Residential Irrigation Water Use in the Central Piedmont of North Carolina. I: Measured Use and Water Requirements

Irrigation is the most common and standard practice for maintaining turfgrass and landscape plants in residential settings. The main objective of this study was to quantify the residential irrigation water use in Cary, North Carolina. A subobjective was to develop distributions of monthly reference evapotranspiration (${\mathrm{ET}}_o$) and gross irrigation requirements (GIRs) for the area using long-term weather data. As part of the subobjective, an investigation was performed to compare daily measured solar radiation with two empirical methods of estimating solar radiation ($R_s$). The goal of this investigation was to determine the relative accuracy of these methods and to evaluate the impact of using $R_s$ estimates in the computation of ${\mathrm{ET}}_o$ and GIR for the Raleigh-Durham area, where Cary is located. Irrigation water use data from 2005 to 2007 for 120 randomly sampled residences were evaluated. The average lawn area of the sampled residences was $713{\mathrm{m}}^2$. None of the sampled households used drip irrigation. A negative correlation existed between the irrigated area and applied water depth. The May-October period accounted for 84–89% of the total annual outdoor water use for the 3 years evaluated. A trend of applying more water than the gross irrigation requirement was observed during the lawn renovation period (mid-September through late October). Solar radiation was computed using two methods: the Hargreaves and Samani method based on difference in maximum and minimum temperature, and the Doorenbos and Pruitt method based on percent possible sunshine. Solar radiation estimated using the percent possible sunshine method showed a strong relationship ($R^2=0.88$) with measured solar radiation and proved to be a much more reliable way of estimating solar radiation than the temperature difference method for the Raleigh-Durham region. Estimates of mean, median, and $P_{80}$ (80th percentile) monthly ${\mathrm{ET}}_o$ and GIR constructed from a 60-year data set show that ${\mathrm{ET}}_o$ tends to peak in July at an average of 154 mm, whereas GIR peaks in June with an average of 105 mm. The distribution of GIR is substantially more variable than ${\mathrm{ET}}_o$ due mostly to variability in rainfall, illustrating the difficulty in applying the correct amount of water consistently from year to year in subhumid to humid regions.