Trends in aquatic ecology research associated with urbanization evolution during three decades in Santa Catarina Island/SC

This paper systematically assessed trends of aquatic ecology publications related to the urbanization growth during three decades (1990-2010) in Santa Catarina Island. The study was carried out in fifteen watersheds of Santa Catarina Island, located in Florianópolis city, Southern Brazil. Geoprocessing and scientometry tools were used to analyze the evolution of urban areas and publications, respectively, related to these watersheds during three decades. The QGIS software was used to delimitate the catchment area of each watershed and contrast those areas with the built-up area for each decade, therefore obtaining the percentage of area covered with buildings and impervious surfaces. A gradient was created allowing the classification of the watersheds related to the urbanized area percentage, resulting in 5 groups of urbanization (0-5%; 5-10%; 10-20%; 20-30%; >30% of urbanization). Aquatic ecology publications were obtained from several scientific and academic databases and used in the scientometric analysis. The number of publications by decade, document type, knowledge area and watershed of study were recorded. Growth of urbanized areas was mainly evidenced since the 2000s and showed a moderate positive relationship with the number of publications in ecology of aquatic systems in the Santa Catarina Island during the same period. However, differences between watersheds were observed, since the most urbanized watersheds were not necessarily the ones with the largest number of publications. Urban gradient evidenced in Santa Catarina Island offers an opportunity to investigate the impacts of human disturbance on aquatic ecology, water quality and stream communities. Furthermore, the low number of research projects might be contributing to the decline in biological integrity due to the lack of knowledge for evaluation and management in these environments.

the dynamics of urban environments themselves as ecosystems (Grimm et al., 2000).
Santa Catarina Island is localized in a subtropical climate area and has been a key touristic destination in Brazil (Wemple et al., 2017). Consequently, there has been observed an increased urbanization and population growth in the last decade (Guerra et al., 2016). On the island, the aquatic ecosystems are mainly formed by lagoons, rivers and streams usually of short extension (Pêgas Filho & Tirloni, 2009). The continued development of urban ecosystems on the island has the potential to change the integrity of the aquatic ecosystem, where the understanding of the integrative context will be important for water resources management.
Scientometric tools have been used to measure scientific progress, and are a common research instrument for systematic analysis (Van Raan, 2005). The results could provide a better understanding of the local trend in aquatic ecosystem research and, potentially, give some guidance for future studies and assist policy-makers for water management.
Hence, this paper aims to assess, systematically, the trends of aquatic ecology publications associated with the urbanization evolution during three decades (1990, 2000 and 2010) in Santa Catarina Island, using geoprocessing and scientometry tools.

Introduction
Freshwater systems on subtropical islands are undergoing substantial alteration as urbanization increases, and these watersheds become different from native stream communities (Brasher, 2003). Human population continues to move to cities and managing water quality and sanitation is probably the biggest challenge of this century (Kaushal et al., 2010).
Changes in populations and associated socioeconomic factors drive the intensity, extent, and stage of urban development, and may affect freshwater ecosystems on a temporal-spatial scale (Parr et al., 2016). Uncontrolled urbanization has the potential to affect ecosystem functions of aquatic ecosystems by altering the hydrologic regime (runoff), habitat loss (riparian cover removal), and physicochemical characteristics (inputs of nutrients and pollutants) (Walsh et al., 2005;Parr et al., 2016), threatening the environmental health (Cooper et al., 2013;Liao & Huang, 2013) Effective water quality management of streams in urbanized watersheds requires the identification of the most important contributors to changes in the environment, such as sources of pollutants (Hatt et al., 2004). Thereby, urbanization has led to increased research on ecology in urban ecology in the last decades worldwide (Smucker & Detenbeck, 2014;Walsh et al., 2005).
the area established as permanent preservation areas (IPUF, 2004). However, since the 20th century, urbanization contributed to the development of the city, and it has attracted tourists, resulting in even more growth (Wemple et al., 2017). In 2010s, almost 100% of the urbanizable areas of Santa Catarina Island -in other words, areas with no physical limitations (mainly mountain chains)were occupied (IBAM, 2015;Santiago et al., 2014).
Here we studied the fifteen major watersheds of Santa Catarina Island (Figure 1).

Geoprocessing
An urbanization gradient was used, which enabled the investigation of urbanization evolution over three decades . Firstly, a Digital Elevation Model (DEM) with 30m of spatial resolution from the Santa Catarina State data repository (SIGSC, 2017) was accessed. To delimitate the drainage area of each watershed, the algorithm r.watershed (GRASS GIS 7), was implemented in the QGIS software (QGIS, 2019). Afterwards, the watershed perimeter with the built-up areas for each decade (1993, 2003 and 2013) was clipped to obtain the percentage of area covered by buildings. The present built-up areas (scale: 1:50.000) were provided by the Institute of Urban Planning of Florianópolis -IPUF. Thus, a gradient was created to classify the watersheds according to urbanized area percentage: beige (0-5% of urbanization), yellow (5-10%), orange (10-20%), red (20-30%) and dark red (>30%) (Figure 1).

Scientometrics
To evaluate the publications on aquatic systems in Santa Catarina Island, we have applied a workflow, as suggested by PRISMA (2020) ( Figure 2). The used databases were obtained from the Institute for Scientific Information (ISI), the Web of Science database, the Science Citation Index (SCI), the Google Scholar and the Institutional Repository of Federal University of Santa Catarina (DSpace). The keywords used for retrieving data were "aquatic ecosystem" or "water ecosystem" or "hydrology", or "ecology", or "limnology" and the watershed's name. These terms yielded many non- relevant studies, being necessary an inspection to verify whether they really addressed aspects related to aquatic ecology. Searches were restricted to the period from 1979 to 2018, and were finalized in May 2019.
Afterwards, the publication outputs, document type, number of publications by decade, and studied watershed information were noted from the selected studies. The information was then grouped by knowledge areas and stored in a database of technical (congress abstracts and technical reports) and scientific (articles, doctoral thesis and master's dissertations, completion of course works) documents related to Florianópolis watersheds. From the 212 publications registered, 181 were carried out for a single watershed, while the others addressed two or more watersheds. The studies investigating more than one watershed were computed with repetition to identify the exact number of researches for each watershed individually.
We defined seven knowledge areas within aquatic ecology publications as follows: zoology, land use and land cover, hydrology, geology, water quality, socio-environmental and botany.
We used a non-parametric Kruskal-Wallis H test to compare variations of the urbanization percent and numbers of aquatic ecology studies among the three decades previously described. Statistical significance was calculated using Chi-square test (χ 2 ) and set p-value < 0.05. When a significant difference was detected, a pairwise test for multiple comparisons of ranked data (Conover's post-hoc test) with a Bonferroni correction was performed in order to explore these differences between decades using the PMCMRplus package (Dinno, 2017).
Spearman's Correlation Analysis was performed using all studied watersheds, in order to evaluate the correlation of total urbanization increase with aquatic ecology studies. Besides, we analyzed the correlation between the urbanized areas of each watershed and the number of publications and watershed drainage area and the number of publications. Statistical significance was calculated using the Chi-square test (χ 2 ) and set p-value < 0.05. For this analysis, we used the package 'pspearman' (Savicky, 2015). We used nonparametric statistics after no normal distribution was found (p-value > 0.05; Shapiro Test). All statistical analyses were performed using the software R Studio version 1.1.383 (R Development Core Team, 2015).

Gradient of urbanization
The greatest urbanization boom occurred between the 1990s and the 2000s. The Santa Catarina Island showed a total urban area of 8% in 1993, increasing to 14% in 2003 and reaching 16% in 2013 (Figure 3).
Although total urbanization has increased two-fold in three decades, the watersheds showed a different urban growth. However, there were no significant differences in urbanization increase of the watersheds between the three decades (χ2 (df=2 N=45) =5.11, p-value = 0.07). Over the three decades, the watersheds that showed the greatest growth in urbanized areas were Saco Grande, Rio Tavares, Morro das Pedras, Santo Antônio de Lisboa, Ingleses, Costeira, Cacupé, and Itacorubi (Table 1).
Watersheds that showed in the 90s about 6 and 10% of urbanized area, such as Saco Grande, Rio Tavares, and Morro das Pedras, revealed a strong increase in the following decades. Morro das Pedras watershed showed the highest degree of urbanization in Santa Catarina Island in the 2010s, exceeding 41% of the total area (Figure 1). Although it is the smallest watershed on the Island, almost 50% of the total area is urbanized.
In the 1990s, Santo Antônio de Lisboa, Ingleses, Costeira and Cacupé already had considerable urban areas between 11 and 20% of the total area, which increased in the subsequent decades. Cacupé showed the most expressive growth, reaching 35% of urbanized area in the 2010s. Itacorubi was the most urbanized watershed during the 1990s, and in the following decades, it was considered the second most urbanized watershed of Santa Catarina Island (Figure 1).

Scientometrics
Researches on aquatic ecology in Santa Catarina Island watersheds began in 1979, with a technical report about floristics in nine watersheds. Our results showed 212 publications, demonstrating an increasing rate in the last three decades (Figure 4).
During the period from 1979 to 1999, 29 publications were found. The fast-growing period started in the second decade (2000s), when the number of publications increased more than two-fold, reaching 77 publications. The same pattern was observed in the third decade analyzed, with 106 publications. The growth in the number of publications showed significant differences between the three decades evaluated (χ 2 (df=2, N=45) = 9.21, p-value = 0.009). Where the first decade differs from the two subsequent decades (p-value=0.015 and p-value= 0.022, respectively).
Despite the number of publications having increased in the second decade, the number of articles was similar (Dec1 = 15, Dec2 = 14). In the same period, there was a substantial increase in academic research, such as dissertations, CCW, congress abstracts and thesis (Figure 4). On the other hand, the publication of articles increased in the third decade, reaching up to three times more than the previous decades (Dec3 = 51).  Five watersheds have shown the largest number of publications: Lagoa do Peri (n = 66), Itacorubi (n = 53), Lagoa da Conceição (n = 46), Ratones (n = 36) and Rio Tavares (n = 31). The other watersheds showed less than fifteen (15) publications over the selected three decades (Table 1).
Hydrology was the only subject not studied in all decades, with no registered studies in the 1990s. Land use and land cover, water quality, and socioenvironmental comprised about 21.8%, 20.9%, and 19.9% of the total studies, respectively. Furthermore, there was an increase in researches in these knowledge areas in the last two decades, mainly in the 2000s. The third decade also showed an increase in zoology and hydrology studies ( Figure 5).
Studies on land use and socio-environmental are present in all watersheds, but the more urbanized and more studied ones also presented a greater diversity of subjects explored (Figure 6).  Concerning the total urbanized area of Santa Catarina Island and total number of publications, Spearman's correlation indicated that there was a significant positive association between urbanization and the number of publications in the study area, (r s (29) = 0.56, p-value = 0.0008). However, the correlation was not significant when analyzed the urbanized areas for each watershed (r s (13) =-0.14, p-value = 0.06), that is, more urbanized watersheds were not necessarily the ones showing the largest number of publications. On the other hand, the watershed drainage areas were positively correlated with the number of publications (r s (13) = 0.63, p-value = 0.01).

Discussion
Urbanization evolution in Santa Catarina Island, mainly after the 2000s, showed a positive correlation with the number of publications on aquatic ecology, when it was analyzed the total area urbanized with the total number of publications. However, most urbanized watersheds were not necessarily the ones with the largest number of publications, while larger watersheds showed higher amount of publications.
Urban expansion in Santa Catarina Island began in the late 1960s (Campos, 2010) with the implementation of highways and the Federal University of Santa Catarina (IBAM, 2015). In addition, the 98 km of coast and beaches are an important tourist spot, boosting the development of new roads and resulting in subsequent urbanization (Silva et al., 1996) and a population increase of 23% in the last decade (Guerra et al., 2016). Our results showed this fast urbanization taking place between the 1990s and the 2000s, with almost twice as much of built area, mainly on the coastal area, where the occupation is concentrated (Figure 3).
We observed, in the same decades, a strong increase in the number of publications on aquatic ecology related to Santa Catarina Island, associated with the increase of urbanization. Indeed, land cover change has adversely affected aquatic ecosystems (Walsh et al., 2005;Grimm et al., 2008) and concerned scientific community, becoming an important research field, mainly after the 21st century (Liao & Huang, 2013).
Our results suggest that urbanization may be one of the drivers for the increasing publications on aquatic ecology. Currently, urban streams are prevalent throughout the world, and the altered ecosystem structure, with a subsequent reduction in ecosystem services, has led to an exponential increase in the number of research projects regarding urban ecosystems (Smucker & Detenbeck, 2014).
Aquatic ecosystems have gained more attention from researchers around the world and became one of the most dynamic fields within ecology (Kong et al., 2002;Liao & Huang, 2013). Thus, the progressive increase in the number of publications during the three decades evaluated may have led to a greater diversity of document types over time. The increase in the number of articles at the expense of the smaller number of thesis and dissertations accompanies an international trend of consolidating regional knowledge (Almeida & Guimarães, 2013). Also, to speed up the writing and evaluation of the thesis, institutions and graduate programs in several countries, including Brazil, have chosen to replace the writing of the thesis chapters with articles.
The sub-areas found in this study were also explored in several studies of urban ecology (Karaer & Küçükballi, 2006;Cooper et al., 2013) because they are directly affected by increased urbanization, which reinforces our findings. Indeed, studies have shown that the main causes of contamination in the Brazilian coastal zone are related to population density (Braga et al., 2006;Pagliosa et al., 2005). Others studies have shown that, despite the variety of watershed characteristics and urban development, efforts in the last two decades have been focused primarily on describing and analyzing commonalities among urban streams (Paul & Meyer, 2001;Walsh et al., 2005;Wenger et al., 2009).
Although there is a positive relationship between total urban growth and the number of aquatic ecology studies in Santa Catarina Island, the fifteen studied watersheds have not shown a proportional increase of publications in agreement with the urban gradient (Figure 1). In other words, the most urbanized watersheds were not the same watersheds that presented the largest number of publications, evidenced by the lack of a significant correlation (p > 0,05).
Nevertheless, a positive correlation between the number of publications and drainage areas (watersheds size), shows that the largest watersheds are also the most studied. The watersheds with the largest number of publications were Lagoa do Peri, Itacorubi, Ratones, Lagoa da Conceição and Rio Tavares.
The high level of economic importance of these watersheds associated with the presence of large research groups from the universities of Santa Catarina in these locations (UFSC and UDESC) probably has contributed to the increase in the number of studies. It is known that water quality and good ecological status within watersheds are important for the economic activities developed, as supplying drinking water (Lagoa do Peri watershed), recreation (Lagoa do Peri, Lagoa da Conceição, Itacorubi watershed) and for the maintenance of the mangrove ecosystem (Itacorubi, Ratones, and Rio Tavares) (Silva, 2019; Barbosa & José, 1998;Torres et al., 2002).
On the other hand, watersheds which have shown less than ten (10) publications over the three decades exhibited the highest gradient of urbanization (Figure 1), except Lagoinha do Leste, Saquinho and Ribeirão da Ilha, which showed their drainage areas free of urbanization or a low-level of urban areas. It is possible that the low gradient of urbanization in these watersheds may be due to their smaller drainage areas (Table 1), constituted in a mountain chain of a high slope, corroborating with Pegas-Filho & Tirloni (2009). Nevertheless, even in relatively pristine watersheds, alterations on streams caused by urban activities may result in decreased flow velocity and water depth, reducing habitat availability (Brasher, 2003).
Additionally, the lower number of publications associated with the higher values of the urbanized area in Costeira, Santo Antônio and Ingleses watersheds (between 22 and 29%) and Cacupé and Morro das Pedras watersheds (more than 30%) suggest that these streams may have their water quality and integrity compromised.
So, it is necessary to consider two reasons for the lower number of publications; first, low investment in research in the emergent and developing countries (Andrade et al., 2010), leaving gaps in the current understanding of the ecological systems. Second, most of the researchers have their fields of study focusing on one aspect or another, such as research groups that orientate their studies to a specific location (Liao & Huang, 2013). Thus, the low number of studies could be contributing to the lack of knowledge of the water quality and availability, which is a key element for the conservation of water resources and subsequent management of these environments.
In recent years, some studies have focused on urban streams, but many questions about the mechanisms driving important changes remain unanswered (Wenger et al., 2009). Santa Catarina Island, with watersheds ranging from the relatively pristine to the highly degraded, may offer an opportunity to examine the impacts of human disturbance on water quality and native stream communities. Those coastal areas -that are sensitive to the modifications due to urbanization -are understudied and this may lead to the degradation of these resources, damaging the local economy. Understanding the relationships between habitat alteration and aquatic systems structure is critical for developing management strategies.