Beach landscape Dataset of Fernando de Noronha Island (Brazil)

Beach landscape Dataset of Fernando de Noronha Island (Brazil), using a checklist with 26 physical and human parameters. Fernando de Noronha beaches were divided into sectors according to the landscape diversity. In total, 19 sectors were evaluated based on observations done during walks in the area, observations from viewpoints, with remote data. The evaluations were performed during fieldwork from 2014 (summer) and 2016 (spring). The landscape quality evaluation of Fernando de Noronha was performed using the Coastal Scenery Evaluation System. This method converts qualitative-quantitative data in quantitative data by estimating weights for 26 parameters (18 physical parameters – P - and 8 human-related parameters – H). The main parameters that define the landscape quality are classified from 1 (absence/bad quality) to 5 (presence/excellent quality). A mathematical model based on fuzzy logic was utilized to integrate the parameters weights in a special system for the scenarios classifications resulting in a value named D. The D-value is the indicator of the attractiveness of the evaluated place. The beaches are divided into classes ranging from 1 (extremely attractive natural site) to 5 (unattractive urban areas).


a b s t r a c t
Beach landscape Dataset of Fernando de Noronha Island (Brazil), using a checklist with 26 physical and human parameters. Fernando de Noronha beaches were divided into sectors according to the landscape diversity. In total, 19 sectors were evaluated based on observations done during walks in the area, observations from viewpoints, with remote data. The evaluations were performed during fieldwork from 2014 (summer) and 2016 (spring). The landscape quality evaluation of Fernando de Noronha was performed using the Coastal Scenery Evaluation System. This method converts qualitative-quantitative data in quantitative data by estimating weights for 26 parameters (18 physical parameters -Pand 8 human-related parameters -H). The main parameters that define the landscape quality are classified from 1 (absence/bad quality) to 5 (presence/excellent quality). A mathematical model based on fuzzy logic was utilized to integrate the parameters weights in a special system for the scenarios classifications resulting in a value named D. The D-value is the indicator of the attractiveness of the evaluated place. The beaches are divided into classes ranging from 1 (extremely attractive natural site) to 5 (unattractive urban areas

Value of the data
• The data can be useful for coastal management in island and protected areas.
• Any researcher that deals with the Fernando de Noronha Island can benefit from these datasets. In addition, researchers from other fields of knowledge, decision makers and public officials. • The data can be useful for further researches that deal with any coastal scenarios or coastal management research. • Beach landscape Dataset create new perspectives for sustainable development based on the singularities of this touristic resource -the landscape. • The data investigate extensively the effect of tourism on changes in the coastal scenery, including natural and anthropogenic parameters.

Data Description
The dataset consists in the classification of Fernando de Noronha's beaches in classes (1-4), according to the Coastal Scenarios. The classification is in Fig. 1 . The highest diversity of classes is in the APA/Mar de Dentro area (Fernando de Noronha Environmental Protection Area-APA). On the other hand, in the PARNAMAR/Mar de Fora area (Fernando de Noronha National Marine Park-PARNAMAR), there is only one beach in class 3 (mostly natural areas, with some landscape parameters that stands out), and all the other beaches are in classes 1 (extremely attractive natural sites) and 2 (natural, attractive areas with high landscape value sites) ( Fig. 1 and Table 1 ).   The D value (attractiveness indicator) is presented for summer and spring periods ( Table 1 ). Histograms provide a visual summary of the physical and human parameters obtained through the application of the checklist and are useful for immediate evaluation of high and low ranking attributes. The values defined for each parameter (physical and human) are shown in Fig. 2 Table 1 ) ( Fig. 3 ). Attribute values 4 or 5 produces a high scenic value (high rating). The predominance of values 4 and 5 in phisical and anthropogenic parameters in the histogram can be observed for Leão Beach and Sancho Bay ( Fig. 3 ). On the other hand, attribute values 1 or 2 produces a high scenic   In this assessment, it is generally the human parameters that reduce the assessment, such as Porto/Pier Beach ( Fig. 2 ).

Experimental Design, Materials, and Methods
The methodology used to perform the beach landscape quality evaluation of Fernando de Noronha island beaches (Brazil) was the Coastal Scenery Evaluation System [1][2][3] . This method consists of estimating weights for 26 parameters (18 physical parameters -P -and 8 humanrelated parameters -H) and converting qualitative-quantitative data in quantitative data.  cording to the Evaluation System, the 26 parameters were considered essential for an attractive coastal landscape and are shown in Table 1 . The parameters are weighted from 1 to 5, where: 1 refer to the item's "absence or bad quality" and 5 refers to the item's "presence or excellent quality". For the evaluation of Fernando de Noronha island beaches, the beaches were divided in 19 homogeneous landscape sectors, (shown in Fig. 1 ). Some beaches, due to its and landscape variation (heterogeneity), were divided into more than one sector (e.g. Atalaia/rocky beach and Atalaia/sandy beach; Porto/Pier Beach and Porto/Natural Beach). From the total beach sectors, 10 of them are inside the APA area and 9 of them are inside the PARNAMAR area ( Fig. 1 ).
The evaluation of each beach sector and checklist filling was performed (i) in fieldworkusing landscape observations during walks in the beaches and/or from viewpoints, and (ii) in data remote-check using Google Earth imagery. Due to seasonal variability, landscape evaluations were performed during spring and during summer -summer fieldwork performed in 2014 and spring fieldwork performed in 2016. The professionals involved in the field evaluations were from the areas of biosciences and geosciences/geography.
After each parameter's evaluation (checklist), data processing was performed, in order to integrate the parameters weights in a special system for the scenarios classifications. A graphical summary of the investigated sceneries were obtained/generated from the weighted averages and association degrees and histograms [4] . Beach scenery is better when most of the parameters scores "5" (which result in a right-leaning association degree curve), and in the same way the potential status of the scenic assessment are indicated from the weighted average -the more parameters scoring "5", the better the coastal scenery. For this integration, a mathematical model based on fuzzy logic is used and the result obtained from this model is a value named D (D-value), which is the indicator of the attractiveness of the evaluated beach. According to the Method [1] , there are five possible beach classes, according to the D-value obtained ( Table 2 ).
Class 1 beaches are extremely attractive natural sites, with a D-value > 0.85. Class 2 beaches are natural, attractive areas with high landscaping value site and a D-value between 0.65 and 085. Class 3 beaches are mostly natural areas with some landscaping value highlighted, and D-value between 0.65 and 0.40. Class 4 beaches are urban areas, mainly unattractive, with few landscaping values highlighted, and D-value between zero and 0.40. Class 5 beaches are unattractive urban areas, with intense development and low landscaping value, and D-value below zero.
The data obtained in this study are available as georreferenced files (.kmz)