brazilian Journal of Irrigation & Drainage, 2021, p754
- ISSN
1413-7895
- Publication type
Article
- DOI
10.15809/irriga.2021v1n4p754-764