Understanding hydroclimatic trends in semi-arid river basins is vital for sustainable water management under changing climate conditions. This study introduces a modified innovative trend analysis (MITA) to address the temporal ordering limitations of the conventional innovative trend analysis (ITA). MITA complements established nonparametric approaches - Mann-Kendall (MK), modified Mann-Kendall (MMK), and Sen's slope estimator - applied here in a four-method comparative framework. We applied these four methods to nine hydroclimatic indices - annual, monsoon, and non-monsoon rainfall; wet-day frequency and intensity; and temperature extremes - derived from high-resolution gridded data (1971-2021) across 62 grid points in the Manjira River Sub-basin (MRSB), India. Comparative analyses revealed distinct patterns: The Mann-Kendall test detected significant rainfall declines and notable warming trends (Z > 6.0, p < 0.01, slope: +0.014 to +0.017 degrees C year(-1)). Accounting for autocorrelation, MMK reduced the significance of rainfall trends. ITA, which disrupts chronology by sorting data halves, produced artificially rising rainfall trends. Conversely, MK/MMK indicated declining rainfall (Z = -2.32 to -2.56, p < 0.01), while MITA - by correcting ITA's artificial sorting bias and preserving temporal order - revealed pronounced drying (-26.7 to -126.6 mm), consistent with the conservative but realistic estimates of MK/MMK. Cohen's kappa coefficients (kappa = 0.89 for MK-MMK; kappa = 0.76 for ITA-MITA) underscore that preserving temporal sequences yields more realistic trend detection, which is crucial for accurate water resources planning under climate change.