Ethiopia's rapid deforestation and climate variability threaten ecosystems and livelihoods, especially in the ecologically vital Metekel Zone. Despite increasing forest loss and climate instability, understanding their dynamics, causes, and mitigation from environmental and educational perspectives remains limited. This study aimed to integrate spatiotemporal analyses of deforestation and climate change with environmental sustainability education. It also sought to propose action-oriented content and pedagogical strategies for the secondary school geography curriculum in the Metekel Zone. A pragmatist mixed-methods design, integrating a longitudinal biophysical assessment with cross-sectional and convergent embedded surveys were employed to explore the educational facets. A multistage sampling technique combining purposive, simple random, and quota sampling was used to identify survey sites and subjects. Land use and land cover data were derived from Landsat imagery, climate data from the National Meteorological Agency, and crop yields from the Central Statistical Agency. In addition, both primary and secondary data were gathered through standardized tests, surveys, key informant interviews (KIIs), focus group discussions (FGDs), curriculum content analysis, and field and classroom observations. Land use land cover changes were assessed using Landsat images of 1986, 2002 and 2019 using Random Forest and Support Vector Machine classifiers, along with vegetation indices such as the normalized difference vegetation index (NDVI), soil- adjusted vegetation index (SAVI), normalized difference water index (NDWI), and leaf area index (LAI). Climate variability was analyzed using the Coefficient of Variation, Mann–Kendall trend test, Standardized Precipitation Index, and ARDL model to assess rainfall, temperature, and crop yield relationships. Educational data were analyzed with descriptive statistics, chi- square tests, t-tests, correlations, and binary logistic regression to identify factors influncing environmental knowledge, attitudes, and teachers’ implementation of Problem Based Learning (PBL). Qualitative data from KII, FGDs, curriculum reviews, and observations were thematically analyzed to contextualize and enrich the quantitative results. Findings revealed a significant forest cover decline from 51.37% in 1986 to 17.20% in 2019, propelled by agricultural expansion, resettlement, charcoal/fuelwood extraction, and weak institutional enforcement. Climate analysis showed significant warming trends, with annual maximum and minimum temperatures rising by 0.33 °C and 0.31 °C per decade, respectively, with a rainfall decline of 5.32 mm per decade. Rising temperatures have severely affected temperature-sensitive crops in Dangur and Bullen districts, whereas fluctuations in rainfall patterns have diminished agricultural yields in Dibate and Wombera districts. Curriculum review indicated moderate climate change coverage (17.8% of minimum learning competencies in upper secondary grades) but negligible deforestation integration (3.85%), prioritizing cognitive domains (60.8%) over affective (27.5%) and psychomotor (12.5%) outcomes. Students exhibited low environmental knowledge (52.2%) and unfavorable attitudes (52.7%) toward environmental protection; determinants included age (p = 0.007), gender (p = 0.003), residence (p = 0.003), parental education (p = 0.018), family income (p = 0.002), access to environmental information (p = xxiii 0.000), grade level (p = 0.000), and participation in environmental clubs (p = 0.010) significantly influenced awareness and attitudes. In contrast, social pressure, limited contextual learning, curriculum challenges, resource gaps, and restricted personal agency negatively influenced awareness and engagement. Investigation of teachers’ practices showed that geography teachers in Metekel Zone demonstrate moderately positive beliefs in the value of problem based learning (M = 2.56, t = −3.884, p < .001), though beliefs in heuristic approaches (M = 2.49, t = −4.341, p < .001) and contextualizing lessons to students’ real-life experiences (M = 2.47, t = −4.590, p < .001) remain low. Only 24–29% of teachers applied inquiry-based, real-world, and independent problem-solving strategies, while 58–61% expressed negative views toward heuristic methods. Teachers’ beliefs strongly correlated with their classroom practices (r = 0.726, p < 0.01). Binary logistic regression revealed that teachers’ adaptation of PBL (p = .006), emotional intelligence (p = .003), teachers’ perception of students’ environmental identity (p = .002), teacher’s perception of student motivation and engagement (p = .005), and teacher’s content experience (p = .008), significantly enhanced PBL implementation, whereas large class sizes (p = .038), reliance on traditional methods (p = .024), and time-intensive nature of PBL (p = .044) constrained it. Despite teachers’ conceptual support, structural and pedagogical barriers including traditional assessment systems, inadequate logistics, and extensive curriculum content limit effective PBL adoption. This study integrates spatiotemporal environmental evidence with curriculum and pedagogy to illuminate the profound effects of deforestation and climate change on livelihoods and agricultural productivity in Ethiopia. It reveals critical gaps in contextual, action-oriented learning while showing that incorporating real-world ecological data significantly enhances student engagement. By connecting biophysical data with curriculum-pedagogy gaps, this study underscores deforestation/climate impacts on livelihoods and advocates embedding real-world data to boost engagement. It calls for collaborative reforms among educators, curriculum developers, policymakers, and communities to advance sustainable environmental education in Ethiopia.