Technical Abstract: Water quality criteria, irrigation recommendations, and reclamation guidelines were all developed in an era of abundant high quality water and land resources. The objective was to select resources that would avoid problems of salinity or reduced yield. Criteria were inevitably very cautious, as a result of this abundance, limited understanding of the processes and response by the ecosystem, as well as the need to simplify recommendations. Currently we face limited water and land resources, due to competition from growing municipal and industrial needs. The agricultural sector no longer has the resources to select the optimal water for irrigation or over-irrigate or over- apply amendments to ensure high productivity. Knowledge that the resources deemed unsuitable can in fact often be used, results in a lack of confidence in the standards, as the outcome depends on factors that are not considered in the criteria. There is an urgent need to improve and refine these criteria and recommendations, incorporating more science and inevitably making the decision more complex. The needs are twofold: To improve out knowledge base to include all factors and interactions and to incorporate this knowledge into a decision tool. The complexity of the interactions suggests that computer modeling will be essential to predict system response. Examples are given demonstrating the usefulness of models such as UNSATCHEM to analyze management options, especially if they are able to utilize transient conditions. However much additional experimental research is needed before the models can be used without field confirmation of predictions. Quantitative knowledge of the interaction of many physical and chemical processes is necessary for predicting short and long term consequences of varied management practices.