Four squares (2 sq m each) were excavated: two on the northern boundary of the site (Area A) and two along its southern boundary (Area B). Care was taken to meticulously control and separate the excavated material, whereby each square was subdivided into four squares of 1 sq m each, and a separate basket was assigned to each sub-square for every 10 cm of depth, as the excavation descended. All of the excavated material was sifted through a 0.5 cm mesh. About a quarter of all the excavated material underwent fine-sifting (1 mm mesh) and floatation (10 liters for every 10 cm of depth) in order to collect tiny archaeozoological and archaeobotanical finds, coins and other very small material finds. Sediment samples and charred botanical material were collected from the exposed sections of the excavation areas. These were sent to laboratories for a geo-archaeological analysis regarding the formation of the refuse and for 14C dating along with a statistical model employing the Bayesian method. Work on these samples is in progress.
 
Area A. The excavation was concentrated mainly in this area, near the bathhouse; two squares were opened (1, 4; depth 1.3 m). Square 1 was excavated at the top of the large refuse mound, on the northwestern fringes of the site (230.65 m asl). Due to its elevation and volume, the mound stands out prominently in the landscape overlooking the Haluza ruins. The excavation in Sq 1 produced large amounts of pottery sherds, glass items, coins, floral remains and animal bones. Judging by the types of finds, the refuse in the upper portion of the mound can be dated to the late Byzantine period (fifth–sixth centuries CE). A deep section revealed a micro-stratigraphy of thin lenses (each c. 5 cm thick; Figs. 3, 4) consisting of alternating light sandy lenses and dark lenses containing large amounts of ash and charred botanical finds. The lenses in the northern and southern sections of Sq 1 slanted toward the west, with the incline of the slope of the mound, and they probably represent separate incidents of refuse being piled up on the mound. The excavation in this square did not reach virgin soil.
Square 4 was excavated at the foot of the ‘garbage mound’, c. 50 m west of Sq 1, in an area where a shallow, flat refuse mound was observed (224.44 m asl). This mound also yielded a large amount of material finds, albeit at a lower density than was found in Sq 1. Virgin soil was not reached in this excavation square either. The finds that were uncovered in the square predate those that were exposed in Sq 1 and include pottery from the late fourth and early fifth centuries CE. 
 
Area B. Two squares were excavated (6, 7; depth c. 0.5 m). Square 6 was opened near the northern fringes of the refuse mound and yielded a small number of finds. Square 7, which was opened several meters to the south, near the top of the mound (234.85 m asl), produced a rich amount of material finds. A deep probe in this square exposed dark layers of ash alternating with light sandy layers, similar to the sequence exposed in Sq 1. A glance at the sections in Sq 7 reveals almost horizontal layers with only a slight incline, even though the square was excavated into a southward-facing gentle slope. The outline of the original mound has evidently eroded and changed over the years. The density and variety of the finds on the southern mound were lower than those in Sq 1. A count of the types of pottery from Sq 7 shows an overwhelming majority of Gaza jars (c. 48%).
 
Pottery. The assemblage consisted of locally produced 'Haluza Ware' pottery vessels (Fabian and Goren 2002) from the early Byzantine period, which included bowls (Fig. 5:1–5) and jars (Fig. 5:6–9), and from the late Byzantine period, comprising bowls (Fig. 5:10–14), jars (Fig. 5:15–18), a juglet (Fig. 5:19), a cooking pot (Fig. 5:20) and tubuli (Fig. 5:21, 22), as well as two 'Haluza Ware' wine jars (Fig. 5:15, 16). Vessels produced in ‘Aqaba were also found: jars (Fig. 6:1, 2), one of which is a ‘ration jar’ (Fig. 6:1), and a flask (Fig. 6:3). In addition, Gaza jars (Fig. 6:4–14), which constituted a significant portion of the ceramic assemblage, and wheel-made sandal lamps (Fig. 6:15–17) were found. Two types of Gaza jars, produced in the southern Shephelah, Majcherek’s Forms 2 (fourth – mid-fifth centuries CE) and Form 3 (mid-fifth – mid-sixth centuries CE; Majcherek 1995: Pl. 3:2–3). Among the many ceramic finds discovered in Sq 1 was a vessel in the shape of a rooster’s head (Fig. 7), possibly a figurine, which dates from the late Byzantine period.
 
The excavation in the refuse of Byzantine Haluza produced rich material finds, including a wide variety of floral and faunal matter. The density, nature and distribution of the finds, as well as preliminary geo-archaeological indications point to the organized removal of refuse from within Haluza outside the boundaries of the settlement throughout the Byzantine period. These initial observations are not consistent with Negev’s conclusions (Negev 1993), as they indicate a prolonged accumulation of household waste during the Byzantine period—an activity that continued at least until the middle of the sixth century CE. The assemblages that were collected will assist in providing a highly detailed characterization of the social, economic and environmental parameters of the different phases of the city’s existence. These finds will also allow the dating of activity in the city with a high level of accuracy and the examination of geomorphological processes and landscape dynamics in the environs of the city.
Large amounts of bio-archaeological matter were collected in each of the city’s refuse areas. The botanical finds include a large number of grape, date and olive seeds, cereals and legumes. Numerous bones of sheep/goats, cattle and pig were identified in a preliminary examination of the animal remains. The composition of the species and the multitude of skeletal parts rich in meat indicate that most of the refuse was household garbage rather than butchering waste. Numerous young animals were identified among the pig remains, and many butchering marks were discovered on the bones. A large amount of fish bones from the Red Sea and shellfish from the Mediterranean Sea were also identified.
The Byzantine ‘garbage’ provides a unique time capsule that preserves an abundance of organic refuse and rich material finds. The refuse included an especially high density of sherds belonging to vessels that were used for storage, cooking and serving, with one of the most prominent elements being the large number of Gaza jars that were used to store wine. Some of the pottery, such as the Haluza Ware, was produced in the city or nearby, while others, including luxury items, were imported from distant places. The vessels that were produced in the Red Sea port of ‘Aqaba are known from previous research (Dolinka 2003); it is also known that ‘Aqaba (Ayla) served as a base for the Tenth Roman Legion into the Byzantine period. ‘Aqaba Ware, especially the jars that contained combat rations (Erickson-Gini 2010:165), were discovered in strongholds and in settlement sites with a Roman military presence, such as the fortress of ‘Avdat and the Hazeva (Erickson-Gini 2010:131). Presumably, most of the refuse was piled up during the fifth–sixth centuries CE, a time when the city seems to have reached the height of its economic prosperity. When the city’s urban infrastructure collapsed, for reasons that are still unclear, sometime in the mid-sixth century CE, the organized garbage removal was most probably halted, as the city itself was abandoned or continued to exist without a central urban government and with a dwindling population. Further research will aim at determining the quantitative composition of both the pottery assemblage and the animal bones and floral remains. This analysis will allow us to reconstruct the household economy and the dietary habits of the residents of the Byzantine city and will be instructive regarding other aspects of Haluza’s economy, such as its trade routes.