An investigation of seasonal precipitation patterns for rainfed agriculture in the Southeastern region of the United States
Introduction
Worldwide, rainfed agriculture is a very common practice which plays an important role in food and livestock production (Rockström et al., 2010). Rainfed agriculture generally depends on crops, soils, environment, and most importantly, seasonal precipitation totals and their distribution over the crop-growing season. However, rainfed agriculture is vulnerable to climate hazards and its practice does not ensure crop yield stability (Rockström et al., 2010). For instance, when precipitation events are irregular, crops are often subjected to short-term water deficits which cause crop yield losses (Kistner et al., 2018). Indeed, several studies reported the negative impact of short-term water deficit on annual crops yields (Bauer et al., 2009). Particularly, under rainfed conditions, substantial yield gap values (i.e. difference between average and yield potential) are often reported as the consequence of water stress (Lobell et al., 2009). In humid regions such as the Southeastern United States (US), the cumulative annual precipitation totals are relatively high and virtually sufficient to grow a wide range of crops under rainfed conditions. Yet, the sole consideration of annual precipitation amounts may shade critical patterns of precipitation irregularity during the crop growing seasons. This is particularly true in the Southeastern US where short-term precipitation deficits are usual during the crop season (Mo and Schemm, 2008). These short-term precipitation deficits are not necessarily a consequence of a decrease of the total annual or seasonal precipitation as they also seem to be a result of changes in the timely distribution of precipitation events (Kistner et al., 2018). Studies evidenced that changes of precipitation frequency and the subsequent short-term water deficits reduce biomass production, shorten grain filling periods, and lower crop yields (Högy et al., 2013; Katerji et al., 2004). Therefore, an understanding of precipitation patterns (total amount and frequency) is useful for the decision making toward crops management and yields stability.
The frequency of precipitation is critical in rainfed agriculture because it indicates how often the soil water is replenished by natural precipitation events during a given time slice (e.g. month, season, year). In general, the number of precipitation events is assimilated to the number of wet days which can be defined based on a precipitation threshold. Unfortunately, not all precipitation amounts are effective for crops (Rockström et al., 2003; Dastane, 1974). Hence, it is important to define the frequency of precipitation events by referring to daily total thresholds. For instance, the threshold of 5 mm/day (0.2 inch/day) was used in different irrigation scheduling manuals as a minimal daily water need even though the actual need depends on the crop growth stage, the atmospheric water demand, and the antecedent soil moisture condition (Curwen and Massie, 1994). Nevertheless, the 5 mm threshold has been consistently used to study the trend and effects of precipitation events (Goswami et al., 2006; Sala and Lauenroth, 1982). As an example, Mulhouse et al. (2017) considered daily precipitation events of >5 mm to address vegetation response to precipitation timing. Our study also uses the 5 mm threshold to define and address the seasonal number of precipitation events in the Southeastern US.
The climate of the Southeastern US is humid, and the region receives an annual precipitation above 1200 mm. Historically, farmers in the Southeastern US relied on natural precipitations to grow annual crops such as corn, cotton, soybean, peanut, etc. However, the absolute reliance on natural precipitations is nowadays threatened by cases of short-term precipitation deficits occurring during crop growing seasons. Although these precipitation deficits often cause a decrease in crop yield (Rockström et al., 2010), their occurrence is still very unpredictable and therefore needs research attention. To improve the understanding of the occurrence and severity of short-term precipitation deficits, our study had the following three objectives: (i) propose a coherent precipitation regionalization for the Southeastern US, (ii) examine the long-term patterns of precipitation, and (iii) establish a probabilistic framework to support agricultural water management.
Section snippets
Data
The study addresses the part of the Southeastern US encompassing the States of Georgia (GA), North Carolina (NC), and South Carolina (SC). The study aims to regionalize the Southeastern US based on the long-term patterns of precipitations. Hence, data were collected from a total of 422 land-based weather stations distributed across the Southeastern US (Fig. 1). For the 422 stations, all the available daily precipitation records over the period 1960 to 2017 were obtained from the National
Principal components and cluster analysis
The PCA conducted on the seasonal precipitation total and the seasonal frequency of precipitation events >5 mm yielded respectively 19 PCs and 15 PCs with an eigen value greater than one. After the varimax rotation, the 19 PCs obtained from the seasonal precipitation total captured 84% of the total variability in the time-series, while the 15 PCs obtained from the seasonal number of precipitation events >5 mm captured 80% of the variability. The original m by n (i.e. 208 × 232) matrix of
Synthesis, discussion and application
The study analyzed precipitation patterns across the Southeastern US of the United States and outlined implications for agricultural water management. The analyses used 58 years (i.e. period 1960 to 2017) daily precipitation data obtained from 208 synoptic stations distributed across the Southeastern US. The yearly seasons winter (DJF), spring (MAM), summer (JJA), and fall (SON) were considered to generate time-series of seasonal precipitation totals and seasonal numbers of precipitation events
Conclusion
Under rainfed systems, Lobell et al. (2009) indicated that crop yields are in average 50% or less below yield potentials due to water stress. Although this tendency is global, it presumes room for improving yields in regions dominated by rainfed agriculture practices such as the Southeastern US. Indeed, several studies supported the possibility of ceiling yield potential with supplement irrigation (Van Ittersum et al., 2013; Grassini et al., 2011). However, yield gaps vary consistently from one
Disclaimer
Mention of trade names or commercial products in this article is solely for the purpose of providing specific information and does not imply recommendation or endorsement by the U.S. Department of Agriculture.
References (39)
- et al.
High-yield irrigated maize in the Western US corn belt: I. On-farm yield, yield potential, and impact of agronomic practices
Field Crops Research
(2011) - et al.
Impacts of temperature increase and change in precipitation pattern on crop yield and yield quality of barley
Food Chem.
(2013) - et al.
Score normalization in multimodal biometric systems
Pattern Recognit.
(2005) - et al.
Comparison of corn yield response to plant water stress caused by salinity and by drought
Agric. Water Manage.
(2004) The use of satellite data for crop yield gap analysis
Field Crops Research
(2013)- et al.
Managing water in rainfed agriculture—The need for a paradigm shift
Agric. Water Manage.
(2010) - et al.
Quantifying the probabilistic divergences related to time-space scales for inferences in water resource management
Agric. Water Manage.
(2019) - et al.
Yield gap analysis with local to global relevance—a review
Field Crops Research
(2013) - et al.
Power of the Mann–Kendall and Spearman’s rho tests for detecting monotonic trends in hydrological series
J. Hydrol.
(2002) - et al.
Regionalization of precipitation in Switzerland by means of principal component analysis
Theor. Appl. Climatol.
(1997)
A comparison of two cotton cultivars differing in maturity for within-canopy fiber property variation
Crop Sci.
Principal components-based regionalization of precipitation regimes across the southwest United States and northern Mexico, with an application to monsoon precipitation variability
Climate Research
Patterns and causes of Atlanta’s urban heat island–initiated precipitation
J. Appl. Meteorol.
Drivers of long-term precipitation and runoff variability in the Southeastern USA
Theor. Appl. Climatol.
Increasing trend of extreme rain events over India in a warming environment
Science
Principal component analysis
International Encyclopedia of Statistical Science
The application of electronic computers to factor analysis
Educational and psychological measurement
Cited by (22)
Agricultural drought risk and local adaptation measures in the Upper Mun River Basin, Thailand
2024, Agricultural Water ManagementEvaluation of phosphorus runoff from sandy soils under conservation tillage with surface broadcasted recovered phosphates
2023, Journal of Environmental ManagementCitation Excerpt :In particular, the amount of SRP wash-off from plots treated with recovered P from swine manure and turkey litter ash represent small fractions of SRP runoff from plots treated with commercial TSP. The findings of this study are highly relevant for the management of sandy soils of the Southeastern Coastal Plain region known for their poor nutrient holding capacity (Sohoulande et al., 2019, 2020). Indeed, using turkey litter ash or recovered P from swine manure as P fertilizer could be an alternative to enhance soil nutrient management under conservation tillage in the region and substantially increase livestock nutrient recycling at the regional scale.
Drought and flood risk assessment for rainfed agriculture based on Copula-Bayesian conditional probabilities
2023, Ecological IndicatorsSeasonal precipitation pattern analysis for decision support of agricultural irrigation management in Louisiana, USA
2021, Agricultural Water ManagementCitation Excerpt :Thresholds of 5 mm, 12.7 mm (half inch), and 25.4 mm (1 in.) were used to generate gridded time series of seasonal numbers of precipitation events according to Sohoulande et al. (2019, 2014). The higher Jaccard index value corresponded to the optimum number of clusters and identified the precipitation regions.