The Guamá river basin is located in the northeast of the state of Pará and covers an area of about 11,870 km², representing 1% of the area of the state (Figure 1).

Figure 1
Location of the study area.
Source: Hydroenvironmental Studies and Modeling Lab (LEMHA), 2018. Org.: The author, 2018.

The river basin is inserted in the Hydrographic Region of the Atlantic Coast - Northeast, according to Resolution 04/2008 of the Water Resources Council of the state of Pará. The Guamá river basin is a large area formed by 19 municipalities with intense economic activity centered in industry, mining, commerce and agriculture. They are: Capitão Poço, Garrafão do Norte, Irituia, São Miguel do Guamá, Bujarú, Santa Luzia do Pará, Concórdia do Pará, Santa Izabel do Pará, Inhangapi, Castanhal, São Domingos do Capim, Ourém, Mãe do Rio, Belém, Acará, Bonito, Benevides, Marituba and Ananindeua, with a total of nearly 2,700,000 inhabitants according to the last census (BRASIL, 2010).

The calculation of the Water Sustainability Index (WSI) for the Guamá river basin was an adaptation (for the condition of State) of the methodology applied by Chaves and Alipaz (2007), which takes into account hydrological (H), environmental (E), social (S) and management (M) aspects, because the sustainability of a basin is a dynamic process.

For the application of the index in large basins such as Guamá, it is more feasible to divide the basin into sub-basins and calculate its total value in a weighted way.

Thus, the Guamá river basin was divided into eight sub-basins for analysis, namely: Lower Guamá, Apeú Stream, Bujarú River, Middle Guamá west sector, Middle Guamá east sector, Mãe do Rio Stream, Sujo River, and Upper Guamá, delimited according to the river flow system and considering the main river and its tributaries.

To analyze the environmental variable (E), a land use and land cover map was created, defining eight classes adapted according to the IBGE Land Use Manual (BRASIL, 2013): vegetated area, agriculture, occupation areas, uncovered soil, pasture, water bodies, unobserved areas, and others.

The map was generated through the supervised classification method for multispectral images based on the Maximum Likelihood (MaxVer) classifier, performed pixel by pixel, using the ENVI software in orthorectified satellite images of the RapidEye system (high resolution, in a total of 43 scenes, dated 6/29/2011, 7/28/2011, 8/4/2011, 10/23/2011, 7/31/2012, 8/2/2012, 9/13/2012, 10/24 / 2012, 01/08/2013, 04/09/2013, 17/08/2014 and 25/11/2014, which contain five spectral bands, obtained through the Geo Catalog of the Ministry of the Environment (MMA); the selected scenes presented better conditions related to the lower cloudiness index, which favors the analysis and classification of targets), which allowed the analysis of the landscape and the quantification of areas that have forest cover (ratio between total area of each sub-basin and the forested area). Results were validated through field surveys.

Regarding the hydrological variable (H), the hydrological curve number - CN model was used. This model is designed to estimate the surface runoff dynamics in sub-basins, according to the analysis of land use and land cover types. Soil Conservation Service (SCS) procedures, adapted by Calzavara and Fernandez (2015) for Brazilian conditions, were adopted. Precipitation values were obtained from data collected by 11 rainfall stations located in the area of direct influence of the Guamá river basin (series from 1985 to 2015), available from the National Water Agency (ANA) Hydro 1.2 software. These values were spatialized according to the Thiessen Polygon method, which assigned a weighting factor to the precipitated totals in each rain gauge to their respective area of influence (CORREIA; RIBEIRO; BAPTISTA, 2015).

Information on social issues of the municipalities that compose the sub-basins was used in the analysis of the social variable (S), taking into account institutional, political and socioeconomic aspects linked to the population, that directly affect the management of water resources. They included the municipal human development index (MHDI), municipal Gini Index (BRASIL, 2010), water consumption index, and total water supply index per municipality according to the National Sanitation Information System - NSIS (BRASIL, 2016). These data were also investigated from information provided by municipal secretariats and the state government.

After obtaining the variables, arithmetic averages were used to calculate the social indicator for each municipality (Equation 1). This indicator contributed to compare and analyse the municipalities. Then, the weighted average was used so that the social indicators of each municipality were spatialized in their respective sub-basins, i.e. the area of each municipality within the sub-basins was considered (Equation 2). Thus, a social indicator was obtained for each sub-basin and used to determine the WSI, where: S_mi = social indicator per municipality; S_i = variables analyzed; n = number of variables; S = social indicator in the sub-basin; A_i = area of the municipality within the sub-basin (ha); A = total area of the sub-basin (ha).

Finally, the analysis of the management variable (M) assumed arithmetic and weighted averages, likewise in the previous variable, and referred to characteristics that influence the environmental management capacity of the municipalities that are part of the sub-basins (Table 1). They correspond to 13 attributes collected from the databases of IBGE, SNIS, Green Municipalities Program, and websites of the City Halls. Weights from 0 to 1 were assigned to each sub-basin, so that the management conditions of the municipalities could be analyzed and compared, where 0 indicates bad conditions and 1 good conditions.

To calculate the WSI from indicators that fit the evaluated aspects (H, E, S, M), each one of them was divided into five scale scores (0; 0.25; 0.50; 0.75 and 1) to simplify the estimate, where 0 represents the worst conditions and 1 the best conditions (Table 2).

Table 1
Parameters for the evaluation of municipal management.

Org.: The Author, 2018

Table 2
Variables for the application of the WSI.

Org.: The Author, 2018

The WSI corresponds to the sum of variables, according to each parameter considered, divided by the total number of variables adopted (Equation 3). All indicators have the same weight since none of the variables is considered more important than the others, but rather complementary for the sustainability of the basin. They were: hydrological ($p_{wH}$), environmental (_Pwe ), social ($p_{wS}$), and management (_Pwm ) aspects, rated from 0 to 1 (0 indicates poor conditions and 1 indicates optimal conditions). The expression ni represents the variables considered (Table 2). After obtaining the results, they were classified for each sub-basin according to Table 3. With the analysis of the resulting sustainability indices for each sub-basin, a weighted average in relation to the area was applied to obtain the WSI of the entire Guamá river basin.

Equally weighted analysis facilitates structuring the dimensions of the indicators, assuming that they have intrinsic characteristics with water management, are measurable, approachable and comparable, so that redundancy effects among the selected variables are reduced (CARVALHO et al., 2013). The analysis allows the integration of heterogeneous criteria and options simultaneously, as for example in the assessment of water sustainability. These factors are highlighted by Flores-Alsina et al. (2010) in the assessment of the water quality index and by Chuang et al. (2018) in the evaluation of hydroenvironmental indicators in coastal systems.

Table 3
Classification of WSI values.

Source: Adapted from Carvalho et al. (2011).

Complementarily, aiming at an integrated analysis, the results were presented in cartographic format and using cluster analysis. The clustering methodology is a multivariate statistical analysis that enables the identification of groups with homogeneous characteristics (k-means), considering mathematical calculations of proximity (similarity) to all pairs of objects and between each object and subgroups. such that the distances between the members of a subgroup are minimal and distances between subgroups are maximum (YOSHIMITANAKA et al., 2015). Results are presented as groupings and heatmaps, where each cell corresponds to the position occupied by the value of a given variable in a unit of analysis.

The analysis of the land use and occupation map allowed the identification of the percentage of areas with vegetation (Table 4), and this result identified which sub-basins had the best and worst representativeness in terms of environmental indicator.

It was observed that the sub-basins that presented the best conservation stage, with the occurrence of vegetated areas, were Lower Guamá (57.87%) and Bujarú River (55.51%). In turn, the ones with the largest presence of altered areas were Mãe do Rio Stream (24.07%) and Upper Guamá (29.08%).

It is noteworthy that the largest concentration of altered areas was associated with municipalities that have characteristics focused on agribusiness, with small and medium producers that meet the needs of the state and large producers that serve the rest of Brazil and the foreign market (REBELLO et al., 2011).

The results obtained are in agreement with what Watrin et al. (2009) highlight to the northeast region of Pará, with pasture being the dominant pattern of land use. The same was observed by Pereira et al. (2015) for the Apeú Stream basin, and by Nascimento and Fernandes (2017) for a sub-basin of Upper Guamá.

Table 4
Area corresponding to vegetation cover in the sub-basins analyzed.

Org.: The Author, 2018.

Vegetation cover is of fundamental importance for the functioning of several processes that occur in an ecosystem, such as infiltration or runoff processes, energy balance, maintenance of weather conditions, among others. Changes in vegetation cover caused by deforestation or implementation of agricultural activities can lead to several negative environmental impacts, such as river siltation, soil compaction, triggering of erosive processes, among others (SILVA et al., 2017).

Runoff values were obtained for each sub-basin analyzed through the Curve Number method (Table 5). The runoff of the sub-basins varied between 28 and 33% of the incident precipitation values, and the sub-basin that presented the highest result was the Bujarú River, and the one with the lowest value was the Upper Guamá.

The Guamá river basin is basically composed of five dominant soil types, namely, Yellow Latosols, Concretionary Latosols, Fluvic Latosols, Quartzarenic Neosols, and Red-Yellow Argisols. Yellow Latosols occupy most of the basin, 84.91% of the land.

Yellow Latosol is characterized as deep, porous and with high texture, as well as a moderate infiltration rate, resistance and tolerance to erosion, contributing to the median runoff values in all sub-basins. These conditions favor soil fertility, because the smaller the runoff, the smaller the drag of the finer particle size of the soil (FEITOSA et al., 2010).

The performance shown in Table 5 indicates that the best basin framework, according to the physical-water parameters, of dominant soil type is between Groups A and B, which are classified as (SARTORI et al., 2015): Group A - soils with low runoff and high infiltration, very deep (> 2.0m), with high permeability and low erodibility; Group B - less permeable soils with higher potential to generate runoff, consisting mainly of moderately deep to deep soils. These characteristics make it possible to understand the runoff dynamics in the basin. Consequently, the assessment of water yield potential is also important to the planning of measures for conservation of soil and water characteristics and reduction of siltation processes and risks of floods (MUÑOZ-ROBLES et al., 2011).

Table 5
Mean superficial runoff of sub-basins.

Org.: The author, 2018.

Parameters related to the municipalities of the Guamá river basin, such as the municipal human development index, the Gini index and two indices of the National Sanitation Information System related to water supply and consumption were used in the analysis of the social indicator. The weighted average values per municipality were the basis to determine the social indicators of the sub-basins, taking into account the areas of the municipalities in relation to their respective sub-basins (Table 6).

Table 6
Social indicators of sub-basins.

*There is no information regarding the index. Source: Brasil (2010); Brasil (2016). Org.: The author, 2018.

It was observed that the sub-basin that was most prominent in relation to this indicator was Mãe do Rio Stream, with a value of 0.66. This basin is composed by the municipalities of Capitão Poço, São Domingos do Capim, Irituia and Mãe do Rio, which stand out mainly in relation to the water supply.

The least prominent sub-basin was the Middle Guamá east sector, with a value of 0.48, composed by the municipalities of Bonito, Capitão Poço, Irituia, Ourém, Santa Luzia do Pará and São Miguel do Guamá, two of which did not present information about two of the indices analyzed (Bonito and São Miguel do Guamá).

The analysis of management of the municipalities was based on the mean values obtained by assigning weight to the 13 variables analyzed (Table 1) regarding the presence or degree of coverage of: environmental legislation, municipal council environment, environmental licensing, Environment Secretariat, Rural Environmental Registry (RER), Green Municipalities Program, environmental management via internet, environmental education in the municipality, municipal land-use planning, basic sanitation plan, sanitation policy, environmental management and Municipal Fund of the Environment. Based on these variables, a value was generated for the management indicator for each municipality and then weighted for the sub-basin area (Table 7).

Table 7
Management indicators of sub-basins.

Org.: The author, 2018.

The Guamá river basin is heterogeneously composed of 19 municipalities that present different percentages in area (contained in the basin) of different representativeness: Acará (0.80%); Ananindeua (0.15%); Belém (0.96%); Benevides (0.51%); Bonito (0.65%); Bujaru (7.96%); Capitão Poço (21.59%); Castanhal (3.82%); Concordia do Pará (5.40%); Garrafão do Norte (13.92%); Inhangapi (3.87%); Irituia (11.15%); Mãe do Rio (3.38%); Marituba (0.28%); Ourem (3.45%); Santa Izabel do Pará (4.78%); Santa Luzia do Pará (5.48%), São Domingos do Capim (3.83%); and São Miguel do Guamá (8%). Of these, 14 present their municipal headquarters within the basin area. As management indicators refer to the political actions in the municipality as a whole, it was decided to attribute equal weights because it is understood that, once consolidated, management actions will have consequences distributed throughout the territory.

The procedure adopted follows the premise discussed by Carvalho (2014) in stating that the integrity of watersheds must encompass a systemic understanding of sustainability, with water being understood as a whole in spatial relationships, with integrated environmental planning of watersheds associated with territorial planning.

The sub-basin with the highest value for this indicator was the Apeú stream, which comprises the municipalities of Castanhal and Santa Izabel do Pará that obtained values above 0.50, and Inhangapi had a low management indicator. The sub-basins of the Sujo River and the upper Guamá River had the lowest values. This is because both municipalities had indicators below 0.40, indicating that they need more immediate actions to strengthen environmental management.

Water sustainability was initially calculated per sub-basin, and the results were framed according to each indicator: hydrological, environmental, social and management (Figure 2). Figure 3 represents the treatment given to the indicator categories using the clustering methodology. Relationships observed spatially or by cluster indicate existing similarities and factors of greater intervention. The basin can be viewed in 3 major segments: Upper Guamá - Sujo river - Mãe do Rio stream; Middle Guamá - Apeú Stream; Bujarú River - Lower Guamá.

The social (S) and management (M) indicators are the most responsible for the fragilities observed, mainly pressing the Upper Guamá, Apeú Stream and Mãe do Rio Stream sub-basins. The WSI is the global average of the four indicators and this indicator analysis allowed obtaining the water sustainability indices of each sub-basin and their respective performances (Table 8). The analysis resulted in a value of 0.54.

Maynard et al. (2017) obtained a value of 0.66 for the Japaratuba river basin (SE), with the weakest indicators being the hydrological and management ones. Silva (2017) obtained a value of 0.53 for the Piranhas-Açu sub-basin (RN), also identifying that, in general, the indicators related to socioeconomic and management aspects make a considerable contribution to the maintenance of intermediate sustainability.

Figure 2
Water sustainability index based on environmental, hydrological, social and management indicators for the sub-basins: I- Lower Guamá; II- Apeú Stream; III- Bujarú River; IV- Middle Guamá west sector; V- Middle Guamá east sector; VI- Mãe do Rio Stream; VII- Rio Sujo; VIII- Upper Guamá.

Org.: The author, 2018.

Figure 3
(a) Result of analysis by hierarchical clustering. (b) Representation of the cluster analysis through color matrices - Heatmaps.
Org.: The author, 2018.

Table 8
Water sustainability indices of the sub-basins.

Org.: The author, 2018.

Juwana et al. (2012) presented a review of water sustainability indicators and highlighted the influence that the “management” component may have on the others, where unfavorable conditions in the “hydrological” and “environmental” axes may have their impacts reduced with better management conditions of water resources.

The values of the social indicator showed the need to seek alternatives that contribute to the improvement of the indices analyzed, to obtain circumstances that favor an ideal quality of life and assurance of quality basic services for the population. Environmental management is of fundamental importance for the organization of municipalities and, consequently, of sub-basins. Investing in actions focused on this criterion represents a beneficial strategy for the maintenance of sub-basins, in addition to strengthening and conserving the interdependent relationships that occur in their territories (MARTINS et al., 2010).

The global average obtained for the basin showed its situation in relation to water sustainability, urging for the establishment of water management programs capable of enhancing the scenario presented, through more responsible actions by management regulators and other sectors of society. Water sustainability suggests the conditions necessary for basins to have conditions to replenish resources, compatible with the existing demand, always aiming at a consumption that is equal to or lower than their recovery capacity (TAMASAUSKAS et al., 2016).

The indicators analyzed demonstrated that the process of use and occupation of the territory of the Guamá river basin has a high potential to interfere with water sustainability, since they exert a positive or negative influence on it. Where the evolutionary dynamics of a landscape are understood by its historical processes at different time scales and according to the predominant environmental characteristics, both aspects contribute to the establishment of a local distribution pattern (VIEIRA et al., 2007).

The evolutionary dynamics from the perspective of sustainability of the Guamá river basin has several historical landmarks, such as the construction of the Belém-Brasília highway (BR 010), responsible for the emergence of dozens of towns, villages and cities, particularly those in the NE region of the state (TAVARES, 2008). Given this context, land use forms have marked northeastern Pará by a process of de-characterization due to deforestation. The percentage of change due to anthropism in the region has reached almost 25%, concomitant to the growth of logistic services and natural resources exploitation services such as logging, slashand-burn agriculture, and livestock, forming a mosaic of different degrees of plant succession, agricultural crops and pasture areas (CORDEIRO et al., 2017).

Regarding the use of the Guamá river basin, it covers part of the municipalities of the Metropolitan Region of Belém, with the headwaters of its sub-basins having the axis of the BR-360, that connects Pará to Maranhão, and of the BR-010, that makes the connection with the southeast of the state, as water divisors. Thus, their urban centers have the ability to polarize and influence a significant number of smaller cities and articulate relationships of all kinds, with the presence of medium-sized urban centers that make up the metropolitan area (TRINDADE Jr, 2011).

Land use in the Guamá basin faces future prospects of water sustainability for the region, with a high degree of urbanization, concentration of economic activities accompanied by growing poverty and persistent social problems (access to health, education, housing and basic sanitation) necessary for a large portion of the population (SILVA; SILVA, 2008).