High Variability in Temporal and Spatial Nutrient Retention in Mediterranean Streams

Abstract

Nutrient retention studies were conducted in two second—order streams located north of Barcelona (Spain) that differed in watershed lithology, soil type, and vegetation. La Solana drains a calcareous watershed, whereas Riera Major drains a siliceous watershed. Within each stream we defined two reaches (50 m) differing in channel form and canopy cover: bedrock and sand—cobble. The bedrock reaches were characterized by a dominance of bedrock outcrop and a sparse canopy, while the sand—cobble reaches had sand, cobble, and boulder substrata and a dense canopy. Phosphate and ammonium uptake length and uptake rate were measured in each reach by performing several short—term additions (2 h) of phosphate and ammonium. Releases were conducted from summer 1990 to spring 1992 to encompass seasonal changes in environmental conditions. We examined temporal trends in nutrient uptake length to test the influence of watershed geomorphology and local channel form on nutrient retention in these two Mediterranean streams. Because both nutrients were added simultaneously, we were able, by using the ratio of each nutrient's uptake length, to compare ammonium and phosphate retention in each reach. Results of a two—way ANOVA indicated that both stream and reach type had significant effects on phosphate and ammonium uptake lengths. However, channel form had a different effect depending on the stream. In La Solana, shortest phosphate uptake lengths (highest efficiency in phosphate retention) were measured in summer, whereas shortest ammonium uptake lengths were measured in winter in the two reaches. In this stream, both nutrients were retained with greater efficiency in the bedrock reach than in the sand—cobble reach (mean uptake length was 59 ± 12 m for phosphate and 59 ± 14 m for ammonium in the bedrock reach, means ±1 SE, n = 13 for both nutrients; and 119 ± 19 m for phosphate and 182 ± 51 m for ammonium in the sand—cobble reach, n = 12 for each nutrient). In Riera Major, only the bedrock reach exhibited a seasonality in nutrient uptake length that was similar to that found in La Solana reaches. The sand—cobble reach in Riera Major showed higher efficiency in ammonium retention than the bedrock reach (85 ± 16 m, n = 14; and 189 ± 28 m, n = 10, respectively); but phosphate uptake lengths did not differ between reaches to a statistically significant extent (163 ± 26 m in the sand—cobble reach, n = 14; and 191 ± 32 m in the bedrock reach, n = 10). When the two streams were compared, La Solana exhibited higher nutrient retention efficiencies (shorter uptake lengths) than Riera Major. However, nutrient uptake rates for both nutrients were higher in Riera Major because of greater nutrient loadings. Mean phosphate uptake rates at ambient levels for the whole study period were 45.4 ± 10.8 &mu g P.m‐2.min‐1 in La Solana (n = 25 and 175.5 ± 68.6 μg P.m‐2.‐1 in Riera Major (n = 24); and mean ammonium uptake rates at ambient levels were 103.3 ± 27.4 and 186.7 ± 48.6 μg N.m‐2 c.min‐1, respectively. Finally, nutrient uptake lengths from this study are within the range reported for other streams of similar order. Overall results indicate that differences in ammonium retention efficiency in these streams may be tightly related to in—stream processes. In contrast, major between—stream differences in phosphate retention efficiency appear to be a function of overall phosphate availability.