Investigation of the large-scale atmospheric moisture field over the midwestern United States in relation to summer precipitation. Part I: Relationships between moisture budget components on different timescales

Atmospheric moisture budget components are evaluated for a large area (1.23 × 10⁶ km²) in the midwestern United States for all 12-h (1200-0000, 0000-1200 UTC) and 24-h (1200-1200 UTC) periods during the contrasting summers (May-August) of 1975, 1976, 1979, and 1988. The atmospheric moisture flux divergence (MFD, separated into horizontal and vertical advection components. HA and VA) and storage change (dPW) are estimated using a standard finite-difference method applied to objectively analyzed U.S. and Canadian rawinsonde data (50-hPa vertical resolution, surface-300 hPa) for 0000 and 1200 UTC. Area-averaged precipitation (P) totals are derived from approximately 600 relatively evenly distributed (but ungridded) recording rain gauges. Evapotranspiration (E) is estimated as a residual of the moisture budget equation and compares favorably with the few existing observations, especially when totaled for periods of 1 month or longer. Relationships between the budget components are established for the daily, monthly, and seasonal timescales using stratification, correlation, and cross-spectral analyses. On monthly and seasonal timescales, the surface is a net source of water vapor (positive E-P) and the bulk of this surplus is exported from the region, largely through HA. For the daily budget, a threshold P rate (∼4 mm day^-1) separates surplus E-P budgets from deficit budgets. On all timescales, most of the P variance is reflected in the VA component of MFD, while HA explains ∼80% of the variation in dPW. For the monthly and (especially) daily budgets, E has bimodal distributions with P where the minimum E occurs at P ∼ 2.6 mm day^-1 (monthly) and P ∼ 4-5 mm day^-1 (daily). For drier daily P regimes, relatively high E is associated with increased (decreased) dry VA (HA). The correlation of E with P becomes substantially more positive from the daily-to-monthly timescale, confirming the importance of land-atmosphere interactions over longer periods. The above stratification and correlation results are complemented by cross-spectral analyses that identify strong associations between P-HA and P-dPW previously masked by phase differences. The cross-spectral results also prompt the development of a conceptual model that describes the temporal relationships among the budget components for eastward-moving large-scale, "wavelike" disturbances with 3-10-day timescales. The suggested sequence of interactions-moist HA is accompanied by a pronounced PW increase and then followed by a moist VA maximum; this horizontal and then vertical moisture redistribution is first associated with an E minimum and then culminates in a P maximum; after the P event, atmospheric drying occurs through increased (diminished) dry HA (moist VA), which leads to an E maximum and then P minimum.