Incorporating scientific uncertainty as part of science teaching means acknowledging that there may be incomplete or potentially limited scientific information when scientists draw conclusions. In the geosciences, scientists routinely make inferences about the Earth based on observations of the present, and test those observations against hypotheses about Earth's history and processes that are not readily observable. We engage students in thinking about uncertainty as part of writing scientific arguments in an online Earth science curriculum module called, "Will there be enough fresh water?" In this article, we introduce the module, the structured scientific argumentation task format, and students' uncertainty attributions identified across eight argumentation tasks in the module. The impact of the module on student learning of uncertainty attribution was based on 546 middle- and high-school students taught by nine teachers who implemented the module in six states. Students achieved significant gains from pretest to posttest on scientific argumentation (effect size = 1.38 SD, p < 0.001) consisting of claims, explanations, uncertainty rating, and uncertainty attribution. According to our linear regression analysis, students' uncertainty attribution performance during the module was a statistically significant predictor (p < 0.001) for their posttest uncertainty attribution performance, although gender (p = 0.48), ethnic minority status (p = 0.71), English language learner status (p = 0.27), and prior computer use for science learning (p = 0.42) were not. Teachers can use these results to support students' consideration of the strengths and limitations of data and models and the uncertainty of the science itself.