The quantitative assessment of crop yield loss in response to drought is crucial in the development of the agricultural sector to improve the productivity. This study estimated and analyzed the spatiotemporal patterns of crop yield loss in response to drought using the Lagrange interpolation method, wavelet analysis, and sequential Mann-Kendall test in the mountain, hill, and Terai (low-land) regions of Nepal's Koshi River Basin from 1987 to 2016. According to the findings, average crop yield loss was common after 2000, with the Terai, hill, and mountain experiencing the greatest loss in maize, rice, and wheat, respectively. Average annual rice and wheat yield losses rate were highest in the mountains, while maize yield losses were highest in the Terai. There was an abrupt change in wheat yield loss in the mountain, with significant increasing trend. In the hill, significant increment in maize and wheat yield loss, and decrement in rice yield loss, were observed. Between 1987 and 2016, periodic variations of maize, rice, and wheat revealed significant yield loss after 2000. The characteristics of the first and second key periods for crop yield loss demonstrated variation period which predicted that crop yield loss would either enter high yield loss or low yield loss period shortly after 2016. The findings of the study provide a detailed intervention in assessing crop yield loss at the river basin level and can provide an important pathway for developing a crop yield loss mitigation plan in the agricultural sector to achieve self-reliance and sustainable agricultural productivity.