Assessment of conflicts between mangroves and human occupation in Subaé river outfall between the years 1988 to 2017.

)


Introduction
Coastal zones fulfill important ecological, social and economic functions, occupied by marine engineering works, ports, navigable canals, dredging and artificial landfills (PRIMAVERA, et al., 2019). They are affected by the damming of rivers, facilities for leisure areas, tourism, and urbanization, among other interventions. In this sense, geotechnologies appear as a possible option for monitoring and research in these areas, both in operational and cost terms (MOURA and CANDEIAS, 2011). Geotechnologies Provides efficient and agile ways to characterize land use. This method eases the acquisition and manipulation of data, allows repetition in the collection of information, and constant updates of the analyzed areas (MAURYA et al., 2021). Thus, this research sought to determine spatio-temporal changes. Demonstrating loss and gain relationships between classes in the low Subaé, with an emphasis on the areas of mangrove and urban occupation.

Methodology
The study area is located in Baixo Subaé, located in the Northern Reconcavo of the state of Bahia (Figure 1).

Figure 1 -Location of the study area
Satellite images from the Landsat Series (30m) were used for temporal evaluation of the mangrove areas. And, an image from the PlanetScope satellite (3m) was used to determine the accuracy of the classifications.
To allow comparison between maps with different spatial resolutions, we created a vector file with 5000 points of random location in the map area. This vector received the classes values in both images. We arrange the data in a Confusion Matrix. And, calculated the Tau and Geographic Simultaneity (GS) (SILVA et al., 2017) indices, which were classified according to Silva (2018). The methodological steps are shown in Figure 2.

Results and discussion
The results of the classifications are shown in Table 1 Table 1-Classes of Use and coverage for the years studied (Km²).   (Table 3). It can be seen that classes 1,2,6 and 7 obtained an Excellent and Almost Perfect classification for the Geographic Simultaneity and Tau indices, respectively. Class 5 was Almost Perfect for Tau, but it only Very Good for GS. Class 3 and 4 were Very Good (SG) and Substantial (Tau). Table 3 -Concordance indices.

Satellite
The following cross-table shows the changes that occurred (in pixels) between 1988 and 2003 (Table 4). The processing of changes between the periods from 1988 to 2003, allowed us to infer that there was a significant increase in using pasture in the same order as the reduction in agriculture (around 30 km²), and the tree vegetation that had a reduction of 15km², approximately ( Figure 3). These results are compatible with the report of the SEI (2012) which highlighted the state of natural vegetation in this region, which was practically devastated because of the rise to productive activities. Pasture, agriculture and forestry activities were the main responsible for the suppression of natural vegetation    Observing the behavior of mangroves, it was observed a 3 km 2 growth over areas that were previously bare soil. A priori, this growth occurred in an area close to the urban area of Santo Amaro (Figures 6 and 7B).
Originally, mangroves occupied this area, but between 1973 and 1976, this area was devastated. There is a few data regarding the causes mortality in the literature, probably because of the military dictatorship in the period when this happened. And it is worth mentioning that factors such as sedimentation rate, sediment subsidence, fluvial discharge, tides, and changes in sea level, have a direct influence on the growth and survival

Figure 7 -Losses and gains Urban class 1988-2003 (A) and losses and gains Mangrove 1988-2003 (B).
Analyzing the data on the temporal variation for each class between the periods of 2003 and 2017 (Table 5), there were no transitions between the agriculture class for the urban classes and water bodies, as well as there was no assignment of land from the water bodies class to the agriculture class. In Figure 8, it is possible to observe that there was a great variation in losses and gains in pasture class. This class has gone through losses and gains greater than 20 km². On the other hand, the classes Mangrove, Urban and Water Body showed only small variations. The greatest growth is related to the class Arboreal vegetation.
On the other hand, the class that lost the most area was Pasture, which may be related to the implantation of  Evaluating only the variation in area (Figure 9), it can be seen that the class with the greatest loss of area is that of exposed soil, followed by Pasture and Agriculture, while the Water and Urban classes, obtained a small area gain. This variation between classes may be seasonal. Images from December/2003 and June/2017, indicates summer and winter, respectively. Another factor is the temporary peanut, bean, and corn crops grown in Santo Amaro (BAHIA, 2011).
The mangrove class did not show significant variation, considering that the gains and losses were of the same order. This effect emphasizes the importance of observing the geographic location of these changes, since the losses and gains cancel each other out in the area variations. 10 and 12A). This behavior may be related to difficulties in discerning the exposed soil class and the urban area. Most of the growth occurs on areas of exposed soil and pasture, followed by tree vegetation, agriculture and mangroves, on a smaller scale. Like the urban class, there are losses of the mangrove class to water (Figures 11 and 12B). Observing the tide tables, we noted that despite the high tide on both days, the tide height was 1.    Table 6 shows the area values for class transitions during the period between 1988 and 2017. Analyzing the losses and gains for the complete temporal extension of the study (Figure 13), we found the major changes in area for the Pasture, Agriculture, and Vegetation classes, with transitions of 40km², 50Km² and 50Km², respectively. When considering the total variation as the sum between losses and gains, there is also growth, with insignificant losses, in water and urban classes. On the other hand, the agriculture class had the lowest growth at the expense of a large loss of area.  Recognizing the difference between losses and gains, it is possible to see a significant reduction in agricultural areas. This reduction corresponds to 15% of the study area, approximately. In the same way, but with less intensity, there is a reduction in exposed soil areas.
The growth of pasture from 24km² to 48Km² is also notable, being the second largest class found in the area.
The most abundant class is Arboreal Vegetation, mainly because it makes no difference between the Atlantic Forest, Secondary Forest, or Silviculture. In this case, it is possible to explain that the growth of this class was probably driven by the production of eucalyptus (JESUS, 2018) ( Figure 14). Considering the variations that occurred in the Urban Zone during this interstice (Figures 15 and 18A), there is a notable growth in area gain over all classes, except water bodies. The most affected classes, for this growth, were Pasture, Agriculture, and Urban Vegetation.   Another significant changes in the mangrove class is the growth over the Exposed Soil. Alike the transition to 2003, by the recovery of the area previously deforested in Santo Amaro da Purificação (Figures 17 and 18B).

Conclusions
Calculating the general Tau index for the classifications was 0.89, considered to be almost perfect. Class 1, 2, 6, and 7 obtained an Excellent and Almost Perfect classification for the Geographic Simultaneity and Tau indices, respectively. Class 5 was also considered to be Almost Perfect for Tau, but rated as Very Good for SG.
Classes 3 and 4 were considered Very Good (SG) and Substantial (Tau).

Assessment of conflicts between mangroves and human occupation in Subaé river outfall between the years 1988 to 2017.
International Journal for Innovation Education and Research Vol. 10 No. 11 (2022), pg. 103 Carrying out the spatio-temporal analysis observing only the variation of areas may not represent the phenomena that occurred. The result for a class that did not vary over time may indicate two things. First, that there were no changes in this local. Second, it may indicate that a substantial increase was suppressed by a loss of equal magnitude.
During the period 1988-2017 there was a significant reduction in agricultural areas, corresponding to approximately 15% of the study area. The pasture expanded from 24km² to 48Km² and urban area growth occurred across all classes, including mangroves. The mangroves expansion over areas that were previously bare soil, is related to the incident between 1973, and 1976.

Acknowledgment
This