The spatiotemporal variability of precipitation in regions of complex mountainous terrains constitutes one of the most challenging research topics of geosciences. This paper explores hourly precipitation data from a set of 25 stations spanning the period 1998 to 2005 within northwestern Colombia, in the Aburrá Valley and the neighboring San Nicolás plateau (75.16°W−6°N and 75.66°W−6.6°N) which accounts for a land area of ~4,000 km². Our aim is to identify the main features of the diurnal cycle of precipitation over this complex terrain. We found that the average diurnal cycle of rainfall in the study region is bimodal at regional scale although it results from the superposition of two unimodal diurnal cycles shifting its phase throughout the seasons of the year. From October to April, average diurnal rainfall peaks in the afternoon hours (13:00–16:00 LST) but from May to September, the phase of the diurnal cycle changes to midnight hours (22:00–02:00 LST). Three low-level jets (LLJs), namely Caribbean, CHOCO, and the so-called Corriente de los Andes Orientales (CAO), are relevant to explain the seasonal shift of the diurnal cycle given their modulation of the seasonal variation of moisture sources and transport over this region. During June-July-August, moisture from afternoon evaporation processes at the bottom of the inter-Andean Magdalena Valley, located at the east of the study region, is transported by anabatic and easterly trade winds and contribute to explain the midnight and early morning peak. The life cycle of convective processes influences the orographic nature of rainfall distribution and timing in the region since deep convective cores are related with the afternoon peak, whereas wide convective cores with the early morning peak.

In this study, we validate precipitation estimates remotely sensed by the Tropical Rainfall Measuring Mission (TRMM) at monthly and seasonal timescales, during the period 1998–2015, by calculating and analyzing diverse error metrics between the 3B43 V7 product and in situ measurements from 1,180 rain gauges over Colombia, of which at least 987 are fully independent of TRMM. We explore the existence of spatiotemporal patterns to assess the performance of 3B43 V7 over the five major natural regions of Colombia: Caribbean, Pacific, Andes, Orinoco and Amazon. The results show that 3B43 V7 product is able to capture the phase of the annual cycle of monthly mean precipitation, but the performance is not good for the amplitude, in particular over the Andes and Pacific regions owing to complex climatic and topographic conditions. In general, 3B43 V7 exhibits good performance in the low‐lying and plain Amazon, Orinoco and Caribbean regions. Over the Andes region, characterized by complex topography, overestimation errors are identified [root mean squared error (RMSE) ≥83.59 mm·month⁻¹and relative bias (BIAS) ≥4.69%], whereas the extremely wet rainfall regime of the Pacific region is largely underestimated (RMSE ≥253.52 mm ·month⁻¹and BIAS ≤−11.75%). These errors are greater during the wet seasons when the metrics reach worse scores than those reported in similar studies worldwide. Occurrence analyses showed that 3B43 V7 misses very frequent light rainfall events and less frequent but very heavy storms, which contribute to the overall underestimation (overestimation) observed over the Pacific (Andes) region. The error characteristics identified and quantified in this study confirm the well‐documented limitations of remote precipitation sensing and constitute a warning about major challenges that complex climatic and physiographic features can impose on satellite rainfall missions.

We present a simplified overview of land-atmosphere feedbacks at interannual timescales over tropical South America as structural sets of linkages among surface air temperature (T), specific humidity at 925 hPa (q₉₂₅), volumetric soil water content (Θ), precipitation (P), and evaporation (E), at monthly scale during 1979–2010. Applying a Maximum Covariance Analysis (MCA), we identify the modes of greatest interannual covariability in the datasets. Time series extracted from the MCAs were used to quantify linear and non-linear metrics at up to six-month lags to establish connections among variables. All sets of metrics were summarized as graphs (Graph Theory) grouped according to their highest ENSO-degree association. The core of ENSO-activated interactions is located in the Amazon River basin and in the Magdalena-Cauca River basin in Colombia. Within the identified multivariate structure, Θ enhances the interannual connectivity since it often exhibits two-way feedbacks with the whole set of variables. That is, Θ is a key variable in defining the spatiotemporal patterns of P and E at interannual time-scales. For both the simultaneous and lagged analysis, T activates non-linear associations with q₉₂₅ and Θ. Under the ENSO influence, T is a key variable to diagnose the dynamics of interannual feedbacks of the lower troposphere and soil interfaces over tropical South America. ENSO increases the interannual connectivity and memory of the feedback mechanisms.

The Colombian Andes (CA) are located over the northwestern corner of tropical South America (NW‐TropSA), where land and atmosphere interchange moisture and energy in a complex way owing to the orographic influence on the recycling of moisture over land. We aim to understand where and when water vapor evaporated from land turns into rainfall through moisture recycling, using the Water Accounting Model‐2layers (WAM‐2), an offline model to track atmospheric moisture forced with data from ERA‐5 at its native 0.25° resolution during 1980–2020. We define the spatiotemporal distribution of sources (high evaporation recycle ratios, ϵ _C ) and receptors (high precipitation recycle ratios, ρ _C ) of continental moisture at diverse timescales, including monthly, seasonal, annual and interannual (ENSO). Referring to the regional runs of WAM‐2 over NW‐TropSA (4°S–12°N/80°W–66°W), at elevations above 1,000 m.a.s.l., the CA has a mean annual ρ _C of 11% (ranging 6%–16%) and a mean annual ϵ _C of 35% (ranging 27%–40%). Moisture recycling in the CA exhibits a strong annual cycle over the region. The seasonal dynamic of moisture recycling shows two clear‐cut sources of moisture: the eastern foothills of the Eastern and Central ranges of the CA. Both foothills are also regions of high rainfall, although moisture recycling mechanisms differ. Sources of continental moisture grow spatially during September‐October‐November and March‐April‐May. The seasonal availability of moisture recycled coincides with regions where orography interacts with low level jets sourcing humidity. At interannual timescales, sources and receptors of continental moisture in the CA are modulated by the extreme phases of ENSO.

Air pollution includes particle-bound polycyclic aromatic hydrocarbons (PAHs), which eventually reach the placenta, triggering adverse perinatal outcomes through long-term exposure. Lately, air pollution has increased over the Metropolitan Area of Medellin-Colombia (MAMC), but its effects on pregnancy are still unknown. In this research, a real-time analysis of total airborne PAHs was made using a photoelectric sensor for residential places influenced by the industrial and traffic sources affecting the southern and northern MAMC during the second peak of the bimodal tendency for PM2.5 emissions in this region. Additionally, individual PAHs were analyzed by GC/MS coupled with pressurized hot water extraction methodology. The data were applied using an inhalation intake model to assess pregnancy exposure. The average concentration of PAHs in the southern MAMC was three times higher than in the northern MAMC, where the missed abortion rate has been 1.4 times higher according to the database. Previous research indicates that PAHs act as endocrine-disrupting chemicals (EDCs) during pregnancy and that even heavy congeners could reside in umbilical cord blood. Finally, the annual series of missed abortion rates in the MAMC exhibited a significant correlation with the annual average levels of PM2.5, which were associated with PAHs through correlation analysis (r² = 0.69, p < 0.01). While this significant correlation does not imply causality, our results suggest an important connection between both variables, opening a gap for a deeper understanding of how regions with high PAH convergence influence the missed abortion rates in MAMC.