Assessing the sensitivity of lower-atmospheric characteristics to agricultural land use classification over the lower Mississippi River alluvial valley

Authors: DYER, JAMIE - Mississippi State University; Rigby Jr, James

Submitted to: Journal of Theoretical and Applied Climatology
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/3/2020
Publication Date: 7/10/2020
Citation: Dyer, J., Rigby Jr, J.R. 2020. Assessing the sensitivity of lower-atmospheric characteristics to agricultural land use classification over the lower Mississippi River alluvial valley. Journal of Theoretical and Applied Climatology. 142:305-320. https://doi.org/10.1007/s00704-020-03318-w.
DOI: https://doi.org/10.1007/s00704-020-03318-w

Interpretive Summary: In many rural landscapes agricultural irrigation is a major component of the water budget. Irrigation most often represents a withdrawal from groundwater or surface water bodies and inherently supplements soil moisture to achieve maximal crop yields. Inefficiencies in irrigation may also contribute to runoff and deep percolation. The development and use of irrigation is fundamentally a response to the irregularity in timing and amount of precipitation as the only natural supply of water for crops. When considering local water budgets, precipitation is treated in nearly all cases independently of local irrigation practices. Land use, crop type, and precipitation (and/or soil moisture status) each display potentially large-scale spatial coherence contributing to consequent large spatial correlations in irrigation application. Additionally, temporal irrigation scheduling is often significantly impacted by nearest neighbors. Put simply, a farmer is more likely to irrigate if neighboring farmers are irrigating. While substantial research has addressed the impacts of large-scale land-use change on precipitation patterns, the impacts of large-scale manipulation of soil moisture has received less attention. In climates where the growing season is dominated by relatively weak synoptic conditions such that a large percentage of growing season precipitation is contributed by local convection processes, even relatively minor alterations to precipitation patterns may be significant for the management of water resources. Thus, for extensive, irrigated landscapes such as the High Plains, the California Central Valley, and the Lower Mississippi Valley where irrigation represents a spatially coherent large-scale alteration of soil moisture on the landscape, the question arises whether irrigation practices may alter precipitation patterns during the growing season in favorable or unfavorable ways. This work demonstrates changes in regional atmospheric processes deriving from large scale irrigation...

Technical Abstract: The lower Mississippi River alluvial valley (LMRAV) is a key agricultural area within the US, and although it receives a substantial level of annual rainfall, irrigation remains a requirement to sustain high productivity. Since a large percentage of irrigation comes from limited groundwater sources, predictions of surface and lower-atmospheric characteristics associated with convective rainfall processes are critical for planning and managing water resources. While numerical weather models are a key tool in this prediction effort, there is considerable error in the models associated with the correct categorization of regional land use. This is especially true of the LMRAV, where most agricultural land is defined as dry cropland despite the extensive use of irrigation. To improve the accuracy of regional model simulations over the LMRAV, this project investigates the sensitivity of changing the dominate land use category from dry to irrigated cropland within a high-resolution Weather Research and Forecasting (WRF) model simulation. Based on a five month simulation (May – Sept. 2016) over the LMRAV, results of the study show that replacing dry cropland with irrigated cropland leads to substantial variations in the low-level thermal and moisture balances. In general, the northern LMRAV was shown to have the greatest increase (decrease) in latent (sensible) heat flux in August and September, with a corresponding increase in 2-meter dew point temperature where latent heat flux increased. Additionally, boundary layer heights were shown to decrease over the northern LMRAV over the simulation, likely a result of decreased temperatures resulting from a dampened sensible heat flux.