The incremental impacts of climate change on elements within the water cycle are a growing concern. Intricate karst aquifers have received limited attention concerning climate change, especially those with sparse data. Additionally, snow cover has been overlooked in simulating karst spring discharge rates. This study aims to assess climate change effects in a data-scarce karst anticline, specifically Khorramabad, Iran, focusing on temperature, precipitation, snow cover, and Kio spring flows. Utilizing two shared socioeconomic pathways (SSPs), namely SSP2-4.5 and SSP5-8.5, extracted from the CMIP6 dataset for the base period (1991–2018) and future periods (2021–2040 and 2041–2060), the research employs Landsat data and artificial neural networks (ANNs) for snow cover and spring discharge computation, respectively. ANNs are trained using the training and verification periods of 1991–2010 and 2011–2018, respectively. Results indicate projected increases in temperature, between + 1.21 °C (2021–2040 under SSP245) and + 2.93 °C (2041–2060 under SSP585), and precipitation, from + 2.91 mm/month (2041–2060 under SSP585) to + 4.86 mm/month (2021–2040 under SSP585). The ANN models satisfactorily simulate spring discharge and snow cover, predicting a decrease in snow cover between − 4 km2/month (2021–2040 under SSP245) and − 11.4 km2/month (2041–2060 under SSP585). Spring discharges are anticipated to increase from + 28.5 l/s (2021–2040 under SSP245) to + 57 l/s (2041–2060 under SSP585) and from + 12.1 l/s (2021–2040 under SSP585) to + 36.1 l/s (2041–2060 under SSP245), with and without snow cover as an input, respectively. These findings emphasize the importance of considering these changes for the sustainability of karst groundwater in the future.