Erratum ‘Erratum to “Application of WRF/Chem-MADRID for real-time air quality forecasting over the Southeastern United States” [Atmos. Environ. 45 (2011) 6241e 6250]’ Ming-Tung Chuang a, b , Yang Zhang a, * , Daiwen Kang c a Air Quality Forecasting Lab, North Carolina State University, Raleigh, NC 27695, USA b Currently at National Central University, Taiwan, ROC c Computer Science Corporation, Research Triangle Park, NC 27709, USA article info Article history: Received 7 June 2012 Received in revised form 7 July 2012 Accepted 10 July 2012 An error was discovered in two global attributes in some emis- sion files used for real-time air quality forecasting (RT-AQF) using WRF/Chem-MADRID: MMINLU ¼ "", and ISWATER ¼ 0, where MMINLU is a character string and enclosed in quotation marks ("), indicating the type of land use categories, and ISWATER is an integer specifying the land use category of water. Their correct values should have been MMINLU ¼ "USGS" (i.e., the U. S. Geological Survey 24-category Land Use Categories), and ISWATER ¼ 16 (i.e., water bodies). The incorrect ISWATER value in those files led to an incorrect assignment of a value of 1 (Land) to the landmask index over oceanic area (the correct landmask for ocean should have been 0). The impact of the incorrect MMINLU and ISWATER values over ocean on the RT-AQF results and model performance presented in our paper was found to be very small. This is attributed to the fact that the WRF model physics does not predict sea-surface tempera- ture (SST) and sea ice (i.e., no oceanic processes and airesea inter- actions are treated) (NCAR, 2011). Instead, the model simulations used SST and sea ice fields from the National Center for Environ- mental Prediction’s (NCEP)/FNL meteorological forecast. To further verify the impact and our assessment, we compared model simulation results of May 2010 with incorrect and correct MMINLU and ISWATER values. The absolute differences in the forecasted mixing ratios of maximum 1-h average ozone (O 3 ) and maximum 8-h average O 3 are 1.7 and 1.6 ppb, respectively. The corresponding percentage differences are 3.7% and 3.9%, respectively. The absolute difference in the forecasted concentra- tions of 24-h average PM 2.5 is À0.2 mgm À3 . The corresponding percentage difference is À1.9%. The performance of the model from both simulations is evaluated using observations from the U.S. EPA’s AIRNow network. With correct MMINLU and ISWATER values, the normalized mean bias (NMB) changes from À5.6% to À9.1% for maximum 1-h average O 3 , from À4.6% to À8.1% for maximum 8-h average O 3 , and from À8.0% to À6.2% for 24-h average PM 2.5. The normalized mean error (NME) changes from 21.9% to 20.9% for maximum 1-h average O 3 , from 22.1% to 21.0% for maximum 8-h average O 3 , and from 40.3% to 38.3% for 24-h average PM 2.5. The use of incorrect MMINLU and ISWATER values did not change the direction of model predictions (i.e., O 3 and PM 2.5 concentrations remain slightly underpredicted) nor the overall performance (i.e., performance remains satisfactory, NMBs of within Æ15%, Seigneur (2001), U.S. EPA (2001), and Y. Zhang et al. (2006)). These results are consistent with our aforementioned impact assessment. The fact that the model used NCEP/FNL SST fields throughout all months and a consistency between the RT-AQF results in 2009 and those during MayeSeptember 2010e2011 with correct MMINLU and ISWATER values as shown in N. Zhang et al. (2011) support such an impact assessment for RT-AQF results during MayeSeptember 2009 presented in the paper, namely, the inadvertent use of incorrect MMINLU and ISWATER values caused negligible changes in the simulated concentrations of O 3 and PM 2.5 and performance statistics; it had no impact on the conclusions in this paper. Acknowledgements We thank Yao-Sheng Chen and Nan Zhang, former members of Air Quality Forecasting laboratory (AQFL) at North Carolina State DOI of original article: 10.1016/j.atmosenv.2011.06.071. * Corresponding author. Air Quality Forecasting Lab, North Carolina State University, Campus Box 8208, Room 1125, Jordan Hall, 2800 Faucette Drive, Raleigh, NC 27695-8208, USA. Tel.: þ1 919 515 9688; fax: þ1 919 515 7802. E-mail address: yang_zhang@ncsu.edu (Y. Zhang). Contents lists available at SciVerse ScienceDirect Atmospheric Environment journal homepage: www.elsevier.com/locate/atmosenv 1352-2310/$ e see front matter http://dx.doi.org/10.1016/j.atmosenv.2012.07.019 Atmospheric Environment 60 (2012) 677e678