<p>In survey sampling, handling missing or ambiguous data is a recurring problem, particularly when working with intricate and indeterminate datasets. This paper adapts some fundamental neutrosophic imputation frameworks and proposes some optimal neutrosophic imputation frameworks along with their resultant estimators for estimating the population mean under simple random sampling (SRS), with a focus on addressing the missingness in indeterminate data. The mean square error (MSE) expressions of the resultant adapted and proposed estimators are determined up to first order approximation. Comparative analysis of the proposed neutrosophic imputations with the adapted neutrosophic imputations is conducted. The theoretical results are verified by a simulation study based on neutrosophic symmetric and asymmetric data. The paper illustrates the usefulness of these imputations in handling missing indeterminate data and enhancing the accuracy of population mean estimates by applying it to real world climate data. The findings suggest that the optimal neutrosophic imputation frameworks are not only an important tool for climate data analysis but also have broader applications in domains where missingness are prevalent in uncertain and incomplete data.</p>