Remote sensing products have been widely used in water resources assessment and management. However, the accuracy varies in different products. Water budget closure provides an important breakthrough for better assessing the uncertainties and quantifying error sources of satellite-based data products using water budget imbalance, especially in ungauged basins where conventional verification methods are not applicable. However, the commonly used assessment methods mainly refer to closure residual of water budget $\mathrm{\Delta }Res$ or its proportion to corresponding mean precipitation (P) $\mathrm{\Delta }Res/P$, which have apparent limitations due to the dimensional effects in the former method and the high dependence on precipitation in the latter one. More critically, these two methods do not consider measured values (referring to in-situ values in gauged basins or average values of multiple satellite products in ungauged basins as in this study) as benchmarks of water budget components in the assessment of satellite-based hydrological data products, resulting in a wrong choice because the closure result is subject to the mutual cancellation in the errors of water budget components from different satellite-based products. This study proposes a novel error-based method to assess the accuracy of satellite-based hydrological data products for both gauged and ungauged basins based on water budget equation, measured values, and the omission error (OE) which is generally overlooked in existing studies. The method is applied to a typical closed basin, the Tarim River Basin (TRB) of China, to verify the applicability and reliability of the proposed method. Sixty combinations of water budget components from various satellite sources can be designed based on five precipitation (P) products, four evapotranspiration (ET) products and three terrestrial water storage change (TWSC) products. First order reliability method (FORM) is employed to assess the reliability of water budget closure results caused by the errors of satellite sources. The results indicated that different combinations produced different annual $\mathrm{\Delta }Res$ and $\mathrm{\Delta }Res/P$, indicating great water budget closure errors for different combinations due to errors in different satellite sources. The OE error from satellite-based hydrological data products of water budget components clearly showed a greater impact on water budget imbalance than water budget components, indicating that OE error is important for understanding quality of satellite products. By comparing the results of $\mathrm{\Delta }Res$ and $\mathrm{\Delta }Res/P$, it can be seen that our proposed method had great advantages in assessment of satellite products with a low dependence on P, and a better comparability between different combinations. The value of our assessment criterion ranges from 0 to 1 which is easier to understand in practice. The combinations of water budget components from TRMM_3B43 (P), GPM_3IMERGHH (P), GLDAS_NOAH025 (ET), and GRACE (TWSC) products show satisfactory water budget closure results in TRB using our proposed method. The proposed method provides a new aspect for assessing the accuracy of satellite-based hydrological data products especially in ungauged basins, which provides insights into water resources management in basins.

遥感产品已广泛用于水资源评估和管理。但是，不同产品的精度有所不同。水资源预算的关闭为利用水资源预算失衡更好地评估卫星数据产品的不确定性和量化误差来源提供了重要的突破，尤其是在常规检验方法不适用的非流域盆地。但是，常用的评估方法主要是指水费预算的结余。$\mathrm{\Delta }回覆s$ 或其与相应平均降水量（P）的比例 $\mathrm{\Delta }回覆s/P$，由于前一种方法的尺寸效应和后一种方法对降水的高度依赖性，因此存在明显的局限性。更关键的是，这两种方法都没有将测量值（如本研究中所指的是规范盆地中的原位值或未灌流盆地中多个卫星产品的平均值）作为评估基于卫星的水文过程中水预算组成部分的基准数据产品，导致选择错误，因为关闭结果会相互抵消来自不同卫星产品的水预算要素的误差。这项研究提出了一种新的基于误差的方法，可基于水预算方程，测量值，遗漏误差（OE）在现有研究中通常被忽略。该方法应用于典型的封闭流域塔里木河流域（TRB），以验证该方法的适用性和可靠性。可以基于五种降水（P）产品，四种蒸散（ET）产品和三种地面储水量（TWSC）产品，设计来自各种卫星源的水预算要素的60种组合。一阶可靠性方法（FORM）用于评估由卫星源误差引起的水费预算结清结果的可靠性。结果表明，不同的组合产生不同的年度 验证该方法的适用性和可靠性。可以基于五种降水（P）产品，四种蒸散（ET）产品和三种地面储水量（TWSC）产品，设计来自各种卫星源的水预算要素的60种组合。一阶可靠性方法（FORM）用于评估由卫星源误差引起的水费预算结清结果的可靠性。结果表明，不同的组合产生不同的年度 验证该方法的适用性和可靠性。可以基于五种降水（P）产品，四种蒸散（ET）产品和三种地面储水量（TWSC）产品，设计来自各种卫星源的水预算要素的60种组合。一阶可靠性方法（FORM）用于评估由卫星源误差引起的水费预算结清结果的可靠性。结果表明，不同的组合产生不同的年度 一阶可靠性方法（FORM）用于评估由卫星源误差引起的水费预算结清结果的可靠性。结果表明，不同的组合产生不同的年度 一阶可靠性方法（FORM）用于评估由卫星源误差引起的水费预算结清结果的可靠性。结果表明，不同的组合产生不同的年度$\mathrm{\Delta }回覆s$ 和 $\mathrm{\Delta }回覆s/P$，表明由于不同卫星源的误差，不同组合的水预算关闭误差很大。来自水预算组成部分的基于卫星的水文数据产品的OE误差显然显示出比水预算组成部分对水预算失衡的更大影响，表明OE误差对于理解卫星产品的质量至关重要。通过比较结果$\mathrm{\Delta }回覆s$ 和 $\mathrm{\Delta }回覆s/P$可以看出，我们提出的方法在评估卫星产品时具有很大的优势，其对P的依赖性较低，并且不同组合之间具有更好的可比性。我们的评估标准的取值范围是0到1，在实践中更容易理解。使用我们建议的方法，TRMM_3B43（P），GPM_3IMERGHH（P），GLDAS_NOAH025（ET）和GRACE（TWSC）产品的水预算要素组合显示出令人满意的TRB中水预算封闭结果。所提出的方法为评估基于卫星的水文数据产品的准确性提供了一个新的方面，尤其是在非流域的盆地，这为流域水资源管理提供了见识。