Global Navigation Satellite Systems (GNSSs) are currently used in civil aviation to provide aircraft with position and velocity estimates from en-route to Precision Approach (PA) operations. The challenge for the incoming years is to extend the use of GNSSs "from gate to gate" [1]. More particularly, the goal is to use GNSSs to perform automatic taxi and parking operations. In order to support automatic surface operations, future airport navigation systems will combine guidance and control functions [2]. The guidance function will provide indications to the pilot to navigate on the airport surface [3]. The control function will allow the pilot to drive in all weather conditions using steering indications [3]. Some publications propose operational requirements for guidance and control during operations in terms of integrity [3] [4] and underline that these operations require high integrity performance. Integrity is a measure of the trust which can be placed in the correctness of the information provided by the total system. It includes the ability of a system to provide timely and valid warnings when the system must not be used for the intended operation [5]. A number of systems have been implemented to provide integrity monitoring. Among these systems, Aircraft Based Augmentation Systems (ABAS) are of particular interest as they do not need the support of external infrastructures. The integrity algorithms employed within ABAS consist of functions to check measurement consistency of the sensors and to compute Protection Levels (PLs). PLs depend upon the expected error characteristics of the measurements employed, the geometrical configuration as well as the probabilities of failure modes allocated to the responsibility of ABAS. Due to this dependency, realistic threat models are required to correctly assess the performance of ABAS algorithms. Performance of ABAS algorithms considering a realistic threat model adapted to surface operations has not been widely investigated in the literature. More specifically, performance of ABAS algorithms have been mostly investigated considering multipath threat model adapted to in-flight operations [9]. Multipath is the reception of reflected or diffracted replicas of the desired signal by the GNSS airborne antenna [6]. A multipath error model has been standardized for in-flight operations [7] considering that the structure of the aircraft itself is the dominant source of multipath error. However, during surface operations, additional sources of multipath, such as other aircraft and surrounding buildings, may contribute to the resultant pseudo range error [8]. The standard multipath error model is thus not adapted to surface operations and a new multipath threat model must be proposed. In a first part this paper presents a multipath threat error model adapted to surface operations. Firstly a preliminary analysis based on simulations is presented. It reveals that the GNSS ranging measurements are most of the time affected by low amplitude multipath errors (few cm up to few dm) when the aircraft is far from the gate, that is to say during taxi operations. However, high amplitude multipath errors (few m) may temporarily affect the ranging measurements when the aircraft is close to the gate or other parked aircraft. Secondly, and based on this analysis, a classification of the GNSS multipath ranging errors into nominal multipath ranging errors and multipath ranging failures is proposed. The concepts of multipath nominal error, single multipath ranging failure and multiple multipath ranging failures are defined. Thirdly, a methodology to identify the probability of occurrence of the multipath failure modes during surface operations is developed. This methodology is applied to Blagnac airport. Probabilities of occurrence of single multipath ranging failures and multiple multipath ranging failures during taxi and parking operations in Blagnac airport are provided. In a second part this paper presents an analysis of the performance of ABAS algorithms during surface operations. Firstly an overview of the existing ABAS algorithms is proposed and a specific algorithm is selected for the rest of the analysis. Secondly failure modes (other than multipath failure modes) that may represent a threat in terms of integrity for surface operations are identified as well as their probabilities of occurrence. Based on the analysis of the different failure modes that have to be considered during surface operations as well as their related probabilities of occurrence, the paper explains how the total allowed integrity risk is allocated and how the missed detection probabilities for each failure mode are evaluated. Thirdly the horizontal position levels are simulated with the selected ABAS algorithm parameterized by the obtained probabilities of missed detection and by the required probabilities of false alert. These simulations allow analyzing the ability of the selected ABAS algorithm to provide integrity monitoring for surface operations. More specifically, the necessity to develop innovative detection and mitigation techniques for multipath ranging errors in order to improve the performance of ABAS algorithms during surface operations is underlined. Possible detection and mitigation techniques for multipath ranging errors during surface operations are discussed.