Climate change is one of the most significant challenges of the 21 st century. Ethiopia is particularly vulnerable to climate change and variability because its economy relies heavily on climate-sensitive agriculture and has low adaptive capacity. The semi-arid agricultural landscapes of Ethiopia are increasingly affected by climate change, which poses significant challenges to staple crop production, including dry beans (Phaseolus vulgaris L.), an essential source of food and income. Investigating temporal and spatial changes in climate is crucial to assessing the impacts of climate change and identifying potential adaptation options. Understanding the effects of climate change and developing adaptation strategies are necessary to mitigate risks and sustain agricultural production. Most previous studies in the northeastern regions of Ethiopia have focused on short-term field experiments, neglecting the long-term climate variability and future climate scenarios. Traditional agronomic research is often time-consuming, location-specific, and resourceintensive, which limits its ability to meet the growing demand for sustainable agricultural production. Therefore, this study was conducted to achieve several objectives:(i) analyze historical rainfall and temperature variability and trends (1991-2020), (ii) project rainfall and temperature changes and trends in the 2040s (2030–2060) and 2080s (2070-2100) under SSP24.5 and SSP5-8.5 scenarios as compared to the baseline time period (1984–2014), (iii) Analysis the spatiotemporal rainfall and temperature extremes using CMIP6 global climate model, (iv) Assess management options for dry bean (Phaseolus vulgaris L.) under the projected climate change, and (v) modeling impacts of climate change and adaptation pathways for dry beans in semi-arid northeastern Ethiopia using the CROPGRO-Dry Bean model within the Decision Support System for Agrotechnology Transfer (DSSAT v4.8). Simulations were conducted in the Habru and Tehuledari districts of northeastern Ethiopia. The calibration and evaluation field trials were conducted at Sirinka and Jari site in 2022 and 2023 consecutive main crop growing seasons. A simulation study was conducted using a crop model to assess the impacts of climate change and management practices. Three dry bean cultivars (short, medium, and long duration), planting dates (July 3, July 18, and August 3), plant densities (small, medium, and high density), and two production seasons' irrigation regimes (rain-fed and supplemental irrigation) were selected to evaluate potential adaptation strategies for improving dry bean yield in the target areas. The data collected for the study came from both primary and secondary sources. Historical climate data from 1984 to 2020 were obtained from the Ethiopian Meteorological Institute (EMI) and served as the baseline for our analysis. Additionally, gridded climate data from 1984 to 2014 were sourced from the Climatic Research Unit (CRU TS4.07) and the Global Rainfall Climatology Center (GPCCv2020). Projected climate data for the 2040s (2030-2060) and 2080s (2070-2100) were retrieved from the Earth Systems Grid Federation (ESGF). These climate projections utilized an ensemble of selected Coupled Model Inter-comparison Project Phase 6 (CMIP6) climate models under the Shared Socioeconomic Pathways SSP2-4.5 and SSP5-8.5. The ESGF provides re-gridded climate data with a spatial resolution of 0.25° × 0.25°. Soil data were collected from the experimental fields at the Sirinka Agricultural Research Center in northeastern Ethiopia and from various literature sources. Major crop data were gathered from the Early Warning Offices in South Wollo and the Oromia Special Zone in Amhara, Ethiopia. The CROPGRO-dry-bean model was calibrated and evaluated using field trials conducted over two consecutive main cropping seasons (2022 and 2023) at the Sirinka and Jari sites. Climate analysis was performed at the local level, while model calibration and evaluation relied on fieldmeasured data from the two stations. To examine historical and projected climate trends, we employed the modified Mann-Kendall test and Sen's slope estimator, both implemented in R software using the package ‘modifiedmk’ (v4.5.2). The descriptive statistics of climate characteristics were analyzed using XLSTAT-2023 and INSTAT (v3.37) software. The relationship between climate variables and crop parameters was assessed using the Pearson correlation coefficient (r). To evaluate calibrated model performance, statistical indicators such as root mean square error (RMSE), normalized RMSE (nRMSE), coefficient of determination (R²), and the index of agreement (d-statistic) were utilized. Model evaluation was performed using these same statistical metrics based on independent datasets. The crop simulation in the baseline (1984-2014), in the 2040s (2030–2060) and 2080s (2070-2100) under SSP2-4.5 and SSP5-8.5 scenarios were analyzed using analysis of variance (ANOVA). The statistical analysis was performed using SAS v9.4, and treatment means were compared using the least significant difference (LSD) at the 5% significance level. The results indicated low variability in annual and seasonal temperatures and an uneven distribution of annual rainfall, with moderate seasonal rainfall, characterized by irregular fluctuations over the historical analysis period from 1991 to 2020. Both maximum and minimum temperatures consistently increased, whereas seasonal rainfall exhibited inconsistent trends across all monitoring stations. The extremes of these rainfall patterns also showed inconsistencies and non-significant rising trends. Additionally, rainfall events showed low interannual variability, with late onset between July 8 and July 19 and early cessation dates around September 2 and 3. However, there was high intera-nnual variability in the length of the growing period, which ranged from 46.5 to 62.5 days in both the Kiremt and Belg seasons across all stations. The probabilities of experiencing dry spell lengths of 5, 7, 10, and 15 days were above 59%, 43%, 27%, and 13%, respectively, during the Kiremt season. In contrast, the Belg season showed much higher probabilities of dry spells, with 96%, 87%, 68%, and 36% for the exact durations. This suggests that the Belg season has a greater likelihood of prolonged dry spells than the Kiremt season. Dry spells lasting more than 7 days in the Kiremt season and more than 10 days in the Belg season were significant contributors to crop failures in the semi-arid regions of northeastern Ethiopia. Future projections indicate significant increasing trends in mean annual and seasonal rainfall and temperature across all periods and climate scenarios. Annual rainfall is projected to increase by 7.77% to 13.74% under the SSP2-4.5, and 14.02% to 28.48% under the SSP5-8.5 scenario in the 2040s and 2080s, respectively in eastern Amhara, northeastern Ethiopia. The mean annual temperature is expected to rise by 1.27 °C and 1.61 °C in the 2040s and by 2.21 °C and 3.92 °C in the 2080s under the SSP2-4.5 and SSP5-8.5 scenarios, respectively. The projected Kiremt season rainfall shows increasing trends, while the Belg season rainfall is expected to decrease under both climate scenarios and periods. Regarding spatial distribution, the most significant increases in rainfall and temperature are expected in the western and central parts of the study region. Overall, the rise in rainfall and temperature is projected to be greater under the SSP5-8.5 scenario compared to SSP2-4.5 across the study region. These findings identify specific regional hotspots that can inform potential adaptation and mitigation strategies to sustain agricultural production. All assessed rainfall extreme indices, except for extreme wet (R99p) in the baseline and consecutive dry day (CDD) under SSP2-4.5 and SSP5-8.5, showed significant increasing trends with erratic variation. Consecutive dry day (CDD), number of heavy days (R10mm), very wet (R95p) and extremely wet day (R99p)showed high variation (>31%), whereas maximum one-day (RX1day) and consecutive five days (RX5 days), simple daily intensity index (SDII), consecutive wet day and annual total wet day (PRCPTOT) showed low (CV<20%) in the current and future projections. The western and central parts received maximum PRCPTOT, R1ady, R5day, SDII, R10mm, very heavy days (R20mm), CWD, moderate wet (R75p), heavy wet (R90p), and R99p with upward trends. Likewise, to rainfall extremes, all examined temperature indices like annual mean maximum temperature (TMAXmean), annual mean minimum temperature (TNmean), Max Tmax (TXx), Min Tmax (TXn), Max Tmin (TNx), Min Tmin (TNn), Summer days (SU25), cold spell duration indicator (CSDI), and warm spell duration indicator (WSDI), and warm day (TX90p) showed increasing trends, except for warm night (TN90p), cool day (TX10p), cool night (TN10p). Among these, TXmean and TNmean had moderate values with low variability (CV < 20%). WSDI, TX10p, TX90p, CSDI, TN10p, and TN90p had low levels with high variability (CV > 30%) across periods and scenarios. Increasing TXmean and TNmean trends are linked with rising TX90p and TN90p and decreasing of TX10p and TN10p. Most maximum and minimum temperature extremes are projected to increase. This increase is especially expected in the region's northern and eastern parts. The results suggest that the region may be affected by both the variability and the magnitude of climate extreme indices. Spatiotemporally assessed climate changes and extreme indices showed upward trends with irregular distribution; a local-scale study is recommended to emphasize the need for targeted adaptation strategies to mitigate climate-related risks. The calibrated and evaluated results of the CROPGRO-Dry bean model in DSSAT revealed excellent agreement between the simulated and observed values. The model can serve as a valuable tool for predicting climate impacts and identifying optimal management practices like plant density, planting dates, suitable cultivars, and production situation (rain-fed and supplemental irrigation) for dry bean production in the semi-arid areas of tropical lowlands and similar agro-ecologies. According to the impact of climate change, increasing of rainfall during the cropping season, along with higher CO₂ concentrations and evapotranspiration in future climate scenarios may positively influence dry bean yields. However, rising temperatures could have adverse effects. The projected climate changes indicate negative impacts on the duration to flowering and maturity, but positive impacts on biomass and grain yields. The combined use of long-duration cultivars, late sowing dates, and supplementary irrigation could significantly increase grain yield by 46% in the baseline scenario, 38.8% and 38.3% in the 2040s, and 37.6% and 39.6% in the 2080s under the SSP2-4.5 and SSP5-8.5 scenarios, respectively, compared to conventional practices (which involve mid-duration cultivars sown in mid-July with rain-fed systems). Moreover, the individual or combined application of high plant density and long-duration cultivars consistently enhanced simulated grain yields across all study locations under current and future climates. This strategy improved yields by 36.9% in the baseline, 32.6% and 32.8% in the 2040s, and 32% and 28.5% in the 2080s under the SSP2-4.5 and SSP5-8.5 scenarios, respectively. The most significant benefits are expected to occur in the mid-century (2040s) under both SSP scenarios. Adaptation role assessment results showed that long-duration cultivars with all planting dates under supplemental irrigation and plant density levels within the same cultivar had a significant effect on yield in the current climate and an adaptive role for future climates. Despite its positive impact on grain yield, a long-duration cultivar with high plant density, and the interactionbetween late sowing date and supplemental irrigation with the same cultivar, played a decisivepositive adaptation role across climate periods and scenarios. To optimize dry bean productivityamid changing climatic conditions, it is recommended to introduce long-duration cultivars,increase plant density, delay sowing dates, and use supplemental irrigation. Because, thecombination of long-duration cultivars, delay sowing dates, and use supplemental irrigation mayenhance dry bean productivities by 43.5% in the current, 36.4% and 35.8% in the 2040s, as wellas 35.6% and 35.6% in the 2080s under SSP2-4.5 and SSP5-8.5, respectively. Similarly, aninteraction of high plant density and long-duration cultivars, whether applied individually or in combination, consistently outperformed conventional practices in terms of grain yield. Specifically, their combined application yielded improvements of 36.9% under baselineconditions, 32.7% in the 2040s, and up to 30.3% in the 2080s in the above respective climatescenarios. These findings highlight the need to implement integrated agronomic strategies tailored to evolving climate conditions. In conclusion, adopting a combination of late planting dates, longduration cultivars, supplemental irrigation, and high-density planting with these genotypespresents a promising strategy for stabilizing dry bean yields and ensuring food security invulnerable regions, such as northeastern Ethiopia. Future research should include sensitivityanalyses of other environmental and management factors like soil types, pest pressures, andnutrient availability to improve model applicability and support resilient agricultural planning.We also recommend using a multi-model ensemble approach to cross-validate findings andexpanding field experiments to capture broader environmental variability. Integrating crop modeling with socioeconomic surveys will be essential to address adoption barriers. These stepswill enhance the robustness of results and provide more reliable recommendations for optimizingdry bean production under diverse conditions.