The role of antecedent (previous autumn, winter, and spring) land surface conditions (precipitation, surface air temperature, soil moisture, snow cover) in the NAM warm season (Jun-Jul-Aug-Sep) precipitation remains elusive. These relationships are nonetheless important for improving seasonal monsoon predictability, especially for the NAM core region of northwestern Mexico, a semi-arid area where NAMS warm season precipitation greatly affects social and economic life. Unlike its Asian monsoon sister, relatively little previous work has addressed NAM teleconnections through data analysis because of the lack of land surface variables such as soil moisture over a long enough time period, and sufficient spatial extent, to support meaningful analyses. Producing reasonable long-term gridded precipitation, temperature, and other derived variables like soil moisture for Mexico is a major challenge, especially for the period prior to the 1960s because of the sparse gauge station distribution and discontinuous gage records. We combined three different sources of station data : ERIC2, a product (1940-1998) from Mexican Institute of Technology of Water (IMTA) of the SEMARNAP; DATA322, produced by SMN (Servicio Meterorol"gico Nacional, 2000) with some stations dating back to the 1915; and SMN daily historical precipitation data ( 1995 - near real-time). These were used to create gridded forcing data of acceptable quality for long-term (1950-2002) retrospective VIC runs. The land surface predictors used in the study are monthly aggregates from the 1950-2002 VIC archive. The retrospective archive includes gridded precipitation (P), mean surface air temperature (Ts), and Variable Infiltration Capacity (VIC) land surface model-derived soil moisture (Sm) and snow water equivalent (SWE). In a previous similar study over the U.S. Southwest, we proposed a land surface feedback hypothesis and through correlation and composite analysis we found that while the land surface state in spring is affected by the previous winter's precipitation anomaly, that anomaly contributes little to the magnitude of NAM precipitation. In this study, we apply a similar analysis method to evaluate whether a land surface feedback between previous season's land surface state and the strength of the NAMS exists over Northwestern Mexico and if so, how it varies with time. This study is designed to produce a better picture of the importance of land surface feedback in the monsoon system over Northwestern Mexico.