Climate variability in the Apacheta micro-basin has an impact on vegetation behavior. The objective is to analyze the annual trend, anomalies and predict the behavior of vegetation cover (CV) with Landsat images and the MOD13Q1 product in the Apacheta micro-basin of the Ayacucho Region. For this purpose, the CV was classified and validated with the Kappa index (p-value=0,032; 0.05), obtaining a good agreement between the values observed in situ and the estimated in the Landsat images. The CV data were subjected to the Lilliefors normality test (p-value=0,0014; 0,05) indicating that they do not come from a normal distribution. CV forecasting was performed with the auto.arima, forecast and prophet packages, in R, according to the Box-Jenkins and ARIMA approaches, whose two-year future scenario is acceptable, but with higher bias. The results show an anual increasing CV trend of 3,378.96 ha with Landsat imagery and 3,451.95 ha with the MOD13Q1 product, by the end of 2020. The anomalies and the CV forecast also show a significant increase in the last 9 years, becoming higher in the forecasted years, 2021 and 2022.

Vulnerable road users in informal urban environments confront a distinct set of hazards that standard computer vision datasets are ill-equipped to represent: artisanal speed bumps constructed without regulatory compliance, deteriorated road markings, and the mototaxi—a three-wheeled motorized vehicle that constitutes the primary informal transport mode in intermediate Andean cities yet is absent from all major international repositories. This paper presents QHAWAY—from Quechua qhaway, a transitive verb meaning “to look; to observe”—an Advanced Driver Assistance System (ADAS) predicated on instance segmentation, monocular distance estimation via the pinhole camera model, and Time-to-Collision (TTC) computation, developed for the road environment of Ayacucho, Peru (2761 m a.s.l.), a city recognised by UNESCO as a Creative City of Crafts and Folk Art since 2019. A hybrid dataset comprising 25,602 images with 127,525 annotated instances across 12 classes was assembled by combining an original local collection of 4598 images (10,701 instances) captured through four complementary acquisition methods across the five urban districts of the Huamanga province with three established international datasets (BDD100K, BSTLD, RLMD; 21,004 images, 116,824 instances). A three-phase progressive training strategy with monotonically increasing resolution (640, 800, and 1024 pixels) was evaluated as an ablation study. A multi-architecture comparison spanning YOLOv8L-seg and the YOLO26 family (nano, small, large) identified YOLO26L-seg as the best-performing model, attaining mAP50 Box of 0.829 and mAP50 Mask of 0.788 at epoch 179. The integration of ByteTrack multi-object tracking with the pinhole equation $D=(H_{\mathrm{real}}\times f)/h_{\mathrm{px}}$ delineates operational risk zones aligned with the NHTSA forward collision warning standard (danger: <3 m; caution: 3–7 m; TTC threshold ≤ 2.4 s). The system sustains processing rates of 19.2–25.4 FPS on an NVIDIA RTX 5080 GPU. A systematic field survey established that 96% of the audited speed bumps fail to comply with MTC Directive No. 01-2011-MTC/14, constituting the first quantitative record of informal road infrastructure non-compliance in the Andean region. Validation was conducted under naturalistic driving conditions without staged scenarios. Grad-CAM explainability analysis, encompassing three complementary visualisation algorithms (Grad-CAM, Grad-CAM++, and EigenCAM), confirmed that model attention concentrates consistently on safety-critical objects.

El estudio, expone los descriptores de la precipitación observados en la cuenca del río Cachi, durante 1991 al 2008, por la naturaleza de su desarrollo, es cuantitativa descriptiva, no experimental, longitudinal, actuada sobre una serie cronológica de 3016 registros de la precipitación mensual en 22 estaciones hidrometeorológicas administradas por el Gobierno Regional de Ayacucho. Los datos fueron organizados en Excel y procesados en SPSS, luego se realizó el análisis de varianza, aplicando la estadística descriptiva y las comparaciones múltiples de Duncan, observándose que al menos uno de los promedios de la precipitación mensual es diferente de los otros. Los resultados revelaron que, de 1991 al 1999, la precipitación mensual promedio en la cuenca presentó pequeñas variaciones cíclicas que denota fluctuación de un nivel muy bajo a uno ligeramente alto, a excepción del año 1993 con una precipitación acentuada en un nivel alto. En los demás años del periodo, la precipitación muestra mayor variación con picos pronunciados ubicados en un nivel alto y muy alto; además, en los meses de verano la precipitación mensual media es la más alta del año, en los demás meses es muy baja, en especial en los meses de mayo a julio; asimismo, se determinó que la precipitación en la Cuenca Alta varía de 18.97 a 98.00 mm y en la Cuenca Media oscila de 43.92 a 97.90 mm. Con estas estimaciones no se cumple la relación inversa entre la precipitación y la altitud de estación.