干旱是全球最严重的自然灾害之一，其监测存在费时费力、时效性差、准确度不高等问题，给人类防范旱涝灾害带来了巨大的挑战。随着全球变暖趋势日益严重，干旱发生愈发频繁。因此，干旱监测成为水资源管理和生态环境保护亟需解决的重点问题。此外，厄尔尼诺-南方涛动（El Niño-Southern Oscillation, ENSO）是一种异常气候事件，也会导致干旱事件发生。随着全球导航卫星系统（Global Navigation Satellite System, GNSS）技术的快速发展及GNSS气象学的提出，GNSS水汽探测技术凭借其高精度、高时空分辨率、低成本和不受天气影响等优势成为获取天顶对流层延迟（Zenith Troposphere Delay, ZTD）和大气可降水量（Precipitation Water Vapor, PWV）的重要方式之一，并逐步应用于气象干旱监测中。本文在现有研究基础上，提出了无实测气象参数下高精度PWV获取方法，发展了基于GNSS PWV/ZTD的ENSO监测及区域干旱监测理论，主要研究内容如下：

（1）针对大多数GNSS测站未配备气象传感器，难以获取实测气压（Pressure, P）和气温（Temperature, T）参数，导致无法反演高精度PWV的现状，提出了无实测气象参数下高精度PWV反演方法，满足了国际公认标准3 mm的精度要求。首先，对欧洲中期天气预报中心第五代再分析数据集提供的P和T及GNSS ZTD进行评估；其次，对PWV反演的关键参数-大气加权平均温度的五种计算模型进行分析；再次，基于PWV反演流程计算GNSS PWV值；最后，根据误差传播定律推算出PWV的理论误差，并基于无线电探空数据计算的PWV对GNSS PWV进行评估。实验结果表明提出的无实测气象参数下的GNSS PWV反演方法具有较好的精度。

（2）针对大多数研究仅定性地分析了GNSS PWV/ZTD与ENSO之间的关系，并未深入研究GNSS PWV/ZTD用于ENSO事件监测的不足，提出了基于GNSS PWV/ZTD的ENSO监测方法，通过构建监测指数对ENSO及ENSO导致的干旱事件进行监测。首先，利用多通道奇异谱分析方法确定了ZTD、PWV和T的趋势和周期信息；其次，利用经验正交函数方法分析ZTD、PWV和T的时空分布特征；再次，基于移动窗口相关性分析方法确定了ZTD、PWV和T对ENSO的响应阈值；最后，经过最优权重确定和标准化处理构建了标准化ENSO监测指数（Standardized ENSO Monitoring Index, SEMI）和标准化PWV和T指数（Standard PWV and T Index, SPTI）。实验结果表明提出的SEMI/SPTI指数对ENSO及ENSO导致的干旱具有良好的监测能力。

（3）由于现有传统站点干旱监测指数计算公式复杂（如标准化降水蒸散指数），忽略了干旱灾害的区域性特征。本研究基于GNSS PWV/ZTD监测ENSO导致干旱的理论基础，提出了基于GNSS ZTD的区域干旱监测理论与方法，解决了站点干旱指数不适用区域干旱的缺陷。首先，选择了GNSS ZTD和降水数据构建干旱指数，ZTD较PWV包含更多的数据信息，且避免了ZTD转化为PWV过程中利用经验公式计算造成的误差；其次，确定了干旱监测指数的表达式并进行多时间尺度累计；再次，考虑到不同数据量级差异，对提出的干旱监测指数进行标准化处理；最后，基于泰森多边形方法构建了区域降水和ZTD指数（Regional Precipitation and ZTD Index, RPZI）。实验评估了RPZI在2006-2018年中国干旱、半干旱、湿润和半湿润地区及1998-2018年澳大利亚昆士兰州汤斯维尔市旱涝事件的监测效果。结果表明以区域降水滑动指数（Regional Precipitation Smoothing Index, RPSI）为标准，RPZI与RPSI具有良好的一致性，验证了RPZI具有良好的区域干旱监测能力。

Drought is one of the most serious natural disasters in the world, and its monitoring is time-consuming, inefficient and inaccurate, posing a great challenge to human beings in preventing droughts and floods. With the increasing trend of global warming, droughts are occurring more and more frequently. Therefore, drought monitoring has become a key issue that needs to be addressed urgently in water resources management and ecological environmental protection. In addition, El Niño-Southern Oscillation (ENSO) is an abnormal climate event that can also lead to drought events. With the rapid development of Global Navigation Satellite System (GNSS) technology and the proposal of GNSS meteorology, GNSS water vapor detection technology has become one of the important ways to obtain Zenith Troposphere Delay (ZTD) and Precipitation Water Vapor (PWV) with its advantages of high accuracy, high spatial and temporal resolution, low cost and weather independent, and is gradually applied in meteorological drought monitoring. Based on the existing research, this paper proposes a high-precision PWV acquisition method under non-real meteorological parameters, and develops the theory of ENSO monitoring and regional drought monitoring based on GNSS PWV/ZTD, with the following main research contents:

(1) In response to the current situation that most GNSS stations are not equipped with meteorological sensors and it is difficult to obtain the measured pressure (Pressure, P) and temperature (Temperature, T) parameters, which makes it impossible to invert the PWV with high accuracy, a high accuracy PWV inversion method without measured meteorological parameters is proposed, which meets the internationally accepted standard of 3 mm accuracy. Firstly, we evaluate the P and T and GNSS ZTD provided by the fifth generation reanalysis dataset of the European Centre for Medium-Range Weather Forecasts (ECMWF); secondly, we analyze five calculation models for the key parameter of PWV inversion, the atmospheric weighted average temperature; secondly, we calculate the GNSS PWV values based on the PWV inversion process; finally, we derive the theoretical error of PWV based on the error propagation law, and calculate the PWV value based on the radio Finally, the theoretical error of PWV is derived according to the error propagation law, and the GNSS PWV is evaluated based on the PWV calculated from the radio sounding data. The experimental results show that the proposed GNSS PWV inversion method with non-measured meteorological parameters has good accuracy.

(2) In response to the shortcomings that most studies only qualitatively analyzed the relationship between GNSS PWV/ZTD and ENSO and did not deeply investigate the use of GNSS PWV/ZTD for ENSO event monitoring, a GNSS PWV/ZTD-based ENSO monitoring method was proposed to monitor ENSO and ENSO-caused drought events by constructing monitoring indices. Firstly, the trend and period information of ZTD, PWV and T are determined using the multi-channel singular spectrum analysis method; secondly, the spatial and temporal distribution characteristics of ZTD, PWV and T are analyzed using the empirical orthogonal function method; again, the moving window correlation analysis method is used to determine the ZTD, PWV Finally, the standardized ENSO Monitoring Index (SEMI) and the standardized PWV and T Index (SPTI) are constructed after the optimal weight determination and standardization. The experimental results show that the proposed SEMI/SPTI indices have good monitoring ability for ENSO and ENSO-induced drought.

(3) Due to the complicated calculation formulae of existing traditional station drought monitoring indices (such as Standardized Precipitation Evapotranspiration Index), the regional characteristics of drought disasters are ignored. Based on the theoretical basis of GNSS PWV/ZTD monitoring of ENSO-caused drought, this study proposes a theory and method of regional drought monitoring based on GNSS ZTD, which solves the defect that the station drought index is not applicable to regional drought. Firstly, GNSS ZTD and precipitation data are selected to construct the drought index, which contains more data information than PWV and avoids the errors caused by the calculation using empirical formulae in the process of converting ZTD to PWV; secondly, the expressions of the drought monitoring index are determined and accumulated on multiple time scales; again, the proposed drought monitoring index is standardized considering the differences in different data magnitudes Finally, the Regional Precipitation and ZTD Index (RPZI) was constructed based on the Tyson polygon method. The RPZI was evaluated for monitoring the drought and flood events in arid, semi-arid, wet and semi-humid regions in China from 2006 to 2018 and in Townsville, Queensland, Australia from 1998 to 2018. The results showed good agreement between RPZI and RPSI using Regional Precipitation Smoothing Index (RPSI) as the standard, which verified that RPZI has good regional drought monitoring capability.