<p>Addressing the limitations of most existing drought indices, which often select time scales subjectively and exhibit significant variability in drought severity represented by the same index value across different climatic regions, this paper proposes a novel drought index that integrates mathematical and graphical methods and is independent of historical climate data. By hypothesizing a soil evaporation model, we establish a water balance equation that maintains soil drought levels under drought conditions. Leveraging the relationship between soil surface evaporation and water surface evaporation under drought, we transform the soil evaporation variable in the equation into water surface evaporation, a routine meteorological observation, thereby constructing maintenance lines for various soil drought levels based on conventional meteorological observations. These lines divide the cumulative precipitation-cumulative evaporation coordinate plane into regions corresponding to different drought levels. Subsequently, the initial day of accumulation is determined using the rolling maximum distance method on a daily basis, and a drought index is constructed based on the distances from the maximum distance point to the respective soil drought level maintenance lines. Comparisons with other drought indices such as SPI and SPEI, in terms of correlation with soil moisture, capture of drought processes, and effectiveness of historical drought inversions, reveal that the proposed index exhibits strong correlation with soil moisture, sensitive and accurate capture of drought processes, good agreement with actual records of historical drought years, and convenient calculation without the need for historical data sequences. These advantages make it significant for drought monitoring, assessment, and guidance in drought prevention.</p>