1 GIS MODEL FOR PRODUCING HYDRAULIC SOIL GROUP CLASSIFICATION DIGITAL MAP OF BAGHDAD CITY Ahmed A.M. Ali * Mahmoud S. Mahdi ** Nuha Jamal Abdullah*** **Professor in Water Resources Engineering Department/ University of Baghdad. **Assistant Professor in Building and Construction Engineering Department/University of Technology *** M. Sc. Water Resources Engineering Department/ University of Baghdad. ABSTRACT Study and analysis of the soil characteristics is very important for the estimation of surface runoff and infiltration and therefore is a matter of great interest for researchers, planners and scientists who deal with the urban hydrology. The Geographical Information System (GIS) was used to produce a major United States Department of Agriculture (USDA) textural soil classification digital map and subsequently produce the Hydrologic Soil Groups (HSG) classification digital map for Baghdad City. The produced map can be baseline data for estimating runoff using the Natural Resources Conservation Service (NRCS) runoff curve number (CN) method. Soil investigations reports for the 23years (1990 to 2013) were adopted as a reference for 362 boreholes spread randomly within Baghdad City. The soil data of these boreholes and a digital map of Baghdad City were used for estimating HSG. GIS techniques were used to zone the area of study in to small neighborhoods started with digitizing fourteen district boundaries of the digital administrative division map of Baghdad City and to implement a GIS data base for the soil data. These techniques integrated with geo- statistics was applied to analyze the spatial variability in soil properties within the city, where the shape files of soil separates are interpolated using kriging combined with semi-variogram parameters to generate digital maps of sand, silt and clay which are used in preparing major USDA textural soil model. Digital maps of soil separates were used for applying percentage criteria of the twelve major USDA textural soil classes as rules within knowledge based classification model in ERDAS Imagine 11.2 software to predict the required soil characteristics digital map. The Soils within the city were assigned to HSG according to Technical Release 55 (TR55) of USDA-NRCS criteria, by using knowledge based classification model to predict the HSG map. The overall accuracy of HSG model produced from knowledge-based classification is 97.93 % with Kappa coefficient of 0.916 indicate that the produced HSG map was at high level of accuracy. The produced soil textural map shows that among twelve classes, seven classes of soil texture appeared within the area of study which were: Loam, Silt Loam, Sandy Clay Loam, Clay Loam, Silty Clay Loam, Silty Clay and Clay. While the produced HSG digital map shows that most of the area of study covered with D class (89%) followed by B class (10%) and very small part of C class (1%). The results of accuracy assessment and the produced digital maps shows that the implemented model and the produced digital maps can be effectively used for estimation runoff within Baghdad City. KEYWORDS: GIS, HSGs model, USDA textural soil classes, ﺟﻐراﻓﯾﺔ رﻗﻣﯾﺔ ﺗﺻﻧﯾفﯾدروﻟوﺟﯾﺔ賰 اﻟ ﻟﻣدﯾﻧﺔ وﺗﺣﻠﯾلم賰 ﻣ اﻟﺳﯾﺢ ھﻲ أھﻣﯾﺔ ﻛﺑﯾرة ﻟﻠﺑﺎﺣﺛﯾن واﻟﻣﺧططﯾن اﻟذﯾن ﯾﺗﻌﺎﻣﻠون ھﯾدروﻟوﺟﯾﺎ اﻟﺣﺿرﯾﺔ. اﻟﺟﻐراﻓﯾﺔ ﺗﺻﻧﯾف رﻗﻣﯾﺔ اﻟرﺋﯾﺳﯾﺔ ﻣﻌﺎﯾﯾر اﻷﻣرﯾﻛﯾﺔ(USDA) إﯾﺟﺎد ﺗﺻﻧﯾف رﻗﻣﯾﺔﯾدروﻟوﺟﯾﺔ賰 اﻟ ﻟﻣدﯾﻧﺔ. ﯾﻣﻛن اﻟﺑﯾﺎﻧﺎت اﻷﺳﺎﺳﯾﺔ ﯾﻣﻛن اﻋﺗﻣﺎدھﺎ اﻟﺳﯾﺢ طرﯾﻘﺔ اﻟﺳﯾﺢ اﻟطﺑﯾﻌﯾﺔ(NRCS) .