ClimateSearch

Xi’an, Shaanxi

China · 9.6M people · 411 m elevation · Northern Hemisphere

Every month in detail

Thirty-year averages for each calendar month.

Monthly climate normals. Values shown as an em dash are missing from the source data.
MonthJanFebMarAprMayJunJulAugSepOctNovDec
Average high42°F (6°C)49°F (10°C)60°F (16°C)71°F (22°C)80°F (26°C)87°F (31°C)89°F (32°C)86°F (30°C)77°F (25°C)66°F (19°C)55°F (13°C)45°F (7°C)
Average low25°F (-4°C)30°F (-1°C)39°F (4°C)49°F (9°C)58°F (14°C)66°F (19°C)71°F (22°C)70°F (21°C)61°F (16°C)50°F (10°C)38°F (3°C)28°F (-2°C)
Dew point18°F (-8°C)24°F (-4°C)32°F (0°C)41°F (5°C)49°F (10°C)58°F (15°C)66°F (19°C)65°F (18°C)58°F (14°C)47°F (8°C)34°F (1°C)22°F (-6°C)
Rainfall0.5 in (12 mm)0.7 in (18 mm)1.2 in (32 mm)1.8 in (45 mm)2.7 in (70 mm)3.5 in (88 mm)5.5 in (140 mm)5 in (128 mm)5.1 in (130 mm)3.1 in (80 mm)1.6 in (40 mm)0.5 in (13 mm)
Rainy days3467871111121073
Snowfall2.5 in (6.4 cm)2.1 in (5.4 cm)0.7 in (1.9 cm)0.1 in (0.3 cm)0 in (0 cm)0 in (0 cm)0 in (0 cm)0 in (0 cm)0 in (0 cm)0 in (0 cm)1.7 in (4.3 cm)1.6 in (4.1 cm)
Sunshine250 h232 h270 h280 h290 h301 h319 h301 h223 h224 h214 h237 h

Scroll the table sideways for all twelve months.

Projected change, 2031–2050

Model projection, not an observation

How this city’s climate may change, measured against the same models’ own 1995–2014 output. Nothing here affects the match score above.

Change in summer afternoon highs

+2.9°F (+1.6°C)

Models range from +2.0°F (+1.1°C) to +3.7°F (+2.1°C)

Days above 90°F (32°C) a year

40 58

About 1.5× as many

Warming by month
  • January: +2.4°F (+1.3°C)
  • February: +1.9°F (+1.1°C)
  • March: +1.9°F (+1.0°C)
  • April: +2.4°F (+1.3°C)
  • May: +1.6°F (+0.9°C)
  • June: +3.0°F (+1.7°C)
  • July: +2.7°F (+1.5°C)
  • August: +2.9°F (+1.6°C)
  • September: +4.2°F (+2.3°C)
  • October: +2.0°F (+1.1°C)
  • November: +2.0°F (+1.1°C)
  • December: +2.5°F (+1.4°C)
Scenario
SSP5-8.5 (high emissions)
Models
4 · CMIP6 HighResMIP, downscaled

This is a high-emissions pathway — the upper end of plausible futures, not a central estimate. Changes are measured by each model against its own past output and then applied to the observed record, which cancels each model’s bias but not the genuine uncertainty about how much warming occurs. Where the models disagree widely, the range above is the honest answer.