Antonio E. B. Silva1,2
Hermano M. Queiroz3
Daniel P. Oliveira4
Giovanna Bergamim Araujo Lopes3Tamara M. Pereira2Gabriel N. Nóbrega5
Diego Barcellos6Carla F. Rezende2
Tiago O. Ferreira1,7*- 1Department of Soil Science, “Luiz de Queiroz” College of Agriculture, University of São Paulo, Piracicaba, Brazil
- 2Graduate Course of Ecology and Natural Resources, Department of Biology, Federal University of Ceará, Fortaleza, Brazil
- 3Department of Geography, Cidade Universitária, University of São Paulo, São Paulo, Brazil
- 4Research Institute for Meteorology and Water Resources (FUNCEME), Fortaleza, Brazil
- 5Department of Soil Science, Federal University of Ceará, Fortaleza, Brazil
- 6Department of Environmental Sciences, Federal University of São Paulo, Diadema, Brazil
- 7Center for Carbon Research in Tropical Agriculture (CCARBON), University of São Paulo, Piracicaba, Brazil
Intermittent rivers and ephemeral streams (IRES) are central to nutrient transport in semiarid regions, yet their role in phosphorus (P) dynamics under combined natural and anthropogenic pressures remains poorly understood. This study assessed sediments and surface waters along the Cruxati River Basin (Northeast Brazil), comparing upland, piedmont, and lowland zones. Field sampling included the following sediment parameters: pH, Eh, particle size distribution, total organic carbon (TOC), sequential P fractionation, and Fe forms. We also performed the following water analyses: pH, Eh, electrical conductivity (EC), dissolved oxygen (DO), total-P, and orthophosphate-P. Spatial characterization of land use/cover and trophic status of nearby reservoirs supported our environmental interpretations. Sediments were predominantly sandy (93.0 ± 7.2%), with kaolinite as the main mineral phase. Among the P fractions, oxide-P, silicate-P, and residual-P were dominant across the sites, while Ca-P was found only in the upland sites. Statistically significant differences were observed for silicate-P (0.0 ± 0.0 mg kg−1 upland; 15.2 ± 4.0 mg kg−1 piedmont; 29.4 ± 24.8 mg kg−1 lowland), Ca-P (14.4 ± 5.4 mg kg−1 upland; absent downstream), and residual-P (26.4 ± 16.8 mg kg−1 lowland vs. 8.1 ± 7.6 mg kg−1 piedmont). In surface waters, total-P decreased downstream (0.16 ± 0.03 mg L−1 upland; 0.07 ± 0.02 mg L−1 piedmont; 0.05 ± 0.0 mg L−1 lowland), whereas orthophosphate-P showed similar patterns across the sites. Land use mapping indicated >70% forest cover, 17% pasture, 8% agricultural mosaic, and 1% urban areas, with increasing fragmentation downstream. Reservoirs near the sampling zones were predominantly eutrophic, highlighting additional anthropogenic pressures. Our results revealed that P dynamics in IRES are shaped by the interaction between natural retention/release mechanisms and human-driven inputs, with depositional zones acting as critical hotspots. Our findings provide new insights into P cycling in semiarid IRES and their vulnerability under climate and land-use changes.
1 Introduction
Intermittent rivers and ephemeral streams (IRES) alternate between periods of flowing water and drought within their channels (Fovet et al., 2021; Sánchez-Montoya et al., 2023). These streams often occur in drylands, where low amounts of precipitation and seasonality drive distinctive biogeochemical cycles. Such cycles cause pronounced temporal and spatial fluctuations in nutrient concentrations, influencing downstream transport (Mulholland and Webster, 2010; Shumilova et al., 2019; von Schiller et al., 2017a). Moreover, IRES act as important nutrient sources for adjacent ecosystems (e.g., phosphorus: Da Silva et al., 2024).
In contrast to perennial aquatic systems, where phosphorus (P) is often found in bioavailable dissolved forms (Reynolds and Davies, 2001), P in intermittent and ephemeral systems tends to associate with mineral phases such as iron (Fe) oxides and silicates, as well as with refractory organic matter from riverine sediments (Da Silva et al., 2024). Multiple factors influence P dynamics in natural ecosystems, including hydrological transport, microbial activity, organic matter, Fe content, and physicochemical conditions (pH and redox potential – Eh) (Darke and Walbridge, 2000; Attygalla et al., 2016; O'Connell et al., 2018; Cavalcante et al., 2018; Kong et al., 2021; Kim et al., 2016; Barcellos et al., 2019; Holtan et al., 1988; Nóbrega et al., 2014; Peng et al., 2022; Pettersson et al., 1988; Queiroz et al., 2022). In contrast to investigations performed in agroecosystems, researchers have largely overlooked the biogeochemical controls of P in IRES (Acuña et al., 2017), and most of the existing studies in IRES are focusing on population and community ecology (e.g., Gonçalves-Silva et al., 2022).
Although recent investigations have highlighted the role of IRES in the seasonal transport of nutrients to downstream ecosystems-such as estuaries-in semiarid regions (Da Silva et al., 2024), our understanding of the dynamics in the P cycling in these systems remains limited. Local factors, including climate regime, sediment composition, land-use intensity, and the presence of dams along the streams, all strongly influence IRES functioning and may impact ecosystemic nutrient cycling (Chiu et al., 2017; Jaeger et al., 2017; Nabih et al., 2021; Li et al., 2023; Xing et al., 2024). Consequently, the influence of these different parameters demonstrates that biogeochemical cycling P and other key elements are highly context-dependent, reinforcing the need for more site-specific and process-oriented investigations.
In the Brazilian Northeastern semiarid region, where drought conditions prevail, water plays a vital role in human consumption, livestock, and agriculture, often supported by thousands of artificial reservoirs (Teixeira, 2004; Meireles et al., 2007; Mamede et al., 2018). The widespread construction of dams has altered natural flow regimes, increased intermittency, disrupted sediment connectivity, and enhanced upstream P retention; while it also reduces downstream delivery, and ultimately reshaping nutrient dynamics along the river continuum (Maavara et al., 2015; Zhou et al., 2024). These local complexities underscore the importance of considering both regional climatic constraints and human interventions to understand P cycling in IRES. Furthermore, in arid and semiarid regions-where P often limits productivity-rivers are becoming increasingly intermittent due to climate change (Larned et al., 2010). Both climate change and anthropogenic alterations (e.g., damming) affect terrestrial and aquatic ecosystems, either locally or adjacently, highlighting the need to understand spatial and temporal patterns of P availability as key determinants of ecosystem functioning in drylands.
Thus, understanding the spatial dynamics of P in IRES is therefore essential for predicting ecosystem responses, assessing environmental risks in impacted systems, and supporting conservation and restoration efforts (Datry et al., 2014; von Schiller et al., 2017b). Within this context, the present study aimed to investigate spatial patterns of P distribution and forms along an intermittent river to clarify how hydrological and geomorphological gradients regulate nutrient cycling in dryland ecosystems. We hypothesized that (i) P concentrations and the prevalence of specific P fractions vary systematically along the hydrological gradient from upland to lowland zones; (ii) depositional environments in lowland zones accumulate greater proportions of refractory and mineral-bound P fractions; and (iii) topographic position, combined with sediment composition, primarily controls P retention and transformation.
2 Material and methods
2.1 Study site and sampling
The work was conducted on the Cruxati River, located in the state of Ceará, Northeast of Brazil (Figure 1). The region's climate is semi-arid, and annual evapotranspiration exceeds annual precipitation, with averages for precipitation and evapotranspiration of 1,100 and 1,400 mm, respectively, and annual temperature ranging between 26 and 28 °C (Funceme - Fundação Cearense de Meteorologia e Recursos Hídricos., 2018; Duarte et al., 2021). Higher precipitation rates are concentrated in February to April, with a drier period in the remaining months of the year (Porto et al., 2004; Funceme - Fundação Cearense de Meteorologia e Recursos Hídricos., 2018; Duarte et al., 2021). Therefore, most of the watercourses in the region are IRES.
Figure 1. Location and hypsometric characterization of the Cruxati River Basin with sampling points representing upland, piedmont, and lowland zones. (A, B) Hypsometric maps showing basin delineation, drainage network, and sampling sites. (C) Three-dimensional hypsometric view highlighting local relief and spatial distribution of zones. (D) Longitudinal transect across the basin used to generate the elevation profile. (E) Altitudinal profile of the sampling points along the transect.
The Cruxati River is part of the coastal watershed of the state and has an area of approximately 77 km2 (COGERH - Companhia de Gestão dos Recursos Hídricos, 2012), where the predominant geology consists of metamorphic and sedimentary rocks from the Barreiras formation (Nunes et al., 2011). The basin is influenced by various human activities in its surroundings, including discharge of domestic and industrial effluents into the rivers, as well as agricultural and pasture activities that lead to a reduction in dense and riparian vegetation (Duarte et al., 2021). These activities may induce a physical and chemical degradation of soils and sediments, resulting in alterations in the hydrographic system. The basin is predominantly covered by the native Caatinga vegetation, a unique Brazilian biome characterized by a semiarid climate and drought-deciduous vegetation adapted to water scarcity (Funceme - Fundação Cearense de Meteorologia e Recursos Hídricos., 2018). While patches of other vegetation types, such as sub-deciduous tropical rainforest, may occur within the basin, Caatinga predominates, particularly in its dense and open shrub forms. This biome is characterized by plant species with reduced or deciduous leaves, thickened stems, deep or extensive root systems, and water retention mechanisms that enable them to withstand prolonged drought periods (Giulietti et al., 2004).
Hydrological flow data were not directly measured during this study. Sampling was conducted in July 2019 during the period of continuous water flow in the river. Consequently, results represent conditions during flow periods, and potential differences under dry-season conditions are not captured. The selection of this hydroperiod was based on studies involving nutrient dynamics and IRES (von Schiller et al., 2017a,b, 2011). For sediment and water collection, 200 m transects were distributed across three zones as follows: sediment production zone (Upland Zone), sediment transition zone (Piedmont Zone), and sediment arrival zone (Lowland Zone; Figure 1) (Tooth, 2000). Changes in riverine geomorphology in arid lands have been documented and demonstrated that these changes influence sediment inputs and outputs along the channel (Tooth, 2000). Also, studies in IRES systems are strategic for understanding nutrient biogeochemical cycling that are interconnected to the hydrological and sedimentological dynamics (e.g., downstream; von Schiller et al., 2017a,b; Shumilova et al., 2019).
For each sampling zone, we collected sediment samples randomly (e.g., right bank, left bank, and center of the river) within three depths (i.e., 0–3 cm, 3–6 cm, and 6–9 cm). The sampling was performed using polycarbonate tubes (0.05 m internal diameter and 0.5 m length) attached to a sediment sampler specially designed for waterlogged soils and sediments. In total, 27 sediment samples were collected (9 samples per zone; 3 per depth). The pH and redox potential (Eh) were measured in situ for each sample by using a glass electrode calibrated with pH standards of 4.0 and 7.0, and a platinum electrode for Eh probe, respectively.
Surface water samples within 10 cm depth were collected using amber glass bottles previously washed with HCl 10% (v/v). Water samples were filtered (0.45 μm) and refrigerated at 4 °C, and the analyses were performed within 48 h of collection for the determination of total P and orthophosphate (persulfate digestion followed by the ascorbic acid method; Eaton and Franson, 2005; Duarte et al., 2021). The following water parameters were obtained by using a multiparameter probe (YSI 7000): pH, Eh, dissolved oxygen (DO; mg L−1), and electrical conductivity (EC; dS m−1).
2.2 Organic carbon contents, particle size, mineralogical analysis, and iron extraction
We determined the organic carbon content (TOC) from sediments by dry combustion in an elemental analyzer (LECO 144 SE-DR). The particle size distribution (i.e., sand, silt, and clay) was obtained by the pipette method proposed by Gee and Bauder (1986). The mineralogical characterization was performed by using one representative sample from each transect (due to the geological and soil homogeneity along the transects). Before the mineralogical analyses, we conducted a pre-treatment with a sodium hypochlorite solution to oxidize the organic matter from the samples, and afterwards we crushed and sieved (< 150 mesh) the samples for X-ray diffraction (XRD) (Siregar et al., 2005). For the clay fraction, we mounted unoriented slides (powder) and conducted the XRD analyses in a Miniflex II Desktop X-Ray Diffractometer with Cu-Kα radiation, with steps of 0.02° 2θ s−1 and range of 3-60° 2θ (Chen, 1977). The Fe oxides were analyzed following the combined methods proposed by Mehra and Jackson (1960), and Mckeague (1978). These methods allowed the extraction of oxides based on their degree of crystallinity (i.e., crystalline phases extracted using dithionite-citrate-bicarbonate—DCB and low crystalline phases extracted using ammonium oxalate). Due to the high affinity of P for Fe oxides (Dixon and Schulze, 2002), the determination of the contents of both long-range and short-range order Fe phases is important to improve the understanding of P sorption mechanisms in the mineralogical matrix.
2.3 Sequential extraction of P
P contents were determined after an operational geochemical fractionation following the methods proposed by Paludan and Jensen (1995), Paludan and Morris (1999), and Nóbrega et al. (2014) resulting in the separation of six P fractions (Table 1). P concentrations were determined by the colorimetric method ascorbic acid/molybdenum blue, and absorbance obtained by using a spectrophotometer at a wavelength of 885 nm (Murphy and Riley, 1962).
Table 1. Different phases of P obtained by an operational geochemical fractionation (six fractions).
2.4 Spatial characterization of land use and water quality
Two complementary analyses were conducted: (i) mapping of land use and cover, and (ii) evaluation of reservoir water quality through trophic state classification. For both analyses, open-source tools were applied, including QGIS 3.34.6 (Prizren, Switzerland), Google Earth Pro, and RStudio (RStudio Team, 2020). Vector datasets were obtained from the official portals of the following Brazilian agencies: Companhia de Gestão dos Recursos Hídricos (COGERH) and the Brazilian Institute of Geography and Statistics (IBGE).
For the land use and land cover (LULC) map, we used a database from MapBiomas Collection 6.0 (MapBiomas, 2023), which provides a consistent historical series since 1985. The 2019 mosaic was derived from Landsat imagery (USGS/NASA) processed within the Google Earth Engine platform by the MapBiomas project. Land cover classes were identified according to their spectral signatures, with a spatial resolution of 10 m, enabling detailed representation of vegetation types and anthropogenic land uses across the basin.
For water quality assessment, trophic state data of monitored reservoirs were obtained directly from COGERH following a formal data request. The trophic state index was calculated following the methodology established by the National Water Agency (ANA; Table 2) and adapted by COGERH for semiarid reservoirs. The index integrates total phosphorus concentrations, which represent the main causal factor of eutrophication in freshwater ecosystems, and chlorophyll-a concentrations, which indicate the biological response through phytoplankton biomass growth (Paulino et al., 2013). All datasets and satellite images used are publicly available and comply with copyright and licensing restrictions established by their respective institutions (MapBiomas, IBGE, COGERH, and USGS/NASA for Landsat imagery).
Table 2. Thresholds and classification criteria for different trophic state levels according to the trophic state index (TSI).
2.5 Statistical analyses
Statistical analyses were performed using Past 4.03 (Hammer et al., 2001) and R software (R Core Team, 2019). Because residuals did not meet the assumption of normality, differences among sampling zones (upland, piedmont, and lowland zones) were tested using the non-parametric Kruskal–Wallis test, followed by Dunn's post-hoc test with Bonferroni correction (p < 0.05) to identify pairwise differences. Additionally, discriminant analysis was applied to explore associations among variables (EX-P, Oxide-P, Silicate-P, HA-P, Ca-P, Res-P, Total-P, Orthophosphate-P, Crystalline-Fe, Low-crystalline-Fe, TOC, DO, EC, pH, and Eh) and the three hydrological zones (upland, piedmont, and lowland) (Fávero et al., 2009; Manly and Alberto, 2016; Gotelli and Ellison, 2013).
3 Results
3.1 Phosphorus in water
The P contents in water (total-P and orthophosphate-P; Figure 2) presented statistical differences, with a trend of decreasing contents toward downstream. For total-P (p-value = 0.004), the mean concentrations were the following: 0.16 ± 0.03 mg L−1 (upland zone), 0.07 ± 0.02 mg L−1 (piedmont zone), and 0.05 ± 0.0 mg L−1 (lowland zone; Figure 2A). Meanwhile, for orthophosphate-P (p-value < 0.001), the concentrations were the following: 0.03 ± 0.0 mg L−1 (upland zone), 0.01 ± 0.0 mg L−1 (piedmont zone), and 0.01 ± 0.0 mg L−1 (lowland zone; Figure 2B).
Figure 2. Means ± standard deviation for P concentrations in water from the Cruxati River across three studied zones (upland, piedmont, and lowland) for (A) Total-P and (B) orthophosphate-P. Different lowercase letters above the bars indicate statistically significant differences (p < 0.05), based on the Kruskal-Wallis test.
3.2 Physicochemical parameters for sediments and water (pH, Eh, EC, TOC and DO)
We observed statistical differences in the pH values for sediments among the studied zones (p = 0.007; Table 3), with the highest values in the lowland zone, while upland and piedmont zones showed similar values. In contrast, redox potential (Eh), and electrical conductivity (EC) in sediments did not vary significantly across the studied zones (Table 3). In surface waters, pH and Eh were also statistically similar among zones, whereas EC differed significantly (p < 0.001), presenting the highest values in the lowland zone, followed by piedmont and upland zones.
Table 3. Means ± standard deviation values for pH, redox potential (Eh), and electrical conductivity (EC) in sediments and surface waters across the three hydrological zones (upland, piedmont, and lowland).
Additionally, sediment TOC and DO contents in surface waters did not differ significantly among the studied zones (Table 4). Despite the absence of statistical differences, TOC values were relatively uniform across zones (~4.5–5.5 g kg−1; Table 4), suggesting that organic matter availability is consistent along the hydrological gradient. Similarly, DO concentrations remained high (7.5–8.2 mg L−1; Table 4), indicating well-oxygenated surface waters across all zones. Together, these results suggest limited spatial variation in organic matter inputs and oxygen conditions, implying that other factors (e.g., hydrological connectivity and sediment composition) may play a stronger role in shaping P dynamics in these IRES systems.
Table 4. Means ± standard deviation contents for total organic carbon (TOC) in sediments and dissolved oxygen (DO) in surface waters across the three hydrological zones (upland, piedmont, and lowland).
3.3 Particle distribution and mineralogical characterization
Particle size distribution showed that sediments were predominantly sandy in all zones, with no significant differences among the three zones. Mean values for sand, silt, and clay fractions were 93.0 ± 7.2%, 4.4 ± 4.7%, and 2.6 ± 2.9%, respectively. XRD analysis of the clay fraction indicated that kaolinite was the dominant mineral, with characteristic diffraction peaks at 7.14, 4.36, 3.58, 2.49, and 2.34 nm. Similar patterns were observed across upland, piedmont, and lowland zones (Figure 3).
Figure 3. X-ray diffraction (XRD) patterns of the clay fraction from sediment samples collected in the upland, piedmont, and lowland zones of the Cruxati River. Diffractograms were obtained from unoriented powder mounts using Cu-Kα radiation. Kaolinite (K) was consistently identified as the predominant crystalline phase across zones.
3.4 Low-crystallinity and crystalline Fe forms
The chemical extraction of Fe oxyhydroxides indicates a predominance of crystalline Fe phases compared to low-crystallinity Fe in the collected sediments across the different zones (Figure 4). Crystalline Fe oxide forms were, on average, 77-fold more prevalent than low-crystallinity forms across all studied zones. However, there were no statistically significant differences in Fe fractions among the studied zones. For crystalline Fe oxyhydroxides, the recorded values were 8.80 ± 8.59 g kg−1 (upland zone), 5.18 ± 1.50 g kg−1 (piedmont zone), and 5.13 ± 1.34 g kg−1 (lowland zone; Figure 4A). Means of 0.09 ± 0.08 g kg−1 (upland zone), 0.05 ± 0.02 g kg−1 (piedmont zone), and 0.17 ± 0.14 g kg−1 (lowland zone) were measured for low-crystallinity Fe oxyhydroxides (Figure 4B).
Figure 4. Means ± standard deviation contents of (A) crystalline Fe oxide forms and (B) low-crystallinity Fe oxide forms, from sediment samples collected in the upland, piedmont, and lowland zones of the Cruxati River. Bars followed by the same lowercase letter indicate no statistical differences among the zones investigated by Kruskal-Wallis test at 5% significance level.
3.5 Phosphorus fractionation of sediments
The predominant P fractions across all zones were Oxide-P, Silicate-P, and Res-P (Figure 5). In the upland zone, Silicate-P was absent, whereas Ca-P occurred exclusively in this zone, with no measurable values in the piedmont and lowland zones. HA-P remained below detection limits in all samples. A consistent decrease in P content with depth was observed in every zone (Figure 5).
Figure 5. Phosphorus fractionation in sediments across the three hydrological zones: (A) Upland, (B) Piedmont, and (C) Lowland. P fractions included the following: Exchangeable P (EX-P), oxide-bound P (Oxide-P), silicate-bound P (Silicate-P), humic acid-bound P (HA-P), carbonate/apatite P (Ca-P), and residual P (Res-P). Oxide-P, Silicate-P, and Res-P were the most representative fractions, whereas Silicate-P was absent in the upland zone and HA-P was below detection limits in all samples.
Comparisons among zones showed no significant differences in EX-P (Figure 6A) and Oxide-P (Figure 6B) (p > 0.05). HA-P was absent in all cases. Significant differences were detected in Silicate-P (p < 0.001), Ca-P (p < 0.001), and Res-P (p = 0.048). Mean Silicate-P values were 0.0 ± 0.0 mg kg−1 (upland), 15.18 ± 4.0 mg kg−1 (piedmont), and 29.41 ± 24.84 mg kg−1 (lowland) (Figure 6C). For Ca-P, means were 14.40 ± 5.36 mg kg−1 (upland), 0.0 ± 0.0 mg kg−1 (piedmont), and 0.0 ± 0.0 mg kg−1 (lowland) (Figure 6D). Res-P contents averaged 15.03 ± 8.37 mg kg−1, 8.12 ± 7.6 mg kg−1, and 26.40 ± 16.8 mg kg−1 for upland, piedmont, and lowland, respectively (Figure 6E).
Figure 6. Means ± standard deviation contents for the P fractions (0–9 cm depth) across upland, piedmont, and lowland zones: (A) EX-P, (B) Oxide-P, (C) Silicate-P, (D) Ca-P, (E) Res-P, and (F) Total-P. Different lowercase letters indicate significant differences among zones (p < 0.05; Kruskal–Wallis test with Dunn's post-hoc test, Bonferroni correction). Total-P corresponds to the sum of all individual P fractions for each zone.
Total-P showed no statistical differences, but values tended to be higher in the lowland (102 ± 18 mg kg−1), almost twice those of the upland (62 ± 12 mg kg−1) and 2.5-fold higher than the piedmont (46 ± 8 mg kg−1), indicating an overall increase toward downstream areas (Figure 6F).
3.6 Discriminant analysis
The discriminant analysis clearly separated the upland zone from the piedmont and lowland zones, which formed a closer cluster (Figure 7A). The first discriminant function (Dim1; eigenvalue = 20.02) explained 97.84% of the total variability, mainly influenced by variables with higher correlation coefficients, including Ca-P (r = 0.36), Total-P (r = −0.40), Orthophosphate-P (r = −0.24), crystalline Fe (r = 0.31), and sediment pH (r = 0.95), which were associated with the upland zone (Figure 7B). In contrast, Eh in water (r = 0.33) and in sediments (r = 0.02), Silicate-P (r = −0.04), and EC (r = 0.48) showed stronger correlations with the piedmont and lowland zones. The second discriminant function (Dim2; eigenvalue = 4.41) accounted for 2.16% of the variance and was mainly driven by DO (r = −0.75) and EC (r = 0.48).
Figure 7. Discriminant analysis of sediments and water variables across the studied zones. (A) Ordination of upland, piedmont, and lowland zones with respective centroids. (B) Contribution of measured variables to the discriminant functions. Variables include: EX-P (exchangeable P), Oxide-P (Fe oxyhydroxide-bound P), Silicate-P (phyllosilicate/Al hydroxide-bound P), Ca-P (carbonate-bound P), HA-P (humic acid-bound P), Res-P (residual P), Crystalline Fe (Fe in crystalline minerals), Low-crystalline Fe (Fe in poorly crystalline minerals), TOC (total organic carbon), DO (dissolved oxygen), pH (sediments and water), Eh (sediments and water), and EC (electrical conductivity). Dim1 explained 97.84% and Dim2 2.16% of total variability.
3.7 Spatial context of land use and water quality
Land use and land cover patterns in the Cruxati River Basin (Figure 8) showed distinct characteristics across the three sampling zones. In the upland zone, sampling points are located in areas largely surrounded by natural forest formations, which together represent more than 70% of the land cover, with smaller patches of pasture (~17%) and mosaic of uses (~8%) which refers to agro-pastoral areas where it is not possible to distinguish between pasture and cropland, while urban areas account for about 1% (Figure 8). This setting represents the upper section of the basin, where land cover remains relatively preserved, although small-scale anthropogenic activities are present. The piedmont zone is characterized by a more heterogeneous configuration, with pastures and mosaics of uses becoming more prominent and forest patches less continuous. In the lowland zone, closer to the river mouth, the landscape is more fragmented, marked by a greater presence of pastures and urban areas, alongside aquaculture ponds and water bodies. These spatial contrasts highlight a progressive intensification of human activities from upland to lowland, coinciding with the natural tendency for material accumulation in the lower sections of the basin.
Figure 8. (A, B) Land use and land cover in the Cruxati River Basin (MapBiomas Collection 5, 2019), showing land use and land cover in the surroundings of the three sampled zones (i.e., upland zone, piedmont zone, and lowland zone). The classes include Forest Formation, Savanna Formation, Mangrove, Grassland, Pasture, Mosaic of Uses, Urban Area, Other Non-Vegetated Areas, Aquaculture, and Water Bodies. Sampling points represent the upland, piedmont, and lowland zones. Maps were produced in QGIS 3.34 (QGIS.org, 2024) using shapefiles from MapBiomas. The data are publicly available and reproduced in accordance with the project license.
The water quality assessment of monitored reservoirs (Figure 9) also reflects spatial variability within the basin. According to the Trophic State Index (TSI), which integrates total phosphorus concentrations as the causal variable and chlorophyll-a concentrations as the biological response (Table 1), the Poço Verde (PCV-01), Quandú (QUA-01), and Mundaú (MUN-08) reservoirs were classified as eutrophic, while the Gameleira (GAM-01) reservoir was classified as mesotrophic. Spatially, the eutrophic reservoirs are located downstream or in areas with denser land occupation, in closer proximity to the piedmont and lowland sampling zones. In contrast, the Gameleira reservoir is positioned upstream in a relatively preserved portion of the basin, influencing the upland zone. This distribution indicates that the sampling points along the Cruxati River are embedded in a heterogeneous landscape where both preserved and impacted water bodies coexist, reinforcing the importance of spatial context for interpreting environmental gradients across the basin.
Figure 9. Trophic state of monitored reservoirs in the Cruxati Basin, Ceará, Brazil. Reservoirs were classified as eutrophic (Poço Verde – PCV-01, Quandú – QUA-01, Mundaú – MUN-08) and mesotrophic (Gameleira – GAM-01), according to the Trophic State Index (TSI), which integrates total phosphorus and chlorophyll-a concentrations (Table 1). Maps were produced in QGIS 3.34 (QGIS.org, 2024) using hydrological and reservoir monitoring data provided by COGERH and shape files from IBGE.
4 Discussion
4.1 Geology and the surrounding environment control P dynamics in IRES
The observed particle size distribution is consistent with the regional geology, characterized by metamorphic and sedimentary rocks from “Barreiras Formation” (Nunes et al., 2011). Moreover, the predominance of kaolinite (Figure 3), particularly under humid tropical conditions, is noteworthy (Correa et al., 2008; Melo et al., 2002a,b). Under these conditions, the presence of Fe3? commonly induces the formation of short-range order Fe phases, which contains high surface area and commonly adsorb P from soils and sediments (Melo et al., 2002a,b; Correa et al., 2008; Dixon and Schulze, 2002). Fe oxyhydroxides (crystalline and low-crystallinity forms; Figure 4), known for their stability and high surface area, effectively immobilize P (Da-Peng and Yong, 2010; Queiroz et al., 2022). This observation aligns with our findings of elevated P contents associated with silicate-P and oxide-P fractions (Figure 5). Additionally, in semiarid regions, where evapotranspiration exceeds precipitation, surface sediments tend to be more oxidized (Queiroz et al., 2018) and promote the precipitation of Fe oxyhydroxides, further reinforcing P retention (Baldwin and Mitchell, 2000; Da-Peng and Yong, 2010; Attygalla et al., 2016; Kong et al., 2021), particularly in the upper sediment layers (Figure 5).
The activities across the Cruxati (i.e., small-scale agriculture, livestock grazing, and rural settlements are widespread across the Cruxati Basin; Figure 8) contribute to downstream nutrient enrichment, reinforcing the combined influence of natural processes (e.g., organic P mineralization, represented by Res-P; Figure 5) and anthropogenic pressures on P cycling (Kong et al., 2021; Gómez-Gener et al., 2021). In contrast, the upland zone—predominantly forested—is less affected by direct anthropogenic actions. This spatial configuration, combined with the topographic gradient, facilitates the routing and concentration of P fluxes toward the piedmont and lowland zones, which act as depositional environments.
The influence of anthropogenic activities is further evidenced by the presence of reservoirs within the basin (Poço Verde, Quandú, and Mundaú), which are predominantly classified as eutrophic (Figure 9). Originally designed for domestic, agricultural, and industrial water supply, these reservoirs also serve as modulators of nutrient dynamics (Ferreira and Fernandes, 2022; Yu et al., 2023; Li et al., 2024; Liu et al., 2024). Their trophic state reflects elevated nutrient availability-particularly phosphorus-derived from surrounding land uses such as agriculture, livestock, and rural settlements (Figure 8).
By retaining and concentrating upstream inputs, eutrophic reservoirs can shift from nutrient sinks to sources when accumulated P is mobilized under changing hydrodynamic or biogeochemical conditions. Overflow events and flood flushing can resuspend particulate P, rapidly transporting it downstream and generating pulses of total P alongside transient increases in dissolved forms (Wang, 2020; Zhao et al., 2024). Controlled releases and environmental flow discharges may also export P-rich waters, especially when withdrawals occur from stratified or oxygen-depleted layers enriched in dissolved inorganic P formed via sediment diagenesis or algal decay (Yu et al., 2023). Sediment resuspension and internal loading further enhance DIP release under anoxic or warming conditions, where Fe- and Al-bound P is desorbed and organic matter mineralization amplifies endogenous fluxes (Vilmin et al., 2022; Shou et al., 2024).
Once exported, this biogeochemically transformed P remains bioavailable and sustains downstream enrichment, bypassing particle-trapping processes and fueling primary production (Vilmin et al., 2022; Li et al., 2024). In cascaded systems such as the Cruxati Basin, sequential reservoirs - Poço Verde, Quandú, and Mundaú (Figure 9)—can amplify these effects, with cumulative releases increasing basin-scale P fluxes (Li et al., 2024; Shou et al., 2024). In this context, the significant Res-P enrichment in the lowland zone (Figure 6) indicates that organic matter decomposition and mineralization remain key local pathways of P release. However, land use patterns (Figure 8) and trophic state classifications (Figure 9) demonstrate that external contributions from agriculture, livestock, and domestic effluents, combined with reservoir-mediated exports, act as additional drivers of P delivery downstream. Therefore, the elevated P concentrations observed cannot be attributed to a single dominant mechanism but rather emerge from the interplay between organic matter degradation, anthropogenic land use, and reservoir dynamics.
4.2 Spatial distribution of P in IRES
Phosphorus transport and transformation along the Cruxati River follow a systematic spatial gradient governed by sediment deposition, organic matter dynamics, and mineral interactions. Our findings underscore the importance of landscape position in assessing nutrient cycling in intermittent rivers, particularly in semiarid regions where fluctuations in water availability and land use can substantially alter biogeochemical processes. The higher total P concentrations observed in the lowland zone (102.02 ± 18 mg kg−1; Figure 6F) indicate that P transport and retention are largely controlled by topographic position. This downstream P enrichment reflects a source-to-sink dynamic, with the lowland zone acting as a depositional environment that enhances P retention (Tooth, 2000).
A key finding supporting this pattern is the distribution of resistant phosphorus (Res-P; Figure 5E). Res-P represents a long-term storage form of P, primarily associated with recalcitrant organic matter and stable mineral phases, making this P form less bioavailable under natural stream conditions (Kelleher and Simpson, 2006). The accumulation of Res-P in the lowland zone suggests that organic P transformation, enhanced deposition, and potential external inputs contribute to the observed pattern. Likewise, the enrichment of Silicate-P in this zone (Figure 6C), together with the lower concentrations of bioavailable orthophosphate in the water column (Figure 2B), indicates that Res-P mineralization may release P, while adsorption onto silicate surfaces acts as a subsequent sink. The corresponding decline in dissolved orthophosphate-P from upland to lowland (Figure 2B) supports this immobilization process, as the bioavailable orthophosphate released during mineralization is rapidly adsorbed by silicate minerals, reducing its concentration in the water column. These findings indicate a spatial P flow along the stream, where Res-P mineralization in the lowland zone releases P that is subsequently retained as Silicate-P, decreasing the bioavailable P fraction in the water. This mechanism emphasizes the role of sediment composition in regulating P availability across river sections (Da-Peng and Yong, 2010; Schönbrunner et al., 2012; Attygalla et al., 2016; Cavalcante et al., 2018; Zhang et al., 2018).
Unlike Silicate-P, oxide-bound P (Oxide-P) fractions did not show significant spatial variation (Figure 6B), indicating that Fe oxyhydroxides were not the main factors governing P retention and release-despite the well-known P adsorption capacity of Fe oxyhydroxides (Da-Peng and Yong, 2010; Queiroz et al., 2022; Baldwin et al., 2000; Kong et al., 2021). Similarly, Fe oxyhydroxides (both crystalline and poorly crystalline phases) exhibited no significant differences among zones (Figure 4), reinforcing that P cycling in this system is primarily driven by organic matter dynamics and silicate interactions rather than Fe phases.
Considered a stable P sink in semiarid regions (Nóbrega et al., 2014), the absence of the calcium-bound P (Ca-P) in the lower zones may result from greater sediment accumulation and moisture, which promote carbonate dissolution (Jaeger et al., 2017). Under such conditions, increased carbonic acid production enhances Ca-P solubility, leading to P release into the system (Kong et al., 2021). In the upland zone (Figure 6D), lower moisture and reduced carbonic acid production likely limit dissolution, allowing Ca-P to remain a stable P reservoir.
As shown by the discriminant analysis (Figure 7), a clear distinction exists between the upland and the two lower zones (piedmont and lowland), emphasizing the influence of hydrological and geochemical gradients on P distribution. In the lower zones, increased surface–subsurface connectivity enhances sediment moisture, underscoring the role of geomorphology in regulating P cycling during flow periods (Jaeger et al., 2017). The potential mobilization of P from Ca-P in the upland zone, followed by its downstream transport and retention, may further explain the spatial pattern of P transformation. These results highlight that Ca-P should be considered not only a sink but also a dynamic component of P cycling, especially in intermittent river systems where hydrological and geochemical conditions vary markedly across landscape positions. This reinforces the role of IRES as nutrient conduits, where P is not only transported downstream but also transformed and redistributed among biogeochemical compartments, ultimately influencing nutrient availability for primary productivity in adjacent ecosystems (Da Silva et al., 2024).
4.3 Future perspectives of IRES in the semiarid Northeast faced with climate changes
In arid and semiarid regions, climate change is expected to intensify intermittency in river regimes, lengthening dry periods and altering the timing and magnitude of nutrient pulses (Nabih et al., 2021; Da Silva et al., 2024). In the Brazilian Northeast, this challenge is compounded by the proliferation of artificial dams, with nearly 70,000 reservoirs currently disrupting natural flow regimes (Rocha Junior et al., 2024). Although such structures are essential for water storage and supply, they further fragment IRES and may lead to the progressive loss of these ecosystems.
Given that IRES function as critical conduits for nutrient transport, any disruption in their hydrological dynamics has direct consequences for P cycling. Our findings illustrate how intermittency enhances the role of stable sediment fractions in regulating P availability. However, reservoir operation and altered hydroperiods can modulate this balance: cascade reservoirs in semiarid regions retain substantial fractions of particulate and total P, but internal processes such as anoxia, sediment resuspension, and prolonged hydraulic retention times can trigger P release, transforming reservoirs from sinks into episodic sources (Li et al., 2024; Nakulopa et al., 2024; Zhao et al., 2024).
These insights underscore the need to integrate hydrological management with nutrient dynamics. Adaptive strategies such as optimizing hydraulic retention times in reservoirs, incorporating water quality considerations into environmental flow management, and applying measures to reduce internal P loading are essential to maintain downstream nutrient supply and ecosystem services (Yu et al., 2023; Bartoszek and Koszelnik, 2025). Furthermore, soil and riparian restoration in the Caatinga can buffer nutrient losses and enhance ecosystem resilience (Da Silva et al., 2024). Under scenarios of climate change and river regulation in semiarid Brazil, where the interplay among water scarcity, nutrient limitation, and human activities may influence ecosystem functioning and resilience. Protecting these systems is not only vital for the semiarid Brazilian Northeast but also provides broader lessons for dryland rivers worldwide, where increasing intermittency will likely reshape nutrient cycling under future climate scenarios.
5 Conclusion
Our results revealed a clear gradient of P fractions along the upland–piedmont–lowland continuum, with oxide-P, silicate-P, and residual-P as the most representative pools. The exclusive occurrence of Ca-P in upland sediments and the higher residual-P contents in lowland zones demonstrate how geomorphological and hydrological conditions regulate P retention and accumulation. The increase in total P stocks toward the lowland zone highlights the role of IRES as nutrient carriers from upstream to depositional environments, reinforcing their contribution to the productivity of downstream and adjacent ecosystems. The predominance of residual-P in lowland sediments further suggests that this less bioavailable pool may act as a long-term reservoir, providing P during drought periods when river flow ceases and supporting nutrient supply to both the river channel and surrounding ecosystems.
Spatial analysis of land use and land cover corroborated these patterns, indicating a progressive intensification of anthropogenic activities from upland to lowland areas, while the trophic classification of nearby reservoirs revealed predominantly eutrophic conditions. Together, these findings demonstrate that P cycling in the basin arises from the interaction between natural retention and mineralization processes and human-induced pressures associated with land use and water management structures.
Our findings show that there is not a single dominant mechanism governing P dynamic. Instead, they emerge from the interaction among geomorphological controls, mineral and organic transformations, land use patterns, and reservoir-mediated exports. This integrated perspective advances our understanding of nutrient cycling in semiarid intermittent rivers and highlights the need to consider both natural and anthropogenic drivers when assessing watershed functioning and resilience. Long-term studies that monitor these watercourses spatially and temporally are needed, and interdisciplinary approaches that integrate different fields of knowledge are essential to fully interpret the complexity of IRES.
Data availability statement
The raw data supporting the conclusions of this article will be made available by the authors, without undue reservation.
Author contributions
AS: Conceptualization, Data curation, Formal analysis, Investigation, Methodology, Project administration, Software, Validation, Writing – original draft, Writing – review & editing. HQ: Data curation, Formal analysis, Methodology, Software, Supervision, Writing – original draft, Writing – review & editing. DO: Data curation, Investigation, Methodology, Software, Writing – original draft, Writing – review & editing. GA: Methodology, Software, Writing – original draft, Writing – review & editing. TP: Investigation, Methodology, Writing – original draft, Writing – review & editing. GN: Data curation, Methodology, Software, Writing – original draft, Writing – review & editing. DB: Data curation, Methodology, Writing – original draft, Writing – review & editing. CR: Conceptualization, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Visualization, Writing – original draft, Writing – review & editing, Data curation, Formal analysis, Validation. TF: Conceptualization, Data curation, Formal analysis, Funding acquisition, Investigation, Methodology, Project administration, Resources, Software, Supervision, Validation, Visualization, Writing – original draft, Writing – review & editing.
Funding
The author(s) declared that financial support was received for this work and/or its publication. This work was supported by the National Council for Scientific and Technological Development (CNPq, grants number 430010/2018-4, 305996/2018-5 to TF, 307707/2025-3 to HQ, and 311627/2015-3 to CR), Coordination of Superior Level Staff Improvement (CAPES, Finance Code 001), Fundação Cearense de Apoio ao Desenvolvimento Científico e Tecnológico (FUNCAP) from Programa de Apoio a Núcleos Emergentes PRONEM/FUNCAP/ CNPq (PNE-0112-00026.01 0.00/16), and USP Process n. 22.1.09345.01.2 (University of São Paulo, project number 10, Notice Program for Support to New Faculty Members 2024/1).
Acknowledgments
We acknowledge the Center for Carbon Research in Tropical Agriculture (CCARBON).
Conflict of interest
The author(s) declared that this work was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.
Generative AI statement
The author(s) declared that generative AI was not used in the creation of this manuscript.
Any alternative text (alt text) provided alongside figures in this article has been generated by Frontiers with the support of artificial intelligence and reasonable efforts have been made to ensure accuracy, including review by the authors wherever possible. If you identify any issues, please contact us.
Publisher's note
All claims expressed in this article are solely those of the authors and do not necessarily represent those of their affiliated organizations, or those of the publisher, the editors and the reviewers. Any product that may be evaluated in this article, or claim that may be made by its manufacturer, is not guaranteed or endorsed by the publisher.
References
Acuña, V., Hunter, M., and Ruhí, A. (2017). Managing temporary streams and rivers as unique rather than second-class ecosystems. Biol. Conserv. 211, 12–19. doi: 10.1016/j.biocon.12.025
Attygalla, N. W., Baldwin, D. S., Silvester, E., Kappen, P., and Whitworth, K. L. (2016). The severity of sediment desiccation affects the adsorption characteristics and speciation of phosphorus. Environ. Sci. Processes Impacts 18, 64–71. doi: 10.1039/C5EM00523J
Baldwin, D. S., and Mitchell, A. M. (2000). The effects of drying and re-flooding on the sediment and soil nutrient dynamics of lowland river-floodplain systems: a synthesis. Regul. Rivers: Res. Manage. 16, 457–467. doi: 10.1002/1099-1646(200009/10)16:5<457::AID-RRR597>3.0.CO;2-B
Baldwin, D. S., Mitchell, A. M., and Rees, G. N. (2000). The effects of in situ drying on sediment-phosphate interactions in sediments from an old wetland. Hydrobiologia 431, 3–12. doi: 10.1023/A:1004015019608
Barcellos, D., Queiroz, H. M., Nóbrega, G. N., de Oliveira Filho, R. L., Santaella, S. T., Otero, X. L., et al. (2019). Phosphorus-enriched effluents increase eutrophication risks for mangrove systems in northeastern Brazil. Marine Pollut. Bull. 142, 58–63. doi: 10.1016/j.marpolbul.03.031
Bartoszek, L., and Koszelnik, P. (2025). The efficiency of phosphate inactivation in the bottom sediments of small retention reservoirs using a combination of calcium and iron salts, ex situ studies. Sci. Rep. 15:32719. doi: 10.1038/s41598-025-17779-4
Cavalcante, H., Araújo, F., Noyma, N. P., and Becker, V. (2018). Phosphorus fractionation in sediments of tropical semiarid reservoirs. Sci. Total Environ. 619–620, 1022–1029. doi: 10.1016/j.scitotenv.11.204
Chen, P. Y. (1977). Table of Key Lines in X-ray Powder Diffraction Patterns of Minerals in Clays and Associated Rocks. Bloomington, Indiana: Department of natural resources, Indiana Geological Survey.
Chiu, M.-C., Leigh, C., Mazor, R., Cid, N., and Resh, V. (2017). “Anthropogenic threats to intermittent rivers and ephemeral streams,” in Intermittent Rivers and Ephemeral Streams (Amsterdam: Elsevier), 433–454. doi: 10.1016/B978-0-12-803835-2.00017–6
COGERH - Companhia de Gestão dos Recursos Hídricos (2012). Bacias hidrográficas do Litoral: Características gerais. Fortaleza: COGERH.
Correa, M. M., Ker, J. C., Barrón, V., Fontes, M. P., Torrent, J., Curi, N., et al. (2008). Characterizing iron oxides from coastal and central plain soils. Rev. Bras. Ciênc. Solo 32, 1017–1031. doi: 10.1590/S0100-06832008000300011
Da Silva, A. E. B., Queiroz, H. M., de Oliveira, D. P., Nóbrega, G. N., Pereira, T. M., de Freitas Lima, P., et al. (2024). Effects of hydrological dynamics in controlling phosphorus bioavailability in intermittent rivers and implications for estuaries. Front. Water 6:1286968. doi: 10.3389/frwa.2024.1286968
Da-Peng, L., and Yong, H. (2010). Sedimentary phosphorus fractions and bioavailability as influenced by repeated sediment resuspension. Ecol. Eng. 36, 958–962. doi: 10.1016/j.ecoleng.04.014
Darke, A. K., and Walbridge, M. R. (2000). Al and Fe biogeochemistry in a floodplain forest: implications for P retention. Biogeochemistry 51, 1–32. doi: 10.1023/A:1006302600347
Datry, T., Larned, S. T., and Tockner, K. (2014). Intermittent rivers: a challenge for freshwater ecology. Bioscience 64, 229–235. doi: 10.1093/biosci/bit027
Dixon, J. B., and Schulze, D. G. (2002). Soil Mineralogy with Environmental Applications. Madison: Soil Science Society of America, Inc. doi: 10.2136/sssabookser7
Duarte, M. R. N., Pereira, T. M., de Lima, P. F., Pereira, E. C. B., Lopes, F. B., Rezende, C. F., et al. (2021). Limnological dynamics in an artificial reservoir and intermittent river in the semi-arid region as a function of land use and occupation. Rev. Ciênc. Agron. 52, 1–10. doi: 10.5935/1806-6690.20210010
Eaton, A. D., and Franson, M. A. H. (Eds.). (2005). Standard Methods for the Examination of Water and Wastewater.
Fávero, L. P. L., Belfiore, P., Silva, F. L., and Chan, B. L. (2009). Análise de dados: Modelagem multivariada para tomada de decisões. Rio de Janeiro: Elsevier.
Ferreira, D. M., and Fernandes, C. V. S. (2022). Integrated water quality modeling in a river-reservoir system to support watershed management. J. Environ. Manage. 324:116447. doi: 10.1016/j.jenvman.2022.116447
Fovet, O., Belemtougri, A., Boithias, L., Braud, I., Charlier, J., Cottet, M., et al. (2021). Intermittent rivers and ephemeral streams: perspectives for critical zone science and research on socio-ecosystems. WIREs Water 8, 1–33. doi: 10.1002/wat2.1523
Funceme - Fundação Cearense de Meteorologia e Recursos Hídricos. (2018). Projetos e Mapas. Fortaleza: Funceme.
Gee, G. W., and Bauder, J. W. (1986). “Particle-size analysis,” in Methods of Soil Analysis: Part 1-Physical and Mineralogical Methods (Madison: Soil Science Society of America). doi: 10.2136/sssabookser5.1.2ed.c15
Giulietti, A. M., Bocage Neta, A. L., Castro, A. A. J. F., Gamarra-Rojas, C. F. L., Sampaio, E. V. S. B., Virgínio, J. F., et al. (2004). “Diagnóstico da vegetação nativa do bioma Caatinga,” in Biodiversidade da Caatinga: Áreas e Ações Prioritárias para a Conservação (Brasília: MMA/UFPE), 45–90.
Gómez-Gener, L., Siebers, A. R., Arce, M. I., Arnon, S., Bernal, S., Bolpagni, R., et al. (2021). Towards an improved understanding of biogeochemical processes across surface-groundwater interactions in intermittent rivers and ephemeral streams. Earth-Sci. Rev. 220:103724. doi: 10.1016/j.earscirev.2021.103724
Gonçalves-Silva, M., Manna, L. R., Rodrigues-Filho, C. A. S., Teixeira, F. K., and Rezende, C. F. (2022). Effect of drying dynamics on the functional structure of a fish assemblage from an intermittent river network. Front. Environ. Sci. 10:903974. doi: 10.3389/fenvs.2022.903974
Gotelli, N. J., and Ellison, A. M. (2013). A Primer of Ecological Statistics. Sunderland: Sinauer Associates.
Hammer, Ø., Harper, D. A. T., and Ryan, P. D. (2001). PAST: paleontological statistics software package for education and data analysis. Palaeontol. Electronica 4, 1–9.
Holtan, H., Kamp-Nielsen, L., and Stuanes, A. O. (1988). Phosphorus in soil, water and sediment: an overview. Hydrobiologia 170, 19–34. doi: 10.1007/BF00024896
Jaeger, K. L., Sutfin, N. A., Tooth, S., Michaelides, K., and Singer, M. (2017). “Geomorphology and sediment regimes of intermittent rivers and ephemeral streams,” in Intermittent Rivers and Ephemeral Streams: Ecology and Management, eds. T. Datry, N. Bonada, and A. Bouton (London: Elsevier Academic Press), 21–49. doi: 10.1016/B978-0-12-803835-2.00002–4
Kelleher, B. P., and Simpson, A. J. (2006). Humic substances in soils: are they really chemically distinct? Environ. Sci. Technol. 40, 4605–4611. doi: 10.1021/es0608085
Kim, B., Gautier, M., Simidoff, A., Sanglar, C., Chatain, V., Michel, P., et al. (2016). pH and Eh effects on phosphorus fate in constructed wetland's sludge surface deposit. J. Environ. Manage. 183, 175–181. doi: 10.1016/j.jenvman.08.064
Kong, M., Han, T., Chen, M., Zhao, D., Chao, J., Zhang, Y., et al. (2021). High mobilization of phosphorus in black-odor river sediments with the increase of temperature. Sci. Total Environ. 775:145595. doi: 10.1016/j.scitotenv.2021.145595
Larned, S. T., Datry, T., Arscott, D. B., and Tockner, K. (2010). Emerging concepts in temporary-river ecology. Freshwater Biol. 55, 717–738. doi: 10.1111/j.1365-2427.2009.02322.x
Li, H., Zhou, J., and Zhang, M. (2023). Regime of fluvial phosphorus constituted by sediment. Front. Environ. Sci. 11:1093413. doi: 10.3389/fenvs.2023.1093413
Li, L., Chen, L., Chen, S., Zhang, Y., Xu, Y., Zhi, X., et al. (2024). The cumulative effects of cascade reservoirs control nitrogen and phosphorus flux: base on biogeochemical processes. Water Res. 252:121177. doi: 10.1016/j.watres.2024.121177
Liu, D., Chen, Q., Maavara, T., Zhang, J., and Chen, Y. (2024). Nitrogen cycling in reservoir drawdown areas and the impacts on water quality. Global Biogeochem. Cycles 38:e2023GB007919. doi: 10.1029/2024GB008136
Maavara, T., Parsons, C. T., Ridenour, C., Stojanovic, S., Dürr, H. H., Powley, H. R., et al. (2015). Global phosphorus retention by river damming. Proc. Natl. Acad. Sci. 112, 15603–15608. doi: 10.1073/pnas.1511797112
Mamede, G. L., Guentner, A., Medeiros, P. H. A., Araújo, J. C., and Bronstert, A. (2018). Modeling the effect of multiple reservoirs on water and sediment dynamics in a semiarid catchment in Brazil. J. Hydrol. Eng. 23, 1–13. doi: 10.1061/(ASCE)HE.1943-5584.0001701
Manly, B. F. J., and Alberto, J. A. N. (2016). Multivariate Statistical Methods: A Primer. Boca Raton, FL: CRC Press. doi: 10.1201/9781315382135
Mckeague, J. A. (1978). Manual on Soil Sampling and Methods of Analysis. Ottowa: Canadian Society of Soil Science, 212.
Mehra, O. P., and Jackson, M. L. (1960). Iron oxide removal from soils and clay by a dithionite-citrate system bulfered with sodium bicarbonate. Clay Clay Miner. 7, 317–327. doi: 10.1346/CCMN.1958.0070122
Meireles, A. C. M., Frischkorn, H., and Andrade, E. M. (2007). Seasonal quality changes in waters from Edson Queiroz dam, Acaraú basin, in the semiarid of Ceará. Rev. Ciênc. Agron. 38, 25–31.
Melo, V. F., Novais, R. F., Schaefer, C. E. G. R., Fontes, M. P. F., and Singh, B. (2002a). Chemical and crystal properties of kaolinite and iron oxides in the Barreiras group sediments in Aracruz, Espírito Santo state, Brazil. Rev. Bras. Ciênc. Solo 26, 53–64. doi: 10.1590/S0100-06832002000100006
Melo, V. F., Novais, R. F., Schaefer, C. E. G. R., Fontes, M. P. F., and Singh, B. (2002b). Mineralogy of the sand, silt and clay fractions of Barreiras group sediments in Aracruz, Espírito Santo state, Brazil. Rev. Bras. Ciênc. Solo 26, 29–41. doi: 10.1590/S0100-06832002000100004
Mulholland, P. J., and Webster, J. R. (2010). Nutrient dynamics in streams and the role of J-NABS. J. N. Am. Benthol. Soc. 29, 100–117. doi: 10.1899/08-035.1
Murphy, J., and Riley, J. P. (1962). A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 27, 31–36. doi: 10.1016/S0003-2670(00)88444-5
Nabih, S., Tzoraki, O., Zanis, P., Tsikerdekis, T., Akritidis, D., Kontogeorgos, I., et al. (2021). Alteration of the ecohydrological status of the intermittent flow rivers and ephemeral streams due to the climate change impact (Case Study: Tsiknias River). Hydrology 8:43. doi: 10.3390/hydrology8010043
Nakulopa, F., Bärlund, I., and Borchardt, D. (2024). How a reservoir modulates downstream water quality under declining upstream loading and progressing climate change. Sci. Total Environ. 912:169460. doi: 10.1016/j.scitotenv.2023.169460
Nóbrega, G. N., Otero, X. L., Macías, F., and Ferreira, T. O. (2014). Phosphorus geochemistry in a Brazilian semiarid mangrove soil affected by shrimp farm effluents. Environ. Monit. Assess. 186, 5749–5762. doi: 10.1007/s10661-014-3817-3
Nunes, F. C., Silva, E. F., and Vilas Boas, G. S. (2011). Grupo Barreiras: características, gênese e evidências de neotectonismo. Rio de Janeiro: Embrapa Solos.
O'Connell, C. S., Ruan, L. S., and Whendee, L. (2018). Drought drives rapid shifts in tropical rainforest soil biogeochemistry and greenhouse gas emissions. Nat. Commun. 9, 1–9. doi: 10.1038/s41467-018-03352-3
Paludan, C., and Jensen, H. S. (1995). Sequential extraction of phosphorus in freshwater wetland and lake sediment: significance of humic acids. Wetlands 15, 365–373. doi: 10.1007/BF03160891
Paludan, C., and Morris, J. T. (1999). Distribution and speciation of phosphorus along a salinity gradient in intertidal marsh sediments. Biogeochemistry 45, 197–221. doi: 10.1023/A:1006136621465
Paulino, W. D., Reis, R., Oliveira, A., and Avelino, F. F. (2013). Classificação do estado trófico para o gerenciamento de reservatórios no semiárido: a experiência da Cogerh no estado Ceará. Fortaleza: ABRH, Associação Brasileira de Recursos Hídricos, 1–8.
Peng, Z., Du, Y., Niu, S., Xi, L., Niu, Y., and Li, Y. (2022). Differences in nitrogen and phosphorus sinks between the harvest and non-harvest of Miscanthus lutarioriparius in the Dongting Lake wetlands. Front. Plant Sci. 13:989931. doi: 10.3389/fpls.2022.989931
Pettersson, K., Boström, B., and Jacobsen, O. S. (1988). Phosphorus in sediments-speciation and analysis. Hydrobiologia 170, 91–101. doi: 10.1007/BF00024900
Porto, M. M., Andrade, E. M., Costa, R. N. T., Lemos Filho, L. C. A., and Meireles, M. (2004). Identificação de bacias hidrográficas com características físicas similares no Estado do Ceará. Rev. Ciênc. Agron. 35, 17–25.
Queiroz, H. M., Ferreira, T. O., Fandiño, V. A., Bragantini, I. O. B. F., Barcellos, D., Nóbrega, G. N., et al. (2022). Changes in soil iron biogeochemistry in response to mangrove dieback. Biogeochemistry 158, 357–372. doi: 10.1007/s10533-022-00903-1
Queiroz, H. M., Nóbrega, G. N., Otero, X. L., and Ferreira, T. O. (2018). Are acid volatile sulfides (AVS) important trace metals sinks in semi-arid mangroves? Marine Pollut. Bull. 126, 318–322. doi: 10.1016/j.marpolbul.11.020
Reynolds, C. S., and Davies, P. S. (2001). Sources and bioavailability of phosphorus fractions in freshwaters: a British perspective. Biol. Rev. 76, 27–64. doi: 10.1111/j.1469-185X.2000.tb00058.x
Rocha Junior, C., Araújo, F., and Becker, V. (2024). Influence of land use on spatial distribution of mobile phosphorus forms in the sediment of a tropical semi-arid reservoir. Sci. Total Environ. 914:169836. doi: 10.1016/j.scitotenv.2023.169836
Sánchez-Montoya, M. M., Datry, T., Ruhi, A., Carlson, S. M., Corti, R., and Tockner, K. (2023). Intermittent rivers and ephemeral streams are pivotal corridors for aquatic and terrestrial animals. BioScience 73, 291–301. doi: 10.1093/biosci/biad004
Schönbrunner, I. M., Preiner, S., and Hein, T. (2012). Impact of drying and re-flooding of sediment on phosphorus dynamics of river-floodplain systems. Sci. Total Environ. 432, 329–337. doi: 10.1016/j.scitotenv.06.025
Shou, C.-Y., Yue, F.-J., Zhou, B., Fu, X., Ma, Z.-N., Gong, Y.-Q., et al. (2024). Chronic increasing nitrogen and endogenous phosphorus release from sediment threaten to the water quality in a semi-humid region reservoir. Sci. Total Environ. 931:172924. doi: 10.1016/j.scitotenv.2024.172924
Shumilova, O., Tockner, K., Gurnell, A. M., Langhans, S. D., Righetti, M., Lucía, A., et al. (2019). Floating matter: a neglected component of the ecological integrity of rivers. Aquat. Sci. 81, 1–20. doi: 10.1007/s00027-019-0619-2
Siregar, A., Kleber, M., Mikutta, R., and Jahn, R. (2005). Sodium hypochlorite oxidation reduces soil organic matter concentrations without affecting inorganic soil constituents. Eur. J. Soil Sci. 56, 481–490. doi: 10.1111/j.1365-2389.2004.00680.x
Teixeira, F. J. C. (2004). Modelos de gerenciamento de recursos hídricos: análises e propostas de aperfeiçoamento do sistema do Ceará. Brasília: Banco Mundial e Ministério da Integração Nacional.
Tooth, S. (2000). Downstream changes in dryland river channels: the northern plains of arid central Australia. Geomorphology 34, 33–54. doi: 10.1016/S0169-555X(99)00130-0
Vilmin, L., Bouwman, A. F., Beusen, A. H. W., van Hoek, W. J., and Mogollón, J. M. (2022). Past anthropogenic activities offset dissolved inorganic phosphorus retention in the Mississippi River basin. Biogeochemistry 161, 157–169. doi: 10.1007/s10533-022-00973-1
von Schiller, D., Acuña, V., Aristi, I., Arroita, M., Basaguren, A., Bellin, A., et al. (2017a). River ecosystem processes: a synthesis of approaches, criteria of use and sensitivity to environmental stressors. Sci. Total Environ. 596–597, 465–480. doi: 10.1016/j.scitotenv.04.081
von Schiller, D., Acuña, V., Graeber, D., Martí, E., Ribot, M., Sabater, S., et al. (2011). Contraction, fragmentation and expansion dynamics determine nutrient availability in a Mediterranean forest stream. Aquat. Sci. 73, 485–497. doi: 10.1007/s00027-011-0195-6
von Schiller, D. V., Bernal, S., Dahm, C. N., and Martí, E. (2017b). “Nutrient and organic matter dynamics in intermittent rivers and ephemeral streams,” in Intermittent Rivers and Ephemeral Streams: Ecology and Management, eds. T. Datry, N. Bonada, A. Bouton (London: Elsevier Academic Press), 135–160. doi: 10.1016/B978-0-12-803835-2.00006–1
Wang, F. (2020). Impact of a large sub-tropical reservoir on the cycling of nutrients in a river. Water Res. 186:116363. doi: 10.1016/j.watres.2020.116363
Xing, X., Wang, P., Wang, X., Yuan, Q., Hu, B., Liu, S., et al. (2024). Dams alter the control pattern of watershed land use to riverine nutrient distribution: comparison of three major rivers under different hydropower development levels in Southwestern China. Water Res. 260:121951. doi: 10.1016/j.watres.2024.121951
Yu, C., Xu, Z., Li, Y., Yang, Y., Cai, Y., Yang, Z., et al. (2023). Rethinking environmental flow management strategies in reservoir operations from an integrated water quality perspective. J. Hydrol. 626:130196. doi: 10.1016/j.jhydrol.2023.130196
Zhang, W., Jin, X., Meng, X., Tang, W., and Shan, B. (2018). Phosphorus transformations at the sediment water interface in shallow freshwater ecosystems caused by decomposition of plant debris. Chemosphere 201, 328–334. doi: 10.1016/j.chemosphere.03.006
Zhao, B., Zeng, Q., Wang, J., Jiang, Y., Liu, H., Yan, L., et al. (2024). Impact of cascade reservoirs on nutrients transported downstream and regulation method based on hydraulic retention time. Water Res. 252:121187. doi: 10.1016/j.watres.2024.121187
Keywords: freshwater ecosystems, IRES, limiting nutrients, P geochemistry, sediment geochemistry, semiarid ecosystems
Citation: Silva AEB, Queiroz HM, Oliveira DP, Araujo Lopes GB, Pereira TM, Nóbrega GN, Barcellos D, Rezende CF and Ferreira TO (2026) Landscape-driven phosphorus distribution and transformations in an intermittent river of semiarid Brazil. Front. Water 7:1716672. doi: 10.3389/frwa.2025.1716672
Received: 30 September 2025; Revised: 22 December 2025; Accepted: 29 December 2025;
Published: 26 January 2026.
Edited by:
George P. Karatzas, Technical University of Crete, GreeceReviewed by:
Carlos Noriega, Federal University of Pernambuco, BrazilMaría Isabel Delgado, National University of La Plata, Argentina
Copyright © 2026 Silva, Queiroz, Oliveira, Araujo Lopes, Pereira, Nóbrega, Barcellos, Rezende and Ferreira. This is an open-access article distributed under the terms of the Creative Commons Attribution License (CC BY). The use, distribution or reproduction in other forums is permitted, provided the original author(s) and the copyright owner(s) are credited and that the original publication in this journal is cited, in accordance with accepted academic practice. No use, distribution or reproduction is permitted which does not comply with these terms.
*Correspondence: Tiago O. Ferreira, dG9mZXJyZWlyYUB1c3AuYnI=