Background <p>Dengue transmission is modulated by climatic conditions through their effects on vector ecology and viral dynamics; however, these relationships are context-dependent and operate across multiple temporal scales. Evidence from ecologically diverse regions remains limited, particularly in settings where contrasting climatic regimes coexist within the same geographic area. This study aimed to characterize interannual climate variability and to evaluate immediate and lagged associations between temperature, precipitation, and dengue incidence in two climatically distinct cantons of southern Ecuador (Loja and Macará) from 2019 to 2024.</p> Methods <p>An ecological time-series study was conducted using weekly dengue case counts from the national surveillance system and meteorological data (minimum temperature, maximum temperature, and accumulated precipitation). Interannual variability was assessed using one-way ANOVA with Games-Howell post hoc comparisons. Climate-dengue relationships were first explored using Spearman correlation across lag structures from 0 to 16 weeks, complemented by Pearson correlation and covariance to describe linear association and joint variability. Temporal dependencies and potential non-linear patterns were examined using cross-correlation function (CCF) analysis and locally estimated scatterplot smoothing (LOESS). Multivariable generalized linear models with a negative binomial distribution were fitted separately for each canton to estimate independent and lagged effects of climatic variables, accounting for overdispersion in dengue case counts.</p> Results <p>Both cantons exhibited significant interannual variability in temperature and precipitation, with greater climatic amplitude in Macará. Dengue incidence increased markedly in recent years, with earlier and more pronounced outbreaks in Macará compared with Loja. In exploratory analyses, minimum temperature and accumulated precipitation showed the most consistent lagged associations with dengue incidence, although these patterns varied by canton and year. In Macará, precipitation showed a strong and consistent association with dengue incidence across all lag models (IRR ≈ 1.01 per mm/week; all <i>p</i> &lt; 0.001), representing the most consistent climatic predictor across lag periods. This finding was consistent with cross-correlation analysis, which identified precipitation as the strongest leading climatic signal (peak <i>r</i> = 0.48 at Lag 1), with significant positive correlations across multiple lag periods, supporting its role as a robust temporal predictor of dengue incidence in Macará. Mean temperature showed weaker and lag-dependent associations, with no consistent concurrent effect and positive associations emerging only at longer lag periods (12–16 weeks). However, cross-correlation analysis indicated that the temperature signal was weak (peak <i>r</i> &lt; 0.15), suggesting a secondary and less consistent role compared with precipitation. In contrast, Loja presented heterogeneous and less temporally structured associations. However, given the low case burden and wide uncertainty, these findings should be interpreted as exploratory and hypothesis-generating rather than confirmatory.</p> Conclusions <p>Climate-dengue relationships in southern Ecuador are heterogeneous and shaped by local climatic context and temporal dynamics. Precipitation was the climatic variable most consistently associated with dengue incidence, particularly in semi-arid settings. In temperate highland settings, observed associations may largely reflect shared seasonal patterns rather than independent climate-related effects. These findings highlight the importance of incorporating lag-specific and context-sensitive climate indicators into epidemiological surveillance and early warning systems and support the use of multivariable approaches to better characterize the combined and delayed associations between climatic factors and dengue incidence.</p> Trial registration <p>Not applicable.</p>