Relationship between groundwater level variations using Grace satellite data and rainfall

Data from the Gravity Recovery and Climate Experiment (Grace) satellite and stations were used to estimate variations of the monthly groundwater level in the Zayanderud basin, Iran, over the period 2000–2018. In addition, the annual and seasonal storage of groundwater in this basin was estimated and verified using Grace satellite data, and compared with data from the previous rainfall. The results showed that the current groundwater level depends on the amount of rainfall in past years. When rainfall is on a downtrend, the trend of underground water storage fluctuations is affected by rainfall to a greater extent. In addition to the time trend, the pattern of the normalised difference vegetation index (NDVI) depends on the amount of groundwater storage variations in the Zayanderud basin. The lowest NDVI (0·078) in 2016 coincided with the lowest amount of underground water storage and recharge rate (27·36 cm). Statistical correlation analysis of Grace satellite data and recorded data from wells indicated a root mean square error of 2·23 cm, a mean absolute error of 3·28 cm and a mean bias error of 1·02 cm on the seasonal scale. The results show that Grace data can provide a good understanding of long-term variations when studying the relationship between groundwater level and rainfall, especially in large basins.