Impact of irrigation with gypsiferous mine water on the water resources of parts of the upper Olifants basin.

Abstract

The generation of large quantities of mine wastewater in South African coal mines and the needs for a cost effective, as well as an environmentally sustainable manner of mine water disposal, have fostered interests in the possibility of utilizing mine water for irrigation. Such a possibility will not only provide a cost-effective method of minimizing excess mine drainage, as treatment using physical, chemical and biological methods can be prohibitively expensive, but will also stabilize the dry-land crop production by enhancing dry season farming. Considering the arid to semi-arid climate of South Africa, the utilization of mine water for irrigation will also boost the beneficial exploitation of the available water resources and relieve the increasing pressure on, and the competition for, dwindling amounts of good quality water by the various sectors of the economy. The disposal of excess gypsiferous mine water through irrigation has been researched in a few collieries in the Witbank area. In this study, the assessment of the impacts of using gypsiferous mine water for irrigation were carried out in parts of the Upper Olifants basin upstream of Witbank Dam, using the ACRU2000 model and its salinity module known as ACRUSalinity. The study area was chosen on the bases of locations of previous field trials and the availability of mine water for large-scale irrigation. The primary objectives of the study were the development of relevant modules in ACRU2000 and ACRUSalinity to enable appropriate modelling and assessment of the impact of large-scale irrigation with mine water and the application of the modified models to the chosen study area. The methodology of the study included the modifications of ACRU2000 and ACRUSalinity and their application at three scales of study, viz. centre pivot, catchment and mine scales. The soils, hydrologic and salt distribution response units obtained from the centre pivot scale study were employed as inputs into the catchment scale study. The soils, hydrologic and salt distribution response units obtained from the catchment assessment were in turn applied in similar land segments identified in the mine used for the mine scale study. The modifications carried out included the incorporation of underground reservoirs as representations of underground mine-out areas, multiple water and associated salt load transfers into and out of a surface reservoir, seepages from groundwater into opencast pits, precipitation of salts in irrigated and non-irrigated areas and the incorporation of a soil surface layer into ACRUSalinity to account for the dissolution of salts during rainfall events. Two sites were chosen for the centre pivot scale study. The two sites (Syferfontein pivot of 21 ha, located in Syferfontein Colliery on virgin soils; Tweefontein pivot of 20 ha, located in Kleinkopje Colliery on rehabilitated soils) were equipped with centre pivots (which irrigated agricultural crops with mine water), as well as with rainfall, irrigation water and soil water monitoring equipment. The pivots were contoured and waterways constructed so that the runoff could leave the pivots over a weir (at Tweefontein pivot) or flume (at Syferfontein pivot) where the automatic monitoring of the quantity and quality of runoff were carried out. The runoff quantities and qualities from the pivots were used for verification of the modified ACRU2000 and ACRUSalinity. The catchment scale study was on the Tweefontein Pan catchment, which was a virgin area mainly within the Kleinkopje Colliery, draining into the Tweefontein Pan. The data on the water storage and qualities in Tweefontein Pan, as well as the soil water salinities in the irrigated area located within the catchment were used for verification of results. In the catchment scale study, different scenarios, including widespread irrigation on virgin and rehabilitated soils, were simulated and evaluated. For the mine scale study, the Kleinkopje Colliery was used. The colliery was delineated into 29 land segment areas and categorized into seven land use types, on the basis of the vegetation and land uses identified in different parts of colliery. The centre pivot and catchment scale studies indicated that the impacts of irrigation with low quality mine water on the water resources are dependent on the soil types, climate, the characteristics and the amount of the irrigation mine water applied, whether irrigation was on virgin on rehabilitated soils and the status of the mine in terms of whether a regional water table has been re-established in an opencast mining system or not. The studies further indicated that the irrigation of agricultural crops with low quality mine water may lead to increases in soil water salinity and drainage to groundwater, but that the mine water use for irrigation iii purposes can be successfully carried out as most of the water input onto the irrigated area will be lost through total evaporation and a significant proportion of the salt input, both from rainfall and irrigation water, will either be precipitated in the soil horizons or dissolved in the soil water of the soil horizons. By irrigating with a saline mine water therefore, the salts associated with the low quality mine water can be removed from the water system, thereby reducing the possibility of off-site salt export and environmental pollution. On-site salt precipitation, however, may lead to accumulation of salts in the soil horizons and consequent restriction of crop yields. Therefore, efficient cropping practices, such as leaching and selection of tolerant crops to the expected soil salinity, may be required in order to avoid the impact of long-term salinity build up and loss of crop yields. The simulated mean annual runoff and salt load contribution to Witbank Dam from the Kleinkopje Colliery were 2.0 x 103 MI and 392 tons respectively. The mean annual runoff and salt load represented 2.7% and 1.4% of the average water and salt load storage in Witbank Dam respectively. About 45% of the total water inflow and 65% of the total salt load contribution from the study area into Witbank Dam resulted from groundwater storage. From the scenario simulations, the least salt export would occur when widespread irrigation is carried out in rehabilitated areas prior to the re-establishment of the water table due to a lower runoff and runoff salt load. It may therefore be a better water management strategy in active collieries if irrigation with mine water is carried out on rehabilitated soils. In conclusion, this research work has shown that successful irrigation of some (salt tolerance) crops with low quality mine water can be done, although increases in the soil water salinity of the irrigated area, runoff from the irrigated area and drainage to the groundwater store can occur. Through the modifications carried out in the ACRU2000 model and the ACRUSalinity module in this research work, a tool has been developed, not only for application in the integrated assessment of impact of irrigation with mine water on water resources, but also for the integrated assessment and management of water resources in coal-mining environments in South Africa.