Leaf-litter decomposition in headwater streams: a comparison of the process among four climatic regions Jesús Pozo1,10, Jesús Casas2,11, Margarita Menéndez3,12, Salvador Mollá4,13, Inmaculada Arostegui5,14, Ana Basaguren1,15, Carmen Casado4,16, Enrique Descals6,17, Javier Garcı́a-Avilés7,18, José M. González8,19, Aitor Larrañaga1,20, Enrique López2,21, Mirian Lusi2,22, Oscar Moya6,23, Javier Pérez1,24, Tecla Riera3,25, Neftalı́ Roblas9,26, AND M. Jacoba Salinas2,27 1 Dpto. Biologı́a Vegetal y Ecologı́a, F. Ciencia y Tecnologı́a, UPV/EHU, Apdo. 644, 48080 Bilbao, Spain 2 Dpto. Biologı́a Vegetal y Ecologı́a, Universidad de Almerı́a, Ctra. Sacramento s/n, La Cañada, 04120 Almerı́a, Spain 3 Dpto. Ecologı́a, F. Biologı́a, Universidad de Barcelona, Avda. Diagonal 645, 08028 Barcelona, Spain 4 Dpto. Ecologı́a, F. Ciencias, Universidad Autónoma de Madrid, Darwin 2, 28049 Madrid, Spain 5 Dpto. Matemática Aplicada e Investigación Operativa, F. Ciencia y Tecnologı́a, UPV/EHU, Apdo. 644, 48080 Bilbao, Spain 6 Instituto Mediterráneo de Estudios Avanzados, IMEDEA (CSIC), Miquel Marqués 21, 07190 Esporles, Mallorca, Spain 7 Dpto. de Ecologı́a, F. de Ciencias Biológicas, Universidad Complutense de Madrid, C/ José Antonio Novais, 2, 28040 Madrid, Spain 8 Dpto. Biologı́a y Geologı́a, Universidad Rey Juan Carlos, C/ Tulipán, s/n, 28933 Móstoles, Madrid, Spain 9 Centro de Investigaciones Ambientales de la Comunidad de Madrid, Ctra. M-607 km 20, 28760 Tres Cantos, Madrid, Spain Abstract. The main purpose of our work was to elucidate factors responsible for the geographical differences in leaf-litter decomposition rates in Spanish oligotrophic headwater streams. Decomposition experiments with alder (Alnus glutinosa) leaf litter were carried out in 22 headwater streams in 4 different climatic regions across the Iberian Peninsula (Cornisa Cantábrica, Cordillera Litoral Catalana, Sierra de Guadarrama, and Sierra Nevada). Streams that were similar in size, flowed mainly over siliceous substrate in catchments with scarce human settlements and activities, and fell within a range of low nutrient concentrations were chosen in each region. Breakdown rates were regionally variable and were low (0.109– 0.198% ash-free dry mass [AFDM]/degree day [dd]) in the Cornisa Cantábrica, the most mesic and Atlantic region, and high (0.302–0.639% AFDM/dd) in Sierra de Guadarrama, one of the coldest and most inland areas. Temperature was not the determining factor affecting differences in breakdown rates among regions, and breakdown rates were not related to concentrations of dissolved nutrients. However, microbial reproductive activity (sporulation rates) was significantly correlated with dissolved P concentration. Breakdown rates were explained better by presence and feeding activities of detritivores than by decomposer activity. Incorporation of breakdown rates in assessment schemes of stream ecological status will be difficult because leaf processing does not respond unequivocally to environmental factors when climatic regions are considered. Thus, regional adjustments of baseline standards in reference conditions will be required. Key words: leaf litter, decomposition, headwater streams, invertebrates, fungi, eutrophication, Spain. 19 jmgonzalez@escet.urjc.es 20 aitor.larranagaa@ehu.es 21 emlopez@ual.es 22 mirianlusi@hotmail.it 23 oscarmoyamesa@gmail.com 24 javier.perezv@ehu.es 25 triera@porthos.bio.ub.es 26 neftali.roblas@madrid.org 27 mjsalina@ual.es 10 E-mail addresses: jesus.pozo@ehu.es 11 jjcasas@ual.es 12 mmenendez@ub.edu 13 salvador.molla@uam.es 14 inmaculada.arostegui@ehu.es 15 ana.basaguren@ehu.es 16 c.casado@uam.es 17 ieaedc@uib.es 18 ciam03@bio.ucm.es J. N. Am. Benthol. Soc., 2011, 30(4):935–950 ’ 2011 by The North American Benthological Society DOI: 10.1899/10-153.1 Published online: 6 September 2011 935 Fluvial ecosystems have been impaired by and continue to deteriorate because of a wide array of human impacts of varying magnitude, ranging from severe alterations with conspicuous effects to subtle and cryptic modifications. Headwater streams, which rep- resent .95% of the total number of stream segments (Wallace and Eggert 2009), are less affected by humans than other water bodies and are crucial reservoirs of biodiversity. A critical step in preserving or improving the integrity of a river (sensu Karr 1991) is to assess its ecological status adequately with methods sensitive enough to determine the consequences of human effects or to guarantee the success of eventual restoration actions. Traditional evaluations of river health rely on physicochemical characteristics (Müller et al. 2008, Fu et al. 2009) or on structural properties of community diversity and composition of several taxonomic groups, mainly macroinvertebrates, algae, or macrophytes (Barbour et al. 1999, De Jonge et al. 2008, Demars and Edwards 2009). Recently, ecologists have advocated use of func- tional components of the ecosystem to evaluate river health and have argued that, in some cases, stressors might change function but not structure (Moulton 1999, Bunn and Davies 2000, Brooks et al. 2002, Gessner and Chauvet 2002, Riipinen et al. 2009, Young et al. 2008). Moss (2008) pointed out that ecological quality is measured accurately by paying attention primarily to the intactness of several fundamental characteristics of ecosystem function rather than to secondary character- istics, such as particular concentrations of substances or species composition. This controversy of structural vs functional indica- tors seems to be implicit in the European Water Framework Directive (WFD) (2000/60/EC). An appar- ent contradiction, noticed by Moss (2008), exists between the definition of a high ecological status of aquatic ecosystems and the instructions given in the WFD on the way that ecological status is to be determined or improved. A high ecological status embraces fundamental characteristics (ecosystem func- tion), but the instructions encourage focus on second- ary details (mainly taxonomic structure) and, hence, may undermine the fundamental improvement of aquatic ecosystems that was intended (Moss 2008). Leaf-litter decomposition in streams is a functional ecosystem variable that integrates the activity of several phylogenetic groups (Gessner and Chauvet 2002, Young et al. 2008). The rate of leaf-litter decomposition depends on natural factors, such as climate, geology, altitude, and latitude, and responds strongly to changes in environmental variables (e.g., temperature, pH, dissolved O2, nutrients, sediments, riparian vegetation) caused by anthropogenic distur- bance (Webster et al. 1995, Molinero et al. 1996, Pozo et al. 1998, Niyogi et al. 2003, Elosegi et al. 2006, Sampaio et al. 2008). Eutrophication is one of the most widespread human effects on inland waters (Withers and Jarvie 2008). Studies on stream eutrophication generally demonstrate that dissolved nutrients enhance decom- position rates of leaf litter by increasing microbial activity (e.g., Suberkropp and Chauvet 1995, Gulis and Suberkropp 2003, Greenwood et al. 2007), at least under moderate nutrient enrichment. However, the effects on macroinvertebrate colonization and leaf- litter consumption seem to be more variable, which could be a result of the community variability between regions or a possible response to other pollutants in eutrophic streams (e.g., Pascoal et al. 2003). Further- more, the effect of eutrophication on stream ecosystem processes can depend on factors other than P or N supplies, such as temperature and flow regimes, substrate, and C supply (Dodds 2007, Withers and Jarvie 2008), which may vary naturally within and across regions (Casas et al. 2006). Therefore, natural variation may hinder the application of functional indices aimed at comparing the effect of eutrophication across different geographical and climatic settings. According to Karr and Chu (1999), an understanding of the baseline of natural variation is the foundation for precise assessment of change caused by humans. Attempts have been made to evaluate ecosystem functioning based on leaf decomposition across large geographical areas (Irons et al. 1994, Young et al. 2004, Lecerf et al. 2007, McKie et al. 2008, Woodward 2009, Hladyz et al. 2010, Pérez et al. 2011). The aim of our study was to compare leaf-litter processing in small headwater streams slightly affected by nutrient en- richment among 4 different geographic and climatic areas of the Iberian Peninsula (Cornisa Cantábrica, Cordillera Litoral Catalana, Sierra de Guadarrama, and Sierra Nevada). We hypothesized that: 1) large-scale abiotic conditions (especially temperature) would influence biotic contributors to leaf breakdown and, therefore, differences in decomposition rates among regions, and 2) leaf-litter processing would respond positively to dissolved nutrients through enhancement of microbial activity. Methods Study sites The study was conducted in 22 low-order streams in the Iberian Peninsula: 6 in the north (Cornisa Cantábrica [CC]), 4 in the northeast (Cordillera Litoral Catalana [CLC]), 7 in the center (Sierra de Guadar- rama [SG]), and 5 in the south (Sierra Nevada [SN]) 936 J. POZO ET AL. [Volume 30 (Fig. 1, Table 1). In each area, streams were similar in size and flowed mainly over siliceous substrate in catchments with scarce human settlements and activities (Table 1). Annual precipitation and mean temperature varied among regions from 310 to 923 mm and from 9.0 to 16.4uC, respectively, and changed with altitude (Table 1). Streams from CC and CLC were more mesic than those of SG and SN, which were at higher altitudes. The Gorzynski Continental- ity Index (GCI) was used as a measure of continen- tality (i.e., climatic gradient) of the geographic areas and for comparative purposes. It is calculated as GCI = 1.7(Mi – mi)/sin(Lat) – 2.4, where Mi and mi are the highest and the lowest mean monthly temperatures (uC), respectively, and Lat is latitude in degrees (www.globalbioclimatics.org). Values were: 7.7 (CC), 17.1 (CLC), 23.6 (SG), and 32.0 (SN). Streams draining larger catchments were chosen to avoid temporary streams in the drier regions. Streams differed in catchment area, but most streams had a mean channel width ,5 m (Table 1). Groups of streams spanned a range of low dissolved nutrient concentrations (particularly P, ,50 mg PO4-P/L) in each area. Streams within this low and narrow eutrophication gradient were diffi- cult to locate in SG and SN because of the sharp transition from oligotrophic to severe eutrophic conditions caused by organic pollution (wastewater). Therefore, in these areas, only streams with low nutrient contents were sampled. As a consequence, nutrient gradients in these areas were lower and narrower than in the other 2 areas. Dispersed human settlements and extensive farming in some areas in northern Spain (CC, CLC) allowed us to meet the required low nutrient-enrichment gradients. Water variables Water temperature was monitored continuously with ACR Smart-Button (ACR Systems Inc., Surrey, British Columbia) or HOBO Pendant (Onset Comput- er Corporation, Bourne, Massachusetts) temperature loggers throughout the study period (autumn–winter 2007–2008) in all streams. Conductivity, pH, and dissolved O2 (WTW multiparametric sensor) were measured in situ, and water samples were taken for nutrient analyses on each sampling date (n = 6). Nutrient analyses were done on water filtered through precombusted glass-fiber filters (Whatman GF/F). NO3 2 concentration was determined by ion chromatography (COMPACT IC1.1; Metrohm, Her- isau, Switzerland) or with the sodium salicylate method (Monteiro et al. 2003). NH4 + was measured with the manual salicylate method (Krom 1980), NO2 2 with the sulphanylamide method, soluble reactive P (SRP) with the molybdate method, and alkalinity through titration to an end pH of 4.5 (APHA 2005). Litter bags and decomposition Alder (Alnus glutinosa (L.) Gaertner) leaves were used as a standard substrate to measure decomposi- tion rates. All leaf litter used was collected in CC to prevent local differences in the initial quality of materials (Lecerf and Chauvet 2008b). Leaves were collected from the forest soil just after abscission in autumn 2007 and air-dried to constant mass. Five grams (6 0.25) of alder leaves were weighed, moistened (spray), and enclosed in mesh bags (15 3 20 cm, 5-mm mesh). Leaf bags (25 in each stream) were tied with nylon lines to iron bars driven into the stream bed along 50-m reaches. Extra sets of 5 bags were immersed in the streams for 24 h and used to correct the initial mass values for leaching. Such a correction is made to better describe processing dynamics once labile compounds have disappeared from leaves (Suberkropp and Chauvet 1995, Ferreira et al. 2006). Leaf incubation was initiated in late autumn (November–December) 2007 to coincide with the seasonal peak in leaf fall. Five bags were retrieved after 7 d (t7) and at dates that roughly corresponded to losses of 20 (t20), 35 (t35), 50 (t50), and 70% (t70) of the initial mass, as estimated from exponential decomposition rates (k) recalculated from previous data at each experimental FIG. 1. Locations of the 4 study regions in the Iberian Peninsula: Cordillera Cantábrica (CC), Cordillera Litoral Catalana (CLC), Sierra de Guadarrama (SG), and Sierra Nevada (SN). 2011] LEAF DECOMPOSITION IN HEADWATER STREAMS 937 T A B L E 1. L o ca ti o n , m ea n an n u al cl im at e v ar ia b le s, ca tc h m en t ar ea , an d la n d u se ch ar ac te ri za ti o n o f st u d ie d st re am s. A b so lu te m in im u m an d m ax im u m v al u es fo r ea ch v ar ia b le ar e in b o ld . D is ch ar g e v al u es ar e ra n g es fo r th e p er io d o f st u d y . C C = C o rn is a C an tá b ri ca , C L C = C o rd il le ra L it o ra l C at al an a, S G = S ie rr a d e G u ad ar ra m a, S N = S ie rr a N ev ad a. S it e L at it u d e L o n g it u d e A lt it u d e (m as l) T em p er at u re (u C ) P re ci p it at io n (m m ) C at ch m en t ar ea (h a) L an d u se (% ) C h an n el w id th (m ) R ip ar ia n tr ee co v er (% ) U p st re am ch an n el sl o p e (% ) D is ch ar g e (L / s) N at iv e v eg et at io n A ff o r- es te d A g ri cu l- tu re C C 1 42 u5 9 9 59 .6 4 0 N 2u 52 95 9. 92 0W 41 5 11 .1 9 2 3 40 0 96 .0 4. 0 0 .0 4. 2 97 .5 11 .7 20 –5 4 C C 2 43 u1 0 9 12 .7 2 0 N 2u 53 92 6. 30 0W 1 0 0 14 .0 87 3 28 0 9. 4 81 .6 9. 0 3. 3 95 .3 8. 8 14 –4 7 C C 3 43 u0 7 9 09 .8 4 0 N 2u 54 93 4. 49 0W 15 0 12 .5 85 0 50 4 4. 8 89 .5 5. 8 3. 7 9 8 .2 14 .5 11 –1 10 C C 4 43 u0 6 9 16 .9 2 0 N 2u 54 92 1. 17 0W 16 5 12 .5 85 0 28 5 0 .6 83 .5 15 .9 3. 6 94 .3 11 .0 4– 57 C C 5 43 u0 8 9 52 .8 0N 2u 50 90 .4 6 0 W 14 5 14 .0 87 3 53 4 1. 1 83 .3 15 .6 3. 9 89 .3 9. 0 23 –1 64 C C 6 43 u0 8 9 59 .2 8 0 N 2u 51 90 .3 6 0 W 12 5 14 .0 87 3 23 7 1. 2 88 .3 10 .5 3. 6 88 .1 11 .9 4– 71 C L C 1 41 u2 9 9 38 .0 8 0 N 2u 16 91 5. 06 0E 49 5 11 .4 61 1 11 65 71 .5 26 .8 1. 7 5. 1 97 .3 18 .0 18 –5 5 C L C 2 41 u2 7 9 54 .1 8 0 N 2u 16 94 1. 27 0E 11 22 11 .4 61 1 24 1 91 .3 5. 5 3. 2 7. 4 92 .0 16 .5 13 –2 5 C L C 3 41 u2 7 9 44 .8 2 0 N 2u 09 94 8. 49 0E 44 5 11 .9 61 1 14 20 88 .2 3. 2 8. 6 6. 8 91 .6 8 .1 2 –1 2 C L C 4 41 u2 7 9 27 .0 7 0 N 2u 09 94 0. 25 0E 44 6 11 .9 61 1 53 0 82 .7 0. 0 1 7 .3 4. 8 79 .7 14 .6 8– 15 S G 1 40 u5 2 9 19 .2 0 0 N 3u 06 93 8. 91 0W 13 80 9 .0 77 2 17 8 48 .9 36 .7 12 .3 2. 5 81 .1 27 .5 2 –7 2 S G 2 40 u2 9 9 13 .9 2 0 N 3u 05 92 4. 18 0W 13 00 9 .0 77 2 1 7 3 53 .1 37 .8 7. 9 1. 9 82 .0 24 .7 3– 21 S G 3 40 u4 6 9 37 .9 2 0 N 4u 0 9 40 .5 8 0 W 13 90 9 .0 77 2 17 5 19 .7 80 .5 0 .0 2. 3 66 .6 2 9 .2 13 -2 6 S G 4 40 u5 4 9 59 .0 4 0 N 4u 07 90 0 0 W 13 00 9 .0 77 2 80 3 5. 2 9 3 .4 0 .0 1 .4 90 .0 17 .1 12 –4 4 S G 5 40 u4 6 9 23 .1 6 0 N 3u 90 96 6 0 W 12 20 9 .0 77 2 16 66 7. 1 16 .4 6. 9 1 1 .2 64 .1 13 .6 6 6 –4 0 6 S G 6 40 u4 7 9 15 .0 0 0 N 3u 83 92 5 0 W 11 90 9 .0 77 2 11 29 26 .1 57 .0 0. 9 9. 4 83 .4 20 .0 24 –2 08 S G 7 40 u3 4 9 58 .8 0 0 N 3u 99 91 1 0 W 14 00 9 .0 77 2 53 2 30 .7 45 .1 0. 6 3. 9 86 .0 21 .9 27 –1 11 S N 1 36 u3 4 9 55 .5 6 0 N 3u 11 90 8 0 W 11 30 1 6 .4 45 6 4 3 5 0 61 .8 22 .6 14 .6 4. 7 80 .3 12 .5 35 –2 63 S N 2 36 u3 5 9 31 .2 0 0 N 3u 09 90 0 0 W 11 20 1 6 .4 45 6 41 74 64 .6 23 .9 11 .5 3. 8 50 .0 10 .8 50 –1 60 S N 3 37 u0 2 9 38 .0 4 0 N 3u 00 94 0. 58 9W 1 6 8 0 1 6 .4 45 6 12 43 45 .9 46 .9 6. 4 2. 2 3 1 .6 12 .2 23 –6 2 S N 4 37 u0 6 9 56 .8 8 0 N 3u 09 90 3. 93 0W 13 16 12 .9 3 1 0 18 50 9 9 .0 0 .0 1. 0 2. 8 67 .1 14 .7 64 –9 1 S N 5 37 u0 5 9 08 .2 3 0 N 3u 02 93 2. 89 0W 14 60 12 .9 3 1 0 19 20 52 .1 47 .8 0. 1 2. 9 81 .1 14 .0 48 –1 25 938 J. POZO ET AL. [Volume 30 site (Mendoza-Lera et al. 2010). t70 was achieved between 46 (SG) and 113 (CLC) d. Initial mass refers to initial ash-free dry mass (AFDM) corrected for leaching. After retrieval, litter bags were placed in individual plastic bags and transported in refriger- ated containers to the laboratory where they were processed immediately. Leaf material from each bag was rinsed with filtered stream water, and the fauna and mineral particles were separated from the leaf litter on a 200-mm sieve. Only fauna from t50 samplings, which generally coincided with coloniza- tion peaks in alder leaves (Hieber and Gessner 2002), were preserved in 70% ethanol for later analysis. Invertebrates were identified to family level under a dissecting microscope, counted, and sorted into functional feeding groups (the most representative for the family) according to Merritt and Cummins (1996) and Tachet et al. (2002). Fungal sporulation rate was determined at t20 (2–3 wk after immersion), which often coincides with the peak of conidial production on alder leaves (Pascoal and Cássio 2004), from 1 set of 5 leaf disks (12-mm diameter) punched from each bag with a cork borer (see below). The remaining leaf material was, as on the other sampling dates, oven- dried (70uC, 72 h) and weighed. A portion was used for nutrient analyses, and the rest was combusted (550uC, 4 h) to determine AFDM. Leaf material for nutrient analyses (C, N, and P) was ground into fine powder (1-mm screen). C and N were determined with a Perkin Elmer series II CHNS/O elemental analyser (Perkin Elmer, Norwalk, Connecticut). P was determined spectrophotometri- cally after mixed acid digestion (molybdenum blue method; Allen et al. 1974). Results were expressed as % leaf-litter dry mass (DM) as it was analyzed and as molecular elemental ratios (C:N, C:P, and N:P). Sporulation of aquatic hyphomycetes (AH) Leaf disks from each bag from t20 were incubated in 100-mL Erlenmeyer flasks with 25 mL of filtered stream water (Whatman GF/F) on a shaker (60 rpm) for 48 h at 10uC. The resulting conidial suspensions were transferred into 50-mL centrifuge tubes and fixed with 2 mL of 37% formalin. An aliquot of the suspension was filtered (Millipore SMWP 5-mm pore size) for conidial identification and counting. Each filter was stained with trypan blue in lactic acid (0.05%), and conidia were identified (after Gulis et al. 2005 and species description protologues) and count- ed under a microscope at 2503. Counting effort was reduced with the assistance of voice recognition and Excel data-entry generator software developed by one of us (OM). Leaf-disk DM was determined as described above for the bulk leaf material. Sporulation rates were expressed as number of conidia produced per mg leaf DM per day of in vitro incubation time. Statistical analyses The relationship between % AFDM remaining (response variable) and elapsed time (predictor variable) was fitted to linear (Mt = M0 2 bt) and exponential models (Mt = M0e2kt), where M0 is the initial AFDM corrected for leaching, Mt is the remaining AFDM at time t, and b is the linear and k the exponential decomposition rate. Streams differed in temperature (Table 2), so breakdown rates were calculated with each model in terms of time (d) and accumulated heat, the sum of mean daily tempera- tures accumulated by the sampling day (degree days [dd]) (Stout 1989). The goodness of fit of the models was evaluated by calculating the coefficient of determination (R2) after expressing the values of the response variable of both models in the same original scale (Kvålseth 1985, Quinn and Keough 2002). Slopes (breakdown rates) were compared with nested anal- ysis of covariance (ANCOVA; %AFDM as dependent variable, streams nested within region and region as factors, and dd as covariate). The ANCOVA model was considered as a particular case of a linear mixed model to account for the correlation resulting from the clustered design of successive measurements at each site (Verbeke and Molenberghs 2000). The model was fitted using the method of restricted maximum likelihood (REML), and the covariance structure used was a 1st-order autoregressive (AR[1]). For other variables (e.g., invertebrates, sporulation rates), com- parisons were carried out by nested analysis of variance (ANOVA; streams nested within region and region as factors; sampling time was a factor when leaf nutrient content was compared). Subse- quent pairwise comparisons were made with Tukey’s test (Zar 2010). The influence of several variables that could be potential predictors of the breakdown rate also was tested. Simple linear regression was fitted indepen- dently for all variables with the breakdown rate as response variable. Region could have influenced the relationships between the selected variables and the breakdown rate, so these regressions were repeated with region as a factor, and the significance of the covariate in the resulting ANCOVA was examined. Simple linear regression and correlation were used when searching for relationships between variables other than breakdown rate. The Shapiro–Wilk test was used to assess normality, and transformation was done when necessary. An arcsine(x) transformation 2011] LEAF DECOMPOSITION IN HEADWATER STREAMS 939 T A B L E 2. M ea n (6 S E ; n = 5– 6) v al u es o f p h y si co ch em ic al v ar ia b le s an d d ai ly m ea n te m p er at u re (r an g e) d u ri n g th e ex p er im en ts . A b so lu te m in im u m an d m ax im u m v al u es fo r ea ch v ar ia b le ar e in b o ld . S R P = so lu b le re ac ti v e P . S it e W at er te m p er at u re (u C ) S R P (m g P / L ) N H 4 + (m g N / L ) N O 2 2 (m g N / L ) N O 3 2 (m g N / L ) p H A lk al in it y (m eq / L ) C o n d u ct iv it y (m S / cm ) C C 1 7. 5 (4 .5 –9 .6 ) 13 .7 6 3. 2 15 .5 6 4. 1 2. 3 6 0. 7 14 7. 3 6 44 .4 7. 61 6 0. 06 0. 74 6 0. 02 13 1. 4 6 5. 8 C C 2 8. 8 (5 .1 –1 1. 3) 24 .7 6 3. 3 20 .6 6 5. 6 1. 6 6 0. 5 29 3. 5 6 40 .1 7. 94 6 0. 07 1. 83 6 0. 06 31 6. 8 6 26 .1 C C 3 8. 2 (4 .2 –1 0. 6) 25 .5 6 7. 5 27 .8 6 4. 4 5. 8 6 1. 6 41 4. 1 6 12 3. 3 8 .1 9 6 0. 01 2. 02 6 0. 02 26 5. 8 6 32 .3 C C 4 8. 3 (4 .9 –1 0. 6) 22 .6 6 6. 7 60 .6 6 16 .5 12 .7 6 2. 6 71 7. 1 6 90 .3 7. 64 6 0. 10 0. 73 6 0. 05 14 9. 5 6 16 .1 C C 5 9. 6 (4 .6 –1 1. 4) 39 .9 6 4. 9 95 .0 6 29 .6 3. 8 6 0. 6 94 7. 6 6 12 3. 2 8. 14 6 0. 06 2. 49 6 0. 12 3 5 1 .4 6 16 .9 C C 6 1 0 .1 (7 .8 –1 2. 0) 5 1 .7 6 8. 8 91 .5 6 29 .4 1 9 .3 6 9. 6 1 1 5 1 .2 6 14 3. 9 8. 06 6 0. 04 2. 28 6 0. 11 32 3. 6 6 31 .8 C L C 1 6. 2 (3 .8 –8 .2 ) 25 .1 6 4. 4 58 .2 6 19 .5 9. 5 6 4. 5 77 7. 9 6 13 0. 3 7. 26 6 0. 02 2. 09 6 0. 05 21 0. 6 6 1. 4 C L C 2 5. 1 (3 .2 –6 .8 ) 37 .1 6 1. 3 40 .7 6 12 .1 2. 1 6 0. 6 38 .7 6 12 .3 7. 75 6 0. 06 0. 53 6 0. 08 63 .4 6 2. 2 C L C 3 4. 7 (1 .4 –8 .4 ) 31 .2 6 3. 5 1 8 7 .4 6 44 .1 15 .7 6 3. 3 24 3. 0 6 24 .2 7. 72 6 0. 07 3 .2 3 6 0. 07 33 0. 8 6 5. 1 C L C 4 5. 6 (4 .4 –7 .9 ) 11 .7 6 2. 0 65 .1 6 22 .1 1. 7 6 0. 7 11 35 .2 6 29 9. 8 7. 32 6 0. 10 2. 68 6 0. 16 30 3. 6 6 3. 9 S G 1 2. 8 (0 .9 –6 .0 ) 10 .6 6 3. 4 21 .6 6 11 .5 2. 2 6 0. 8 15 7. 7 6 22 .0 6. 78 6 , 0. 01 0. 16 6 0. 03 1 3 .2 6 0. 4 S G 2 2. 9 (0 .5 –6 .6 ) 9. 8 6 1. 0 28 .2 6 14 .7 2. 8 6 0. 7 48 1. 2 6 72 .7 6. 78 6 , 0. 01 0. 18 6 0. 02 17 .5 6 0. 8 S G 3 3. 8 (1 .4 –6 .9 ) 15 .5 6 0. 9 16 .2 6 12 .8 4. 8 6 3. 2 27 4. 7 6 5. 3 6. 69 6 0. 04 0. 31 6 0. 04 32 .3 6 1. 7 S G 4 5. 1 (3 .8 –7 .9 ) 12 .3 6 1. 1 5. 4 6 2. 2 7. 1 6 4. 4 33 4. 5 6 25 .5 6. 80 6 0. 05 0. 53 6 0. 05 54 .8 6 4. 3 S G 5 3. 0 (0 .5 –6 .3 ) 7. 3 6 0. 7 23 .4 6 11 .3 0. 7 6 0. 1 23 6. 3 6 33 .5 6 .5 9 6 0. 02 0 .1 4 6 0. 02 14 .0 6 0. 8 S G 6 4. 5 (2 .1 –7 .3 ) 8. 5 6 1. 8 0 .3 6 0. 6 0. 9 6 0. 1 20 6. 2 6 24 .7 6. 60 6 , 0. 01 0. 16 6 0. 02 18 .3 6 0. 6 S G 7 3. 4 (1 .1 –6 .4 ) 9. 4 6 1. 8 5. 6 6 4. 1 0 .5 6 0. 1 15 7. 8 6 24 .7 6. 60 6 , 0. 01 0. 18 6 0. 02 18 .0 6 0. 8 S N 1 7. 3 (5 .0 –9 .1 ) 5. 1 6 0. 5 15 .3 6 2. 7 1. 1 6 0. 2 10 5. 3 6 26 .8 6. 94 6 0. 04 1. 26 6 0. 26 21 4. 6 6 42 .2 S N 2 6. 5 (4 .6 –8 .3 ) 8. 0 6 0. 5 17 .1 6 1. 2 0. 8 6 0. 1 4 .0 6 0. 8 7. 32 6 0. 03 0. 60 6 0. 03 12 8. 2 6 5. 7 S N 3 3. 4 (1 .8 –5 .7 ) 1 .1 6 0. 1 21 .1 6 2. 8 0. 9 6 0. 1 13 4. 9 6 4. 4 7. 08 6 0. 06 0. 25 6 0. 01 68 .0 6 1. 4 S N 4 2 .7 (0 .7 –5 .1 ) 15 .7 6 1. 2 11 .3 6 3. 6 1. 5 6 0. 2 19 7. 4 6 15 .7 7. 60 6 0. 04 0. 78 6 0. 02 10 8. 0 6 0. 8 S N 5 3. 2 (1 .2 –5 .0 ) 1. 9 6 0. 1 16 .8 6 2. 0 1. 1 6 0. 1 13 9. 1 6 7. 5 7. 19 6 0. 07 0. 34 6 0. 02 55 .0 6 , 0. 1 940 J. POZO ET AL. [Volume 30 was applied to percentage data, and !(x) or log(x) transformations were used in the other cases. All analyses were undertaken with SPSS 17.0 (SPSS Inc., Chicago, Illinois) and SAS 9.2 (SAS Institute Inc., Cary, North Carolina). Results Nutrient levels were relatively low (Table 2), but water physicochemical variables differed noticeably within and among regions (nested ANOVA, p , 0.05; Table 3). The highest values of physicochemical variables were found in CC, the most oceanic region according to the GCI (see Study sites), whereas the lowest were registered in SG and SN, the most continental ones. Alder-leaf mass loss corrected for leaching (mean leaching loss < 18%) fit a linear model better than an exponential model in terms of d (R2 values higher in all 22 cases) and dd (R2 higher in 21 of 22 cases) (Table 4). The linear rate based on dd, which corrected for interregional temperature differences, was used for analyses of spatial variation and relationships to environmental variables. Breakdown rates (% mass lost/dd) differed significantly among and within regions (nested ANCOVA, p , 0.001; Table 5) and were high in SG, intermediate in SN, and low in CLC and CC. The initial quality of the leaf material was the same for every stream (% composition 6 SE before leaching, n = 5: C = 47.1 6 1.3; N = 2.48 6 0.14; P = 0.081 6 0.003). Percent C varied little throughout the study (mean CV for all sites < 5%). However, N and P varied noticeably (mean CV < 10% for N and 20% for P). As a consequence, variation observed for C:nutrient ratios mainly depended on changes in N or P in the leaf material. Furthermore, C:nutrient ratios tended to decrease as the dissolved nutrient concen- tration (both N and P) increased (Fig. 2A, B), but differences among regions were not significant (nest- ed ANOVA, p . 0.05). At 19 of 22 sites, leaves lost P relative to C, but this loss decreased as the dissolved P increased (Fig. 2A). In contrast, at all sites, leaves gained N relative to C, i.e., mean C:N decreased during processing. This N enrichment increased as dissolved N increased in streams (Fig. 2B). Mean AH sporulation rates differed greatly among and within regions (nested ANOVA, p , 0.01, SG = SN , CLC = CC). Values ranged from ,0.01 conidium mg21 d21 (SG) to close to 6 (CC) (Table 6) and were positively correlated with variables related to dissolved solids, such as alkalinity (r = 0.747, p , 0.01), conductivity (r = 0.726, p , 0.01) and SRP (r = 0.532, p , 0.05), and temperature (r = 0.558, p , 0.01). A total of 42 identifiable taxa of AH were found (Table 6): 28 in CC, 25 in CLC, 23 in SN, and 19 in SG. The 4 regions had 9 species in common, and Flagellospora curvula was dominant. Macroinvertebrate abundance differed among and within regions (nested ANOVA, p , 0.001, SG , CLC , SN = CC) and was represented by 49 families: 41 in CC, 31 in SN, 27 in SG, and 19 in CLC (Table 7). Shredders were represented by 11, 9, 9, and 7 families, respectively, and were an important component of macroinvertebrate assemblages in terms of abundance in all regions (Table 7). Linear regression analyses between breakdown rate and associated variables showed that conductivity, alkalinity, pH, sporulation rate (negative slope), and channel slope (positive) were the most important predictors of breakdown rate (Table 8). Several variables had extreme values in SG, so regressions were repeated with region as a factor. Conductivity was the most significant variable explaining break- down rate (Table 8), and the region factor was not significant (p = 0.205). Channel slope, shredder density (positive), and alkalinity (negative) were the other variables significantly related to the decay rate (Table 8). Thus, the significant effect of shredders on leaf processing appeared when the influence of region was corrected. Furthermore, when sites from SG were excluded from the regression analysis without region as a factor, the significant effect of shredders also was clear (Fig. 3). The negative highly significant relation- ship between sporulation rate and breakdown rate was clearly influenced by region and disappeared when the regression was adjusted by region (Table 8). No positive relationships were found between nutri- ents and breakdown rate. However, the effect of microbial activity on breakdown rate appeared to depend on concentrations of dissolved nutrients (see results of elemental ratios above). Discussion The main goal of our work was to elucidate factors responsible for regional differences in leaf-litter decomposition rates among oligotrophic headwater streams. Headwater streams in our study were similar and drained siliceous catchments within a narrow low-to-moderate range of dissolved nutrient concen- trations, particularly P. Despite the similarities among streams, inter- and intraregional differences in water- column physicochemical characteristics were appar- ent. In general, nutrient concentrations and their variability decreased with altitude. Before identifying relationships between leaf pro- cessing and environmental variables it was necessary 2011] LEAF DECOMPOSITION IN HEADWATER STREAMS 941 to calculate breakdown rates according to the model that yielded the best fits. In many studies, leaf breakdown is an exponential function of elapsed time (Webster and Benfield 1986, Abelho 2001), but in our study, the correction for leaching resulted in better fits with single linear regressions. Therefore, we focused on linear instead of exponential rates. Exponential rates also are reported because they are frequent in the literature (Irons et al. 1994, Lecerf and Chauvet 2008b). As expected, breakdown rates were noticeably variable within and among regions. Increases in temperature are assumed to enhance biological activities, leaf processing included (Webster and Benfield 1986, Bergfur 2007). However, contrary to our hypothesis, temperature was not a determining factor of breakdown rate. The fastest rates, regardless of the model used for their calculation, were found where mean water temperature and accumulated heat (dd) were the lowest (SG). The range of mean water temperatures among sites (2.7–10.1) should have been great enough to generate important differences in the activity of detritivores and decomposers and in leaf- litter processing rates (Friberg et al. 2009). However, as has been noted in other studies (Fleituch and Leichtfried 2007), other factors must have masked its effects. Gonçalves et al. (2006) found faster decay rates in a temperate stream than in a Mediterranean stream (both on the Iberian Peninsula) or a Neotropical stream. They suggested that the differences might be related to consumer efficiency and proposed that biological differences overrode the temperature effect. If using degree-days eliminates the effect of differing thermal regimes, rates should be similar across latitudes, unless other factors are involved (Irons et al. 1994). When we expressed breakdown rates on a degree-day basis, differences between regions with the warmest and the coldest streams were even greater, as has been observed by others (Hladyz et al. 2010). Thus, other factors in our study were more important than temperature in determin- ing breakdown rates. Similarly, rates (/dd) were much faster in an Alaskan stream than in streams in Costa Rica and Michigan (Irons et al. 1994), results suggesting that interregional differences in litter breakdown rates, as in our study, are not merely consequences of shifts in water temperature. The main factors significantly related to leaf breakdown rate were conductivity, alkalinity (nega- tively), and channel slope (positively). The negative relationship between decay rate and conductivity (or alkalinity) is difficult to explain, and positive rela- tionships are more frequent in the literature (Young et al. 2008). We did find opposite trends at some sites (data not shown), but when the regression analysis was adjusted by region the significant relationship persisted. The negative relationship between decay rate and conductivity might be, in part, an indirect consequence of the effect of channel slope on decay rate because both variables were highly correlated. Channel slope affects water velocity and particle transport, which contribute to physical abrasion on leaves, accelerating leaf fragmentation (Paul et al. 2006) and masking the effect of moderate dissolved nutrient concentration in headwaters (Spänhoff et al. 2007). The importance of physical abrasion on leaf breakdown is context dependent, and some authors have reported that the effect of physical abrasion is trivial compared with the effects of biotic drivers TABLE 3. Results of the nested analyses of variance for physicochemical variables during the experiments. SRP = soluble reactive P. Variable Source of variation df1 df2 F p Water temperature Region 3 18 20.87 ,0.001 Site(region) 18 95 8.26 ,0.001 SRP Region 3 18 10.45 ,0.001 Site(region) 18 95 8.07 ,0.001 NH4 + Region 3 18 10.08 ,0.001 Site(region) 18 95 3.58 ,0.001 NO2 – Region 3 18 5.32 0.008 Site(region) 18 95 6.03 ,0.001 NO3 – Region 3 18 3.38 0.041 Site(region) 18 95 17.86 ,0.001 pH Region 3 18 41.47 ,0.001 Site(region) 18 95 17.52 ,0.001 Alkalinity Region 3 18 14.12 ,0.001 Site(region) 18 95 60.33 ,0.001 Conductivity Region 3 18 20.04 ,0.001 Site(region) 18 95 36.53 ,0.001 942 J. POZO ET AL. [Volume 30 (Hieber and Gessner 2002, Ferreira et al. 2006, Hladyz et al. 2009). In our study, the significance of channel slope diminished and that of shredders appeared when the analysis was corrected by region (Table 8). In contrast, particle sedimentation is a factor com- monly suggested to slow leaf breakdown because deposition of fine sediment on litterbags can limit microbial and macroinvertebrate activity (Zweig and Rabeni 2001, Niyogi et al. 2003, Rabeni et al. 2005, Mesquita et al. 2007, Spänhoff et al. 2007) and, thus, reduce processing rates. We only have indirect measures of this effect (% ash content of leaf litter in the bags), but the significant negative regression between ash content and breakdown rate point to a negative effect of fine sediment on leaf processing. Shredder density positively influenced breakdown rates when SG data were excluded from the regres- sion analyses. The order of mean shredder densities TABLE 4. Mean (SE) leaf-litter breakdown rates, linear b and exponential k, of alder leaves in terms of time (d) and accumulated heat (degree days [dd]). Bold indicates maximum site R2. Site Linear model, b Exponential model, k (% AFDM/d) (% AFDM/dd) (/d) (/dd) Mean SE R2 Mean SE R2 Mean SE R2 Mean SE R2 CC1 1.45 0.09 0.922 0.198 0.013 0.918 0.036 0.004 0.829 0.0049 0.0005 0.846 CC2 1.11 0.06 0.940 0.128 0.007 0.939 0.022 0.002 0.889 0.0026 0.0002 0.891 CC3 1.44 0.11 0.879 0.184 0.015 0.881 0.035 0.004 0.774 0.0045 0.0005 0.777 CC4 1.46 0.09 0.913 0.179 0.012 0.909 0.032 0.003 0.905 0.0040 0.0003 0.902 CC5 1.03 0.06 0.934 0.109 0.006 0.942 0.024 0.003 0.844 0.0025 0.0003 0.856 CC6 1.34 0.07 0.953 0.135 0.008 0.954 0.026 0.002 0.945 0.0026 0.0002 0.944 CLC1 1.26 0.11 0.844 0.207 0.018 0.852 0.033 0.007 0.573 0.0055 0.0010 0.582 CLC2 1.57 0.10 0.911 0.321 0.021 0.918 0.044 0.007 0.639 0.0091 0.0014 0.656 CLC3 0.44 0.03 0.912 0.094 0.007 0.886 0.010 0.000 0.904 0.0014 0.0001 0.892 CLC4 0.53 0.09 0.621 0.090 0.016 0.607 0.011 0.004 0.339 0.0020 0.0006 0.362 SG1 1.65 0.13 0.842 0.639 0.053 0.842 0.053 0.007 0.700 0.0214 0.0024 0.746 SG2 1.38 0.13 0.796 0.496 0.048 0.794 0.031 0.006 0.535 0.0113 0.0020 0.552 SG3 1.62 0.14 0.827 0.432 0.040 0.810 0.053 0.011 0.482 0.0148 0.0028 0.508 SG4 1.85 0.15 0.851 0.369 0.030 0.852 0.074 0.012 0.581 0.0154 0.0024 0.610 SG5 1.41 0.15 0.768 0.496 0.047 0.803 0.035 0.007 0.487 0.0128 0.0022 0.549 SG6 1.31 0.11 0.833 0.302 0.027 0.823 0.029 0.005 0.569 0.0069 0.0012 0.584 SG7 1.62 0.13 0.844 0.482 0.039 0.851 0.037 0.007 0.579 0.0112 0.0018 0.599 SN1 1.05 0.07 0.904 0.143 0.010 0.906 0.023 0.004 0.624 0.0031 0.0005 0.633 SN2 1.30 0.06 0.946 0.198 0.011 0.944 0.027 0.003 0.838 0.0041 0.0004 0.841 SN3 0.74 0.04 0.936 0.226 0.013 0.934 0.013 0.002 0.844 0.0039 0.0004 0.843 SN4 0.96 0.06 0.922 0.359 0.023 0.921 0.019 0.003 0.738 0.0073 0.0010 0.731 SN5 0.83 0.04 0.946 0.265 0.014 0.941 0.014 0.002 0.873 0.0045 0.0004 0.865 FIG. 2. Relationship between alder C:nutrient ratio (molecular elemental ratio) and soluble reactive P (SRP) (A), dissolved inorganic N (DIN) (B). Broken lines correspond to the value of each C:nutrient ratio in leaves before leaching. 2011] LEAF DECOMPOSITION IN HEADWATER STREAMS 943 among the 3 regions was opposite that of mean conductivities, indicating that headwaters usually had low conductivity concurrent with greater shred- der density (relationship not statistically significant). The highest breakdown rates in our study occurred in SG and SN, the regions with the coldest temper- atures and the lowest nutrient levels. Irons et al. (1994) suggested the relative importance of inverte- brates vs microorganisms changes along a latitudinal gradient, with invertebrates more important in colder waters at high latitudes (or high altitudes). If this suggestion is correct, shredders should play a decisive role on leaf breakdown in SG and SN, especially if, as in our case, fungal activity were limited (indicated by low sporulation rates). Some investigators have shown that cold waters can favor some shredders like stoneflies and caddisflies that are adapted to cooler thermal regimes (Danks 2007). This situation could exert a key role on leaf processing and would help explain the faster breakdown rates in the colder areas. Caddisflies and stoneflies were well represented in SG and SN. The degree of eutrophication of our streams was low, but we expected leaf breakdown rates to respond to increases in dissolved nutrients because of enhanced microbial activity (Pozo 1993, Suberkropp and Chauvet 1995, Gulis and Suberkropp 2003). However, neither dissolved nutrients (N and P) nor sporulation rate were positively related to breakdown rate. Poor relation- ships between leaf breakdown rates and water-column nutrients have been found elsewhere. For instance, in a study done along a gradient of water-column nutrient enrichment in south-central Sweden, Bergfur (2007) found little support for the conjecture that decompo- sition rates were related to nutrient enrichment. Perhaps, the potential effects of eutrophication in our low-nutrient, low-variability system were overridden by other factors with more important interregional variation, such as density of shredders. Sporulation rates were positively related to dis- solved solids (alkalinity, conductivity, and SRP) but not with breakdown rate (sporulation rates were highest where breakdown rates were lowest). Sporu- lation rates were measured only on one occasion for each stream, but we assumed that the values could be compared. The time elapsed from implantation to t20 differed among regions, but, in most cases (20 of 22), it was 12 to 23 d, a period of high fungal spore production (including peaks) by AH when a great amount of the incubated leaf litter still remains (Chauvet et al. 1997). This period seems to coincide with the growth phase of mycelia (measured as ergosterol) on leaf litter (Pozo et al. 1998). The relationships between fungal activity, nutrient avail- ability, and leaf decomposition in nutrient-poor waters are probably complex, but the effects on elemental ratios of leaf litter probably are related to microbial activity. All leaves used in the experiments came from the same location, so the variations in leaf quality (N and P content) during breakdown were in response to the local availability of dissolved nutri- ents. Nevertheless, quality acquired (as C:N, C:P, and N:P) and processing rates measured were not parallel. According to Artigas et al. (2008), fungal N demands for sporulation can be fulfilled at levels of dissolved NO3 2 ,300 mg N/L, and no enhancement should be expected with increased dissolved nutrients. How- ever, in streams with low concentrations of dissolved inorganic N (,40 mg/L) and P (,16 mg/L), leaf decomposition and sporulation rates were stimulated only when both nutrients were added together, which suggests that these nutrients potentially colimited fungal activity (Grattan and Suberkropp 2001). Grat- tan and Suberkropp (2001) also reported that when N concentrations were .65 mg/L, decomposition and sporulation rates were stimulated by addition of P to waters with P concentrations ,5 mg/L. In our study, dissolved NO3-N was .65 mg/L in most cases, and mean dissolved PO4-P created a gradient from ,5 mg/L to 52 mg/L. These concentrations were high enough to elicit a response in both decomposition and sporulation rates according to Grattan and Suberkropp (2001), but we observed a response only of sporulation rates. In more eutrophic streams, microbial breakdown rate and spore production are not predictable. Both positive and negative effects have been reported in the literature, but a reduction of species richness of AH involved in leaf processing is often observed in eutrophic streams (Lecerf and Chauvet 2008a). In our study, differences in AH species richness might not be a consequence of impairment but of natural forces because SG, a region characterized by its nutrient-poor waters, circumneutral pH, and low temperature, had the lowest richness. On the other hand, the enhancement of breakdown rates by increases in dissolved nutrients seems to depend on leaf quality (Molinero et al. 1996), which could explain why the decay rate of a high-quality TABLE 5. Results of the nested analysis of covariance for the linear rates in terms of degree days. Source of variation df1 df2 F p Degree days 1 528 3607.1 ,0.001 Region 3 degree days 3 528 198.9 ,0.001 Site(region) 3 degree days 18 528 19.4 ,0.001 944 J. POZO ET AL. [Volume 30 TABLE 6. Sporulation rates (minimum–maximum) for each aquatic hyphomycete taxon or form (no. mg21 leaf dry mass d21). Site abbreviations are given in Table 1. Taxon CC CLC SG SN Alatospora acuminata ‘‘pulchelloid’’a 0.001–0.289 0–0.032 0–,0.001 0–,0.001 Alatospora acuminata sensu neotypea ,0.001–0.005 0–,0.001 0–,0.001 0–0.001 Alatospora acuminata ‘‘subulate’’a 0.011–0.35 ,0.001–0.552 0–,0.001 0–0.002 Alatospora flagellata (J. Gönczöl) Marvanová 0–0.010 Alatospora pulchella Marvanová ,0.001–0.007 0–0.001 0–,0.001 Anguillospora filiformis Greathead 0–0.009 Anguillospora furtiva Descals 0–,0.001 Anguillospora longissima (Sacc. & P. Syd.) Ingold 0.001–0.035 0–0.001 0–,0.001 Anguillospora rosea Descals & Marvanová 0–,0.001 Articulospora tetracladia (Tubaki) Sv. Nilsson 0.001–0.007 0–0.005 0–0.024 0–0.001 Clavariopsis aquatica De Wild 0.001–0.145 0–0.020 0–0.001 0–0.008 Clavatospora longibrachiata Marvanová & Sv. Nilsson 0–0.052 0–0.019 ,0.001–0.008 Crucella subtilis Marvanová & Suberkr. 0–0.025 Culicidospora aquatica R. H. Petersen 0–0.081 Flagellospora curvula Ingold 0.003–5.966 0.011–2.377 ,0.001–0.330 0.119–1.836 Geniculospora grandis (Greathead) Sv. Nilsson Ex Nolan 0–0.061 0–,0.001 Geniculospora inflata Marvanová & Sv. Nilsson 0–0.010 Goniopila monticola/Margaritispora aquaticab 0–0.001 Heliscella stellata (Ingold & Cox) Marvanová 0–0.858 0–0.188 0–0.035 Heliscus lugdunensis Sacc. & Thérry 0.001–0.004 0–0.013 ,0.001–0.014 ,0.001–0.003 Heliscus tentaculus Umphlett 0–0.003 Lemonniera alabamensis Sinclair & Morgan 0–0.006 0–0.084 0–0.009 Lemonniera aquatica De Wild 0–,0.001 0–0.004 0–0.001 Lemonniera centrosphaera Marvanová 0–,0.001 Lemonniera cornuta Ranzoni 0–0.004 ,0.001–0.143 0–0.005 Lemonniera filiformis R. H. Petersen Ex Dyko 0–,0.001 Lemonniera terrestris Tubaki 0.002–0.088 0–0.002 ,0.001–0.028 Lunulospora curvula Ingold 0.004–0.145 0–0.003 0–,0.001 Stenocladiella neglecta Marvanová & Descals 0–1.073 0–0.098 0–0.010 Taeniospora gracilis var. enecta Marvanová 0–0.009 0.001–0.006 0–,0.001 Tetrachaetum elegans Ingold 0.009–0.785 0.001–0.157 0–0.056 0.001–0.110 Tetracladium marchalianum De Wild 0.009–0.158 0–0.035 0–,0.001 0–0.002 Tetracladium setigerum (Grove) Ingold 0–,0.001 Trichocladium angelicum Roldán 0–,0.001 0–,0.001 Tricladium angulatum Ingold 0–0.006 0–,0.001 Tricladium chaetocladium Ingold 0–0.022 0–0.003 Tricladium patulum Marvanová 0–,0.001 Tricladium splendens Ingold 0–,0.001 Triscelophorus acuminatus Nawawi 0–,0.001 Triscelophorus monosporus Ingold 0–,0.001 Tumularia aquatica (Ingold) Descals & Marvanová 0–,0.001 Tumularia tuberculata (Gönczöl) Descals & Marvanová 0–,0.001 Total sporulation rate 0.560–6.208 0.766–3.633 0.003–0.517 0.222–1.924 Taxa number 28 25 19 23 a Alatospora acuminata Ingold 1942 was described without a type. Marvanová and Descals (1985) later detected 2 strains in culture, which have clearly distinguishable conidia. They included both in A. acuminata and referred to them as ‘‘sensu stricto’’ (which the authors designated as neotype) and ‘‘sensu lato’’. However, the bracketed terms above may cause confusion because, by definition, the ‘‘sensu stricto’’ concept should be included in ‘‘sensu lato’’ and this is not the case here, where conidial shapes of both strains do not overlap. We conclude that A. acuminata sensu neotype should be kept as such, the category ‘‘sensu stricto’’ being redundant, and we propose to replace the term ‘‘sensu lato’’ by ‘‘pulchelloid’’, because its conidia strongly resemble those of A. pulchella. We recognize a 3rd conidial shape of what could belong to A. acuminata. It is readily recognized by its strikingly subulate, unconstricted stalk, and we refer to it as ‘‘subulate’’. This shape is relatively abundant in our samples and at many other sites in the Iberian Peninsula and elsewhere, including Hungary (J. Gönczöl, Hungarian National Museum, Budapest, personal communication) and possibly Australia. Pure culture and molecular studies underway will determine whether these 2 forms, pulchelloid and subulate, may be the basis for erecting formal taxa, and whether they should be included in A. acuminata. We included both forms as separate categories under A. acuminata to avoid losing potentially valuable ecological information. b Records of conidia of Goniopila monticola (Dyko) Marvanová & Descals and of typical conidia of Margaritispora aquatica Ingold are lumped because the conidia are indistinguishable on nitrocellulose filters and overlap in size. Atypical forms of the latter species have not been detected in our samples. 2011] LEAF DECOMPOSITION IN HEADWATER STREAMS 945 TABLE 7. Range of per site mean densities of invertebrate families collected from bags (no./g ash-free dry mass). Families are ordered by mean density within each functional group (FG). CC = Cornisa Cantábrica, CLC = Cordillera Litoral Catalana, SG = Sierra de Guadarrama, SN = Sierra Nevada, Shr = shredder, Col = collector, Gat = gatherer, Filt = filterer, Scr = scraper, Pred = predator. Family Order FG CC CLC SG SN Limnephilidae Trichoptera Shr 0–1.64 0–25.46 1.16–19.53 0.13–13.92 Leuctridae Plecoptera Shr 0–8.09 0–2.04 0–7.62 1.52–28.29 Gammaridae Crustacea Shr 0–49.89 0–2.15 Nemouridae Plecoptera Shr 0–2.38 0–17.38 0–4.78 0–13.16 Sericostomatidae Trichoptera Shr 0–0.84 0–27.06 0–10.09 0–0.10 Dryopidae Coleoptera Shr 0.52–15.93 Lepidostomatidae Trichoptera Shr 0–6.11 0–8.35 Capniidae Plecoptera Shr 0–4.81 0–1.88 0.12–3.07 Tipulidae Diptera Shr 0–0.46 0–1.54 0–1.21 0.12–0.86 Limoniidae Diptera Shr 0–0.86 0–0.48 0–0.41 0–1.22 Taeniopterygidae Plecoptera Shr 0–0.13 0–0.14 Odontoceridae Trichoptera Shr 0–0.14 Asellidae Crustacea Shr 0–0.11 Chironomidae Diptera Col–Gat 17.63–540.36 13.27–56.64 0–6.41 66.31–182.83 Oligochaeta Oligochaeta Col–Gat 1.41–80.48 0–1.65 0–0.63 0.46–27.77 Leptophlebiidae Ephemeroptera Col–Gat 0–4.32 0–4.05 0–3.68 Psychodidae Diptera Col–Gat 0–0.27 0–0.21 0–0.41 0.33–11.63 Ephemerellidae Ephemeroptera Col–Gat 0–3.53 0–0.16 Heptageniidae Ephemeroptera Col–Gat 0–0.63 0–1.61 Caenidae Ephemeroptera Col–Gat 0–1.58 0–0.51 Dixidae Diptera Col–Gat 0–0.16 0–0.12 Stratiomyidae Diptera Col–Gat 0–0.13 Hydropsychidae Trichoptera Col–Filt 0–1.65 0–2.55 0–0.82 2.09–24.85 Simuliidae Diptera Col–Filt 0–26.40 0–1.10 0–4.14 Brachycentridae Trichoptera Col–Filt 0–6.95 Philopotamidae Trichoptera Col–Filt 0–0.09 0–2.17 Hydrobiidae Mollusca Scr 0.13–56.86 Ancylidae Mollusca Scr 0–2.50 Scirtidae Coleoptera Scr 0–1.56 0–0.79 0–0.72 Goeridae Trichoptera Scr 0–0.48 Valvatidae Mollusca Scr 0–0.24 0–0.24 Glossosomatidae Trichoptera Scr 0–0.22 Baetidae Ephemeroptera Col–Gat–Scr 1.83–19.01 0–4.69 0–0.47 2.28–24.50 Elmidae Coleoptera Col–Gat–Scr 0–1.82 0–0.16 0–4.75 Hydraenidae Coleoptera Col–Gat–Scr 0–1.90 0–0.16 Planariidae Turbellaria Pred 0–2.32 0–4.29 0–17.25 Perlidae Plecoptera Pred 0–14.10 0–0.09 Empididae Diptera Pred 0–9.78 0–0.10 0.21–3.65 Polycentropodidae Trichoptera Pred 0–7.41 0–0.22 Athericidae Diptera Pred 0–1.13 0–0.70 0–0.09 Rhyacophilidae Trichoptera Pred 0–0.53 0–0.26 0–1.55 Ceratopogonidae Diptera Pred 0–0.47 0–0.58 Chloroperlidae Plecoptera Pred 0–0.39 0–0.40 0–0.65 Dytiscidae Coleoptera Pred 0–0.15 0–0.53 0–0.25 Hydrophilidae Coleoptera Pred 0–0.42 Cordulegasteridae Odonata Pred 0–0.46 0–0.11 Aeschnidae Odonata Pred 0–0.28 Calopterygidae Odonata Pred 0–0.17 Perlodidae Plecoptera Pred 0–0.16 Total shredders 0.75–50.68 0–46.38 7.93–29.10 3.86–73.36 Total invertebrates 57.92–683.30 21.67–121.56 10.85–38.30 114.47–271.29 Family number 41 19 27 31 946 J. POZO ET AL. [Volume 30 leaf species such as alder is less influenced than others by dissolved nutrients (Pozo et al. 1998, Hladyz et al. 2010). However, alder leaf litter decay is sensitive to a slight eutrophication when fine-mesh bags are used for incubations and to changes in riparian vegetation in nutrient-poor waters when coarse-mesh bags are used (Elosegi et al. 2006), results suggesting that this species is sensitive to the activity of both decomposers and detritivores under different stressors. The re- sponses of other species of poor quality (e.g., oak) to moderate eutrophication tend to be higher but later than responses of alder (Gulis et al. 2006), results consistent with the slower decomposition rates of oak leaves. Ferreira et al. (2006) showed that several indicators of the decomposition process respond faster in alder than in oak leaves (e.g., changes in nutrient content, fungal biomass, and sporulation peaks). Thus, alder leaf litter could be considered a better candidate than leaves with slower decay for assessing impacts on stream functioning because it responds faster and its use reduces the risk of bag loss caused by floods. In conclusion: 1) temperature was not the deter- mining factor for differences in breakdown rates among regions nor did rates increase with dissolved nutrients; 2) microbial activity (i.e., sporulation rates) was related to dissolved P, but the effect of nutrients on leaf breakdown rates was negligible; and 3) variability in shredder density explained the geo- graphical differences in breakdown rates, but their role was masked by other factors (e.g., channel slope) locally. Last, incorporation of breakdown rates into assessment schemes of stream ecological status might be hindered by the absence of unequivocal responses of leaf processing to variations of environmental factors among different geographical/climatic re- gions. Precise use of this functional metric would FIG. 3. Relationship between alder breakdown rate (b) and shredder density. Data from Sierra de Guadarrama (SG) were excluded from the regression. AFDM = ash-free dry mass, dd = degree day. TABLE 8. Summary of the regression analyses between the potential explanatory variables and the leaf-litter breakdown rate not adjusted and adjusted by region (analysis of covariance) (right). Sign of the slope is indicated. Asterisks (*) highlight models in which the region factor significantly affects the variable. Variables Unadjusted model Adjusted model Slope R2 p (F–test) Slope R2 p (F–test) Conductivity 2 0.786 ,0.001 2 0.837 0.004 Alkalinity 2 0.645 ,0.001 2 0.806 0.016* Channel slope + 0.644 ,0.001 + 0.789 0.040* pH 2 0.481 ,0.001 + 0.743 0.277* Sporulation rate 2 0.438 ,0.001 2 0.732 0.610* Hyphomycete richness 2 0.286 0.010 2 0.760 0.150* Ammonium 2 0.260 0.015 2 0.729 0.783* Total invertebrate density 2 0.256 0.016 + 0.754 0.194* Shredder density + 0.159 0.066 + 0.797 0.028* Shredder richness + 0.148 0.077 + 0.753 0.201 SRP 2 0.127 0.104 + 0.734 0.543 Nitrate 2 0.112 0.127 2 0.747 0.265 Nitrite 2 0.104 0.144 2 0.729 0.780 Riparian cover 2 0.066 0.247 + 0.729 0.757 Channel width 2 0.044 0.319 2 0.733 0.551 Total invertebrate richness 2 0.042 0.359 + 0.736 0.461 2011] LEAF DECOMPOSITION IN HEADWATER STREAMS 947 require regional adjustments of baseline standards in reference conditions. Acknowledgements This study was funded by the Spanish Ministry of Education and Science (project CGL2007-66664-C04), by the University of the Basque Country (Research grant GIU05/38), and by the Basque Government (Research grants IT-422-07 and IT-302-10). We are grateful to Nadia Arkarazo, Aingeru Martı́nez, Fernando Rodrı́- guez, Carolina Rozas, and Roberto Velilla for help with field and laboratory work. We thank the ‘Cuenca Alta del Manzanares’ Regional Park, the Peñalara Natural Park, the Gorbeia Natural Park, the Montseny Natural Park, and the Sierra Nevada Natural-National Park for sampling permits and assistance, and also acknowledge the Spanish Meteorological Agency (AEMET) for providing data on temperature and rainfall. Literature Cited ABELHO, M. 2001. From litterfall to breakdown in streams: a review. TheScientificWorld 1:656–680. ALLEN, S. E., H. M. GRIMSHAW, J. A. PARKINSON, AND J. A. QUARMBY. 1974. Chemical analysis of ecological materi- als. Blackwell Scientific, Oxford, UK. APHA (AMERICAN PUBLIC HEALTH ASSOCIATION). 2005. Stan- dard methods for the examination of water and wastewater. 22nd edition. American Public Health Association, American Waterworks Association, Water Environment Federation, Washington, DC. ARTIGAS, J., A. M. ROMANÍ, AND S. SABATER. 2008. Effect of nutrients on the sporulation and diversity of aquatic hyphomycetes on submerged substrata in a Mediterra- nean stream. Aquatic Botany 88:32–38. BARBOUR, M. T., J. GERRITSEN, B. D. SNYDER, AND J. B. STRIBLING. 1999. Rapid bioassessment protocols for use in streams and wadeable rivers: periphyton, benthic macroinvertebrates and fish. 2nd edition. EPA 841-B-99-002. Office of Water, US Environmental Protection Agency, Washington, DC. BERGFUR, J. 2007. Seasonal variation in leaf-litter breakdown in nine boreal streams: implications for assessing functional integrity. Fundamental and Applied Limnol- ogy 169:319–329. BROOKS, S. S., M. A. PALMER, B. J. CARDINALE, C. M. SWAN, AND S. RIBBLETT. 2002. Assessing stream ecosystem rehabili- tation: limitations of community structure data. Resto- ration Ecology 10:156–168. BUNN, S. E., AND P. M. DAVIES. 2000. Biological processes in running waters and their implications for the assess- ment of ecological integrity. Hydrobiologia 422:61–70. CASAS, J. J., M. O. GESSNER, P. LANGTON, D. CALLE, E. DESCALS, AND M. J. SALINAS. 2006. Diversity of patterns and processes in rivers of eastern Andalusia. Limnetica 25: 155–170. CHAUVET, E., E. FABRE, A. ELÓSEGUI, AND J. POZO. 1997. The impact of eucalyptus on the leaf-associated aquatic hyphomycetes in Spanish streams. Canadian Journal of Botany 75:880–887. DANKS, H. V. 2007. How aquatic insects live in cold climates. Canadian Entomologist 139:443–471. DE JONGE, M., B. VAN DE VIJVER, R. BLUST, AND L. BERVOETS. 2008. Responses of aquatic organisms to metal pollution in a lowland river in Flanders: a comparison of diatoms and macroinvertebrates. Science of the Total Environ- ment 407:615–629. DEMARS, B. O. L., AND A. C. EDWARDS. 2009. Distribution of aquatic macrophytes in contrasting river systems: a critique of compositional-based assessment of water quality. Science of the Total Environment 407:975–990. DODDS, W. K. 2007. Trophic state, eutrophication and nutrient criteria in streams. Trends in Ecology and Evolution 22:669–676. ELOSEGI, A., A. BASAGUREN, AND J. POZO. 2006. A functional approach to the ecology of Atlantic Basque streams. Limnetica 25:123–134. FERREIRA, V., V. GULIS, AND M. A. S. GRAÇA. 2006. Whole- stream nitrate addition affects litter decomposition and associated fungi but not invertebrates. Oecologia (Ber- lin) 149:718–729. FLEITUCH, T., AND M. LEICHTFRIED. 2007. Electron transport system (ETS) activity in alder leaf litter in two contrasting headwater streams. International Review of Hydrobiology 92:378–391. FRIBERG, N., J. B. DYBKJAER, J. S. OLAFSSON, G. M. GISLASON, S. E. LARSEN, AND T. L. LAURIDSEN. 2009. Relationships between structure and function in streams contrasting in temperature. Freshwater Biology 54:2051–2068. FU, G., D. BUTLER, AND S. T. KHU. 2009. The impact of new developments on river water quality from an integrated system modelling perspective. Science of the Total Environment 407:1257–1267. GESSNER, M. O., AND E. CHAUVET. 2002. A case for using litter breakdown to assess functional stream integrity. Ecological Applications 12:498–510. GONÇALVES, J. F., M. GRAÇA, AND M. CALLISTO. 2006. Leaf-litter breakdown in 3 streams in temperate, Mediterranean, and tropical Cerrado climates. Journal of the North American Benthological Society 25:344–355. GRATTAN, R. M., AND K. SUBERKROPP. 2001. Effects of nutrient enrichment on yellow poplar leaf decomposition and fungal activity in streams. Journal of the North American Benthological Society 20:33–43. GREENWOOD, J. L., A. D. ROSEMOND, J. B. WALLACE, W. F. CROSS, AND H. S. WEYERS. 2007. Nutrients stimulate leaf breakdown rates and detritivore biomass: bottom-up effects via heterotrophic pathways. Oecologia (Berlin) 151:637–649. GULIS, V., V. FERREIRA, AND M. A. S. GRAÇA. 2006. Stimulation of leaf litter decomposition and associated fungi and invertebrates by moderate eutrophication: implications for stream assessment. Freshwater Biology 51:1655–1669. GULIS, V., L. MARVANOVÁ, AND E. DESCALS. 2005. An illustrated key to the common temperate species of aquatic hyphomycetes. Pages 153–168 in M. A. S. Graça, F. Bärlocher, and M. O. Gessner (editors). Methods to study 948 J. POZO ET AL. [Volume 30 litter decomposition: a practical guide. Springer, Dor- drecht, The Netherlands. GULIS, V., AND K. SUBERKROPP. 2003. Leaf litter decomposition and microbial activity in nutrient-enriched and unal- tered reaches of a headwater stream. Freshwater Biology 48:123–134. HIEBER, M., AND M. O. GESSNER. 2002. Contribution of stream detritivores, fungi, and bacteria to leaf breakdown based on biomass estimates. Ecology 83:1026–1038. HLADYZ, S., M. O. GESSNER, P. S. GILLER, J. POZO, AND G. WOODWARD. 2009. Resource quality and stoichiometric constraints on stream ecosystem functioning. Freshwa- ter Biology 54:957–970. HLADYZ, S., S. D. TIEGS, M. O. GESSNER, P. S. GILLER, G. RÎŞNOVEANU, M. PREDA, M. NISTORESCU, M. SCHINDLER, AND G. WOODWARD. 2010. Leaf-litter breakdown in pasture and deciduous woodland streams: a comparison among three European regions. Freshwater Biology 55: 1916–1929. IRONS, J. G., M. W. OSWOOD, R. J. STOUT, AND C. M. PRINGLE. 1994. Latitudinal patterns in leaf litter breakdown: is temperature really important? Freshwater Biology 32: 401–411. KARR, J. R. 1991. Biological integrity: a long-neglected aspect of water resource management. Ecological Applications 1:66–84. KARR, J. R., AND E. W. CHU. 1999. Restoring life in running waters. Better biological monitoring. Island Press, Washington, DC. KROM, M. D. 1980. Spectrophotometric determination of ammonia: a study of a modified Berthelot reduction using salicylate and dichloroisocyanurate. Analyst 105: 305–316. KVÅLSETH, T. O. 1985. Cautionary note about R2. American Statistician 39:279–285. LECERF, A., AND E. CHAUVET. 2008a. Diversity and functions of leaf-decaying fungi in human-altered streams. Fresh- water Biology 53:1658–1672. LECERF, A., AND E. CHAUVET. 2008b. Intraspecific variability in leaf traits strongly affects alder leaf decomposition in a stream. Basic and Applied Ecology 9:598–607. LECERF, A., G. RISNOVEANU, C. POPESCU, M. O. GESSNER, AND E. CHAUVET. 2007. Decomposition of diverse litter mixtures in streams. Ecology 88:219–227. MARVANOVÁ, L., AND E. DESCALS. 1985. New and critical taxa of aquatic hyphomycetes. Botanical Journal of the Linnean Society 91:1–23. MCKIE, B. G., G. WOODWARD, S. HLADYZ, M. NISTORESCU, E. PREDA, C. POPESCU, P. S. GILLER, AND B. MALMQVIST. 2008. Ecosystem functioning in stream assemblages from different regions: contrasting responses to variation in detritivore richness, evenness and density. Journal of Animal Ecology 77:495–504. MENDOZA-LERA, C., J. PÉREZ, E. DESCALS, A. MARTÍNEZ, O. MOYA, I. AROSTEGUI, AND J. POZO. 2010. Headwater reservoirs weaken terrestrial-aquatic linkage by slowing leaf-litter processing in downstream regulated reaches. River Research and Applications. doi: 10.1002/rra.1434 MERRITT, R. W., AND K. W. CUMMINS (EDITORS). 1996. An intro- duction to the aquatic insects of North America. 3rd edition. Kendall/Hunt Publishing Company, Dubuque, Iowa. MESQUITA, A., C. PASCOAL, AND F. CÁSSIO. 2007. Assessing effects of eutrophication in streams based on break- down of eucalypt leaves. Fundamental and Applied Limnology 168:221–230. MOLINERO, J., J. POZO, AND E. GONZÁLEZ. 1996. Litter breakdown in streams of the Agüera catchment: influence of dissolved nutrients and land use. Freshwa- ter Biology 36:745–756. MONTEIRO, M. I. C., F. N. FERREIRA, N. M. M. DE OLIVEIRA, AND A. K. ÁVILA. 2003. Simplified version of the sodium salicylate method for analysis of nitrate in drinking waters. Analytica Chimica Acta 477:125–129. MOSS, B. 2008. The Water Framework Directive: total environment or political compromise? Science of the Total Environment 400:32–41. MOULTON, T. P. 1999. Biodiversity and ecosystem functioning in conservation of rivers and streams. Aquatic Conser- vation: Marine and Freshwater Ecosystems 9:573–578. MÜLLER, B., M. BERG, Z. P. YAO, X. F. ZHANG, D. WANG, AND A. PFLUGER. 2008. How polluted is the Yangtze river? Water quality downstream from the Three Gorges Dam. Science of the Total Environment 402:232–247. NIYOGI, D. K., K. S. SIMON, AND C. R. TOWNSEND. 2003. Breakdown of tussock grass in streams along a gradient of agricultural development in New Zealand. Freshwa- ter Biology 48:1698–1708. PASCOAL, C., AND F. CÁSSIO. 2004. Contribution of fungi and bacteria to leaf litter decomposition in a polluted river. Applied and Environmental Microbiology 70:5266–5273. PASCOAL, C., M. PINHO, F. CÁSSIO, AND P. GOMES. 2003. Assessing structural and functional ecosystem condition using leaf breakdown: studies on a polluted river. Freshwater Biology 48:2033–2044. PAUL, M. J., J. L. MEYER, AND C. A. COUCH. 2006. Leaf breakdown in streams differing in catchment land use. Freshwater Biology 51:1684–1695. PÉREZ, J., M. MENÉNDEZ, S. LARRAÑAGA, AND J. POZO. 2011. Inter- and intra-regional variability of leaf litter break- down in reference headwater streams of northern Spain: Atlantic versus Mediterranean streams. International Review of Hydrobiology 96:105–117. POZO, J. 1993. Leaf litter processing of alder and eucalyptus in the Agüera stream system (North Spain) I. Chemical changes. Archiv für Hydrobiologie 127:299–317. POZO, J., A. BASAGUREN, A. ELÓSEGUI, J. MOLINERO, E. FABRE, AND E. CHAUVET. 1998. Afforestation with Eucalyptus globulus and leaf litter decomposition in streams of northern Spain. Hydrobiologia 373/374:101–109. QUINN, G. P., AND M. J. KEOUGH. 2002. Experimental design and data analysis for biologists. Cambridge University Press, Cambridge, UK. RABENI, C. F., K. E. DOISY, AND L. D. ZWEIG. 2005. Stream invertebrate community functional responses to depos- ited sediment. Aquatic Sciences 67:395–402. RIIPINEN, M. P., J. DAVY-BOWKER, AND M. DOBSON. 2009. Comparison of structural and functional stream assess- 2011] LEAF DECOMPOSITION IN HEADWATER STREAMS 949 ment methods to detect changes in riparian vegetation and water pH. Freshwater Biology 54:2127–2138. SAMPAIO, A., P. RODRÍGUEZ-GONZÁLEZ, S. VARANDAS, R. M. CORTES, AND M. T. FERREIRA. 2008. Leaf litter decompo- sition in western Iberian forested wetlands: lentic versus lotic response. Limnetica 27:93–106. SPÄNHOFF, B., C. AUGSPURGER, AND K. KÜSEL. 2007. Comparing field and laboratory breakdown rates of coarse partic- ulate organic matter: sediment dynamics mask the impacts of dissolved nutrients on CPOM mass loss in streams. Aquatic Sciences 69:495–502. STOUT, R. J. 1989. Effects of condensed tannins on leaf processing in mid-latitude and tropical streams: a theoretical approach. Canadian Journal of Fisheries and Aquatic Sciences 46:1097–1106. SUBERKROPP, K., AND E. CHAUVET. 1995. Regulation of leaf breakdown by fungi in streams: influences of water chemistry. Ecology 76:1433–1445. TACHET, H., P. RICHOUX, M. BOURNAUD, AND F. USSEGLIO- POLATERA. 2002. Invertébrés d’eau douce. Systematique, biologie, écologie. CNRS Editions, Paris, France. VERBEKE, G., AND G. MOLENBERGHS. 2000. Linear mixed models for longitudinal data. Springer-Verlag, New York. WALLACE, J. B., AND S. L. EGGERT. 2009. Benthic invertebrate fauna, small streams. Encyclopedia of Inland Waters 2:173–190. WEBSTER, J. R., AND E. F. BENFIELD. 1986. Vascular plant breakdown in freshwater ecosystems. Annual Review of Ecology and Systematics 17:567–594. WEBSTER, J. R., J. B. WALLACE, AND E. F. BENFIELD. 1995. Organic processes in streams of the Eastern United States. Pages 117–187 in C. E. Cushing, K. W. Cummins, and G. W. Minshall (editors). Ecosystems of the world 22. River and stream ecosystems. Elsevier, New York. WITHERS, P. J. A., AND H. P. JARVIE. 2008. Delivery and cycling of phosphorus in rivers: a review. Science of the Total Environment 400:379–395. WOODWARD, G. 2009. Biodiversity, ecosystem functioning and food webs in freshwaters: assembling the jigsaw puzzle. Freshwater Biology 54:2171–2187. YOUNG, R. G., C. D. MATTHAEI, AND C. R. TOWNSEND. 2008. Organic matter breakdown and ecosystem metabolism: functional indicators for assessing river ecosystem health. Journal of the North American Benthological Society 27:605–625. YOUNG, R. G., C. R. TOWNSEND, AND C. D. MATTHAEI. 2004. Functional indicators of river ecosystem health – an interim guide for use in New Zealand. Cawthron report No. 870 Cawthron Institute, Nelson, New Zealand. (Available from: http://www.cawthron.org.nz/coastal- freshwater-resources/downloads/functional-indicators- guide-final.pdf) ZAR, J. H. 2010. Biostatistical analysis. 5th edition. Pearson, Upper Saddle River, New Jersey. ZWEIG, L. D., AND C. F. RABENI. 2001. Biomonitoring for deposited sediment using benthic invertebrates: a test on 4 Missouri streams. Journal of the North American Benthological Society 20:643–657. Received: 17 November 2010 Accepted: 28 June 2011 950 J. POZO ET AL. [Volume 30