Assessing the ecological integrity after nutrient inputs in streams: the relevance of the descriptor and its associated scale. Sergi Sabater 1,* , Vicenç Acuña 2 , Adonis Giorgi 3 , Helena Guasch 1 , Elena Guerra 2 , Isabel Muñoz 2 , and Anna M. Romaní 1 1 Institute of Aquatic Ecology and Department of Environmental Sciences, Fac. Sciences, Univ. Girona, Campus Montilivi 17071, Girona, Spain. 2 Dept. of Ecology, Fac. Biology, Univ. Barcelona, Avgda. Diagonal 645, 08028 Barcelona, Spain. 3 PIEA, Depto. Ciencias Básicas, Universidad Nacional de Luján. CC 221-6700 Luján, Argentina. * Corresponding author E-mail address sergi.sabater@udg.es Abstract Two different streams receiving high nutrient content were compared in relation to the evidences provided by structural (chlorophyll concentration) and functional (community metabolism) descriptors. The first (Gaià) was an unshaded agricultural stream, where high Net Community Metabolism and chlorophyll concentrations were only modified by flood episodes which caused the sloughing of the benthic community. The second studied stream (Fuirosos) was forested, and an artificial nutrient addition was sustained for 44 days. In this case, the observations expanded also to bacteria and macroinvertebrate, and were compared to a control upstream stretch with low nutrient content. Even though there was an increase of chlorophyll and algal density because of the nutrient addition, the macroinvertebrate community density as a whole was not affected. Moreover, metabolism (Net Community and Respiration) did not experience significant changes. Therefore, the nutrient addition under light limiting conditions was not causing remarkable effects on the stream functioning. However, the observations during a colonization experiment showed that chlorophyll and bacterial density became similar for the different substrata analyzed (sand, rocks) because of the nutrient addition. It is concluded that chlorophyll, algal communities or macroinvertebrate communities may be accurate descriptors of the effect of nutrients on unshaded streams, but that in shaded streams a more precise time scale is necessary to detect structural effects in the system. These structural changes may be associated with significant modifications of the stream functioning (such as the nutrient retention) in the longer term. Key words: algae, bacteria, macroinvertebrates, colonization, monitoring Introduction A healthy ecosystem is defined as a sustainable, active system, which maintains its organization and autonomy over time, and is resilient to stress (Haskell et al 1992). Applying this definition to streams, where water is derived, extracted and polluted, and resources are managed for fisheries, wood extraction or dredging, it might be concluded that streams and rivers do not fit within healthy systems in its great majority. It is true that up to 80% of fluvial systems are regulated or suffer from alteration of the riparian vegetation (Nilsson et al. 1997). However, it is also true that fluvial systems may be highly resilient, i.e. able to recover from a disturbed state with much more