Trend Pattern of Heavy and Intense Rainfall Events in Colombia from 1981–2018: A Trend-EOF Approach

Abstract

The Andes mountain range divides Colombia into various climatic regions over the country, as the Andean, Caribbean, Pacific, Amazon, and Orinoco regions. Given this scenario, knowing the current change in total precipitation and their extremes values are relevant. In this study, the main goal is to assess the spatio-temporal trends of heavy and intense rainfall at a seasonal scale during the last 38 years (1981–2018) using the trend empirical orthogonal function (TEOF). An increase in maximum precipitation during five consecutive days (RX5day), Simple daily intensity index (SDII), and the number of days with precipitation above 20 mm (R20mm) and 30 mm (R30mm) during December–February and March–May was observed in most of the Colombian territory, except for the Amazon region for RX5day. A decrease in total rainfall in June–August was observed in the Andean, the Caribbean, and southern Pacific regions, while, in the northern Pacific, it increased, consistent with the trend patterns of RX5day, SDII, and R20mm. During September–November, there was a reduction in rainfall in the Amazon region and the South Pacific, and an increase in RX5day, SDII, R20mm, and R30mm in the Andean, the Caribbean, and North Pacific regions. The TEOF showed more pronounced spatial trend patterns than those obtained with the traditional Mann–Kendall test. The findings offer a better understanding of the climate extremes impacts in tropical latitudes and help planners to implement measures against the future effects of climate change.

Analysis of Anomalies Due to the ENSO and Long-Term Changes in Extreme Precipitation Indices Using Data from Ground Stations

“…This research is in agreement with the information reported in other studies [33,35,[58][59][60][61][62][63][64]69,72] regarding the influence of ENSO on Colombian precipitation: that is, more rain in NAY, and drier conditions in NOY, such as can be observed in the results obtained from PRCPTOT in terms of positive and negative anomalies during the ENSO phases. In the same way, other studies that have been carried out with satellite precipitation estimates in the northern region of South America show positive anomalies during NAY [87,88].…”

“…Understanding the spatiotemporal changes in extreme precipitation on a regional scale is of great importance for policymakers and the general public [68], because, in the current research, there is still uncertainty about the influence of ENSO [37,48,69]. On the other hand, for the Colombian territory, different studies have been carried out on the long-term trends of the average precipitation values and temperature, finding changes in precipitation between −4% and 6% per decade, although the areas with decreasing/increasing trends are different [58,59,61,64,70,71].…”

The South American precipitation trends under (or not) El Niño‐Southern Oscillation influences and relationship with large‐scale circulation

“…These time series are regressed back to the anomalies to produce the physical trend mode patterns. Therefore, if v is an eigenvector of HQ , the corresponding time series in physical space is given by w = HXv ; and the associated spatial pattern in physical space is the correlation map obtained by correlating w and the precipitation anomaly time series at each grid point (Cerón et al, 2022; Hannachi, 2007). For conciseness, the trend EOF and the corresponding PC time series in the physical space are referred to TEOF and TPC.…”

Satellite derived trends and variability of CO2 concentrations in the Middle East during 2014–2023

“…One of such methods is based on the principal component analysis, in which trend empirical orthogonal functions (TEOFs) and trend principal components (TPCs) are extracted. This technique, proposed by Hannachi (2007) and used in studies such as Barbosa and Andersen (2009) and Ceron et al (2022), is based on the eigen-analysis of a matrix with the inverted ranks instead of the actual values. The procedure followed can be summarized in three steps: i) the XCO 2 gridded daily data, OCO-3 L3 product, for 2015-2021 is used to construct a matrix X n×p (x ij ), where i = 1, 2, .…”