Role of Antecedent Land Surface Conditions in Warm Season Precipitation over Northwestern Mexico

Abstract

Abstract The role of antecedent land surface conditions including precipitation (P), surface skin temperature (Ts), soil moisture (Sm), and snow water equivalent (SWE) anomalies on the onset and intensity of the monsoon during the 1950–99 period in the core of the North American monsoon system (NAMS) region in northwestern Mexico (termed MSa here) is explored. A statistically significant positive relationship is found between monsoon onset date in MSa and previous winter precipitation in the southwestern United States (SW) and northwestern (NW) Mexico, and winter SWE in the southern Rocky Mountains. The linkages are strong during the 1960s–80s and weak otherwise, which is a much shorter period than had been found previously for an SW target area termed monsoon west (MW). In the MW study, the following land surface feedback hypothesis was proposed: more winter P and SWE lead to more spring Sm, hence lower spring and early summer Ts, which induce a weaker onset of the NAMS. This hypothesis broke down in MW due to the small contribution of land surface memory to surface thermal condition, and hence to monsoon strength. The same hypothesis is in this work for MSa by examining three links. First, it is found that in May not only the total column, but also the near-surface Sm, in both SW and NW Mexico have memory from the previous winter precipitation. The spring Sm anomalies correlate negatively with Ts anomalies over most of the continental United States and Mexico except for the desert region of SW and NW Mexico. The monsoon onset is negatively correlated with May Ts over an area roughly consisting of New Mexico and some adjacent areas, suggesting that antecedent land surface conditions may influence the premonsoon surface thermal condition, which then affects monsoon onset. The monsoon-driving force concept that states that the strength of the monsoon should be related to premonsoon land–sea surface temperature contrasts is also confirmed. The confirmation of this concept shows that late monsoon years are associated with colder land and warmer adjacent ocean than early monsoon years. In addition to the apparent land surface feedback, a strong positive relationship between May Ts anomalies and the large-scale midtropospheric circulation (Z500) anomalies is found, which suggests that large-scale circulation may play a strong (possibly more important than land feedback) role in modulating the monsoon onset.