International Journal of Environmental Science and Technology https://doi.org/10.1007/s13762-024-05597-6 ORIGINAL PAPER Change of vegetation cover and land use of the Pómac forest historical sanctuary in northern Peru E. Vera1  · C. Cruz1  · E. Barboza2,3  · W. Salazar2  · J. Canta1  · E. Salazar4  · H. V. Vásquez4,5  · C. I. Arbizu4,6 Received: 11 October 2023 / Revised: 8 February 2024 / Accepted: 10 March 2024 © The Author(s) 2024 Abstract The dry forests of northern Peru, in the regions of Piura, Tumbes, Lambayeque, and La Libertad, have experienced signifi- cant changes as a result of deforestation and changes in land use, leading to the loss of biodiversity and resources. This work analyzed for the first time the changes in vegetation cover and land use of the Pómac Forest Historical Sanctuary (PFHS), located in the department of Lambayeque (northern Peru). The employed approach was the random forest algorithm and visually interpreted Landsat satellite images for the periods 2000–2002, 2002–2004, and 2004–2008. Gain and loss rates were computed for each period, and the recovery process was assessed using the normalized difference vegetation index (NDVI) and enhanced vegetation index (EVI). Results indicate an expansion of agricultural land during each period, result- ing in the deforestation of 102.6 hectares of dense dry forest and 739.9 hectares of open dry forest between 2000 and 2008. The degree of reforestation in the cleared areas was measured using the NDVI and EVI indices, revealing an improvement from 0.22 in NDVI in 2009 to 0.36 in 2022, and from 0.14 to 0.21 in EVI over the same period. This study is expected to pave the way for executing land management plans, as well as the use and conservation of natural resources in the PFHS in a sustainable manner. Keywords Remote sensing · Landsat · Agroforestry · Biodiversity · Protected natural areas Introduction Editorial responsibility: S.Hussain. The tropical dry forest covers 50% of the forested areas in * C. I. Arbizu Central America and 22% in South America (Quiroga et al. carlos.arbizu@untrm.edu.pe 2019). This ecosystem is characterized by receiving 80% of 1 Vista Florida Agricultural Experiment Station, National its precipitation during the stationary period of four months Institute of Agricultural Innovation (INIA), Lambayeque, each year (Maass and Burgos 2011; Espinosa et al. 2012). It Peru is limited by the availability of water resources, especially 2 Division of Supervision and Monitoring in Agricultural in South America, but it provides a significant number of Experiment Stations, National Institute of Agricultural ecosystem services (Miles et al. 2006), with one of the main Innovation (INIA), Lima, Peru ones being the removal of carbon dioxide ( CO2) from the 3 Present Address: Ceja de Selva Research Institute air (Álvarez-Dávila et al. 2017). However, land use change for Sustainable Development (INDES-CES), National University Toribio Rodríguez de Mendoza de Amazonas has made it one of the most threatened ecosystems (Renzo (UNTRM), Chachapoyas, Peru 2003), due to fragmentation and transformation, resulting 4 Division of Agrarian Technological Development, National in a 25% reduction of its original extent (Adán et al. 2014; Institute of Agricultural Innovation (INIA), Lima, Peru Stan and Sanchez-Azofeifa 2019), making it one of the most 5 Present Address: Faculty of Zootechnical Engineering, endangered biomes on the planet (Maass and Burgos 2011). Agribusiness and Biotechnology, National University In Latin America, dry forest ecosystems face significant Toribio Rodríguez de Mendoza de Amazonas (UNTRM), challenges due to deforestation and forest degradation (Jimé- Chachapoyas, Peru nez et al. 2019). Additionally, they are constantly disturbed 6 Present Address: Faculty of Engineering and Agricultural by human activities, particularly various agricultural prac- Sciences, National University Toribio Rodríguez de Mendoza tices (Alejandra and Romero 2016). In Peru, dry forests de Amazonas (UNTRM), Chachapoyas, Peru Vol.:(0123456789) International Journal of Environmental Science and Technology cover 3.11% of the national territory (MINAM 2022), but availability (Gevaert et al. 2015). Satellite sensor images they are facing deterioration primarily due to illegal human were used to map land cover and land use more accurately settlements and economic activities (SERNAP 2016a). In using the random forest algorithm (Zhao et al. 2019). This the northern coastal region of the country, in the depart- enabled the evaluation of forest resources and human settle- ment of Lambayeque, the Pómac Forest Historical Sanc- ments, discriminating soil conditions, vegetation types, and tuary (PFHS), declared a Protected Natural Area (PNA) their states, thus detecting and interpreting land use changes under D.S. N° 034-2001-AG, is not exempt from this issue. for biomes conservation (Stan and Sanchez-Azofeifa 2019). In 2001, there was an illegal human settlement in the area. To date, land use changes and vegetation recovery infor- The invaders cleared approximately 25% of the sanctuary's mation is extremely limited in Peru. The present study aims area for agricultural activities, as well as for building homes to analyze land use changes and vegetation recovery in the and roads to transport their products for commercialization. Pómac Forest Historical Sanctuary between 2001 and 2022, It is important to note that the agricultural activities were using Landsat images, the random forest classification algo- carried out using groundwater through the construction of rithm, and multitemporal analysis based on multi-spectral tube wells (Ramírez et al. 2020). In response to these events, indices. Additionally, it will visualize the process of natural in 2009, an order was issued to evict around 800 invaders. recovery after anthropogenic occupation. This study seeks After this incident, the sanctuary was granted legal security to be a valuable tool for improving the management of PFHS over its territory to ensure its integrity and facilitate physical and may be replicated in other ANPs in tropical dry forest sanitation and recovery efforts. (Source: https:// andina. pe/ ecosystems or similar environments in Peru. agenci a/n otici a-p olici a-r ealiz a-v uelo-r econo cimie nto-s obre- santu ario- histo rico- bosque-p omac-2 14550. aspx). Remote sensing is defined as a technique through which Materials and methods images are acquired, processed, and visually or digitally interpreted to extract information about the land cover and Study area land use in a specific area (Astudillo-Sánchez et al. 2020). It presents advantages over traditional methods due to its rapid The study encompasses a total area of 5887.4 hectares processing of large volumes of data, facilitating decision- located in the Pómac Forest Historical Sanctuary (PFHS), making (Cuello et al. 2015). Moreover, it is a cost-effective in the lower part of the La Leche river, Pitipo district, Ferre- alternative (Zhu et al. 2019). However, it is limited by factors ñafe province, Lambayeque department, Peru (Fig. 1), with such as spatial and tem-poral resolution, as well as images a maximum altitude of 240 m a s l between the geographic Fig. 1 Location of the study area in the Lambayeque region in northern Peru International Journal of Environmental Science and Technology coordinates 6°25′55.49′′ and 6°32′51.84′′ South Latitude, of 30 × 30 m. for the elaboration of land use and vegetation 79°44′15.75′′ and 79°49′10.92′′ West Longitude. The mul- cover maps. titemporal analysis was carried out considering the years The selection of remote sensing images was made consid- 2000, 2002, 2004, and 2008, presenting minimum tem- ering the following criteria, as reported by Chuvieco (1996): peratures of 11.5 °C (July and August), and maximum of (i) Maximum 10% cloudiness and shadow, (ii) absence of 33.1 °C (December–May); average rainfall of 1078 mm, distortions, and (iii) belong to the same time of the year. In concentrated in the months of March and April. It should be addition, to define the study area, we used: (i) the limits of noted that the maximum values of rainfall are recorded in the National System of Natural Areas Protected by the State years that El Niño-Southern Oscillation (ENSO) occurred (SERNANP for its acronym in Spanish), and (ii) the Digital (SERNAP 2016a, b). Elevation Model (DEM) with a resolution of 12.5 × 12.5 m of the ALOS PALSAR (Phased Array Type L-band Syn- Spatial Information and workflow flowchart thetic Aperture Radar) platform. The processing of Land- sat images was conducted using the open-source software Spatial information was obtained from remote sensors from QGIS, and the vegetation cover and human settlements were the Landsat 5 and Land-at 7 satellites (Table 1) of the United calculated using the random forest plugin (Hantson et al. States Geological Survey (USGS), with a spatial resolution 2011). The sequential procedure followed in the study is shown in Fig. 2. Land use change Table 1 Images from satellite sensors for the period 2000–2008 in the study area For land use change, the random forest (RF) algorithm was Acquisition date Satellite/sensor Path/row Spectral band employed, as it is widely used and has demonstrated good thematic accuracy (Zhao et al. 2019; Ramírez et al. 2020). October 07, 2000 Landsat 5 TM 10/65 1,2,3,4,5 y 7 Together with the PFHS map (SERNAP 2016a), five classes September 03, 2002 Landsat 7 TM 10/65 1,2,3,4,5 y 7 of land use coverage (LUC) were identified: dense dry for- August 15, 2004 Landsat 5 TM 10/65 1,2,3,4,5 y 7 est (DDF), sparse dry forest (SDF), agricultural land (AL) May 06, 2008 Landsat 5 TM 10/64 1,2,3,4,5 y 7 and bodies of water (BW) (Table 2). The calibration of the Fig. 2 Methodological workflow to analyze changes in vegetation cover and land use in the dry forest of the Pómac Forest Historical Sanctuary (PFHS) in northern Peru International Journal of Environmental Science and Technology Table 2 LUC classes identified Level I Level II Level III LUC in the PFHS using the CORINE Land Cover methodology 2. Agricultural areas 2.1.1 Farmland 2.1.1. Permanently irrigated land AL adapted for Peru. Source: own elaboration based on MINAM 3. Forests and mostly natu- 3.1 Forests 3.1.1. Low dense forest DDF (2015) ral areas 3.1.2. Low sparse forest SDF 3.1.5. Fragmented forest 3.3 Open spaces with little 3.3.4 Bare land BL or no vegetation 5. Water surfaces 5.1 Inland water 5.1.2. Lagoons, lakes and permanent BW natural swamps AL agricultural land, DDF dense dry forest, SDF sparse dry forest, BL bare land, BW bodies of water classification was carried out based on spectral signatures Spatiotemporal intensity of the rate of change of 157 regions of interest within the PFHS. Among these and transition matrices regions, 28 points correspond to areas of dense dry forest (DDF), 80 to sparse dry forest (SDF), 24 to agricultural The transition matrix allows the analysis of the annual land (AL), 2 to bodies of water (BW), and 23 to bare soil rate of land use change between two different periods, using six bands of the visible and infrared spectrum of the through the equation proposed by Anand et al.  (2020), remote sensors of the Landsat satellites. The spectral dis- for the identification of the main transitions and detection tances of the regions of interest were assessed using the Jef- of changes in the evaluated classes (Pontius and McEach- fries–Matusita (JM) distance in a range from 0 to 2, where ern 2004). The negative values of the equation indicate a the values that tend to 2 present different spectral signatures decrease in LUC, on the contrary, a positive value indi- (Santana et al. 2014). cates an increase in it, and it is calculated as follows: 1 ( ) S2 t2−t1 Thematic accuracy S = − 1 (2) S1 The validation of the thematic classification aims to obtain where, S1 and S2 are the LUC surfaces at date t1 and t2. the quantitative description of the accuracy achieved dur- The obtained results were as follows: (i) exchange (Int) ing the process (Moreno and Chuvieco 2009). The selection between classes, in which the gain of one type of cover- of samples and determination of relationships between the age implies the loss of another; (ii) gain (Gj) is estimated, assigned cover and the reference cover is obtained through based on the differences in the total area of each class j on the construction of the confusion matrix, which indicates the date 2 (P + j), and (iii) the persistence that is expressed in level of correspondence between the relationships (Strahler the diagonal of the matrix (Pjj). The loss (Li), is the differ- et al. 2006). The importance of it is that it allows us to cap- ence between the total area of a class i at date 1 (Pi +) and ture the main conflicts between categories (Congalton and the persistence. To estimate the total change at the class Mead 1983). level (Ct), the gains (Gj) and the losses (Li) are added, The analysis of multiple relationships between classes while, for the net change, which indicates a definitive was carried out using the Kappa statistic, which measures change, it is represented as the difference between the total the difference between the map and the observed reality, change (Ct) and the exchange (Int) (Table 3, 4, and 5). through the following equation (Pontius et al. 2004): ∑ ∑ m X − X X i=1,n ii i=1,n i+ +i Multitemporal monitoring k = ∑ (1) m2 − X X i=1,n i+ +i For the assessment of the recovery process of the PFHS, where, n is the number of rows in the matrix; Xii the num- following the population occupation it underwent between ber of observations in row i and column i; Xi + X + i are the the years 2001 and 2009, a multitemporal analysis of veg- marginal total of row i and column i, respectively, and m is etation indices from Landsat images was conducted. These the total number of observations. images were obtained through the geospatial analysis International Journal of Environmental Science and Technology Table 3 Cross tabulation matrix for LUC in the period 2000–2002 2000 2002 Total 2000 (ha) Exchange rate Loss (Li) Total, change (Tc) Net change (Nc) Interchange (Int) DDF SDF BW AL BL % DDF 1,010.49 69.31 1.50 9.77 2.39 1,093.46 23.66 7.59 68.08 52.91 15.18 SDF 638.37 2,679.35 6.04 72.04 29.16 3,424.96 − 5.34 21.77 33.15 10.39 22.76 BW 4.37 10.92 9.77 0.75 3.99 29.80 − 9.00 67.21 117.25 17.18 100.07 BL 18.77 309.47 7.37 21.65 981.65 1,338.91 − 12.84 26.68 29.34 24.03 5.31 Total 2002 (ha) 1,672.00 3,069.05 24.68 104.21 1,017.19 5,887.13 Gains (Gj) 60.50 11.38 50.03 2.65 DDF dense dry forest, SDF sparse dry forest, AL agricultural land, BL bare land, and BW bodies of water Table 4 Cross tabulation matrix for LUC in the period 2002–2004 2002 2004 Total 2002 (ha) Exchange rate Loss (Li) Total, change (Tc) Net change (Nc) Interchange (Int) DDF SDF BW AL BL % DDF 1,295.75 342.68 0.76 32.65 0.14 1,671.98 − 4.07 22.50 37.02 7.98 29.04 SDF 236.62 2,536.01 5.28 270.73 20.40 3,069.04 1.36 17.37 37.48 2.75 34.74 BW 2.19 13.07 8.90 0.00 0.51 24.67 4.39 63.91 136.81 8.98 127.83 AL 0.00 0.00 0.19 104.01 0.00 104.21 67.75 0.19 293.41 293.04 0.37 BL 3.99 261.55 11.76 2.18 737.71 1,017.18 − 13.63 27.47 29.54 25.41 4.14 Total 2004 (ha) 1,538.55 3,153.31 26.89 409.58 758.76 5,887.08 Gain (Gj) 14.52 20.11 72.89 293.23 2.07 DDF dense dry forest, SDF sparse dry forest, AL agricultural land, BL bare land, and BW bodies of water International Journal of Environmental Science and Technology Fig. 3 Space–time dynamics of coverage and land use for the years 2000, 2002, 2004 and 2008 of the PFHS. Where AL (agricultural land), DDF (dense dry forest), SDF (sparse dry forest), BL (bare land) and BW (bodies of water) platform, Google Earth Engine (GEE). The obtained indices represent annual average values for the years 2009–2022. Through this evaluation, the aim was to demonstrate the recovery process of tropical dry forests that have expe- rienced deforestation. The following indices were used: (a) Normalized Difference Vegetation Index (NDVI). Anand et al. (2020) stated that NDVI is one of the most widely used indices and exhibits good efficiency for ana- lyzing vegetation of both high and low density. NIR − RED NDVI = (3) NIR + RED (b) Enhanced Vegetation Index (EVI). EVI is an optimized vegetation index, originally devel- oped as an improvement over NDVI. It is used to achieve greater sensitivity in the high biomass region by de-coupling background variables and atmospheric influences (Anand et al. 2020). NIR − RED EVI = 2.5 ∗ (4) NIR + 2.4 ∗ RED + 1 Results and discussion Evaluation of land use change in the PFHS Land use change (LUC) studies allow us to know the differ- ent processes that the territories are going through, being the Table 5 Cross tabulation matrix for LUC in the period 2004–2008 2004 2008 Total 2004 (ha) Exchange rate Loss (Li) Total, change (Tc) Net change (Nc) Interchange (Int) DDF SDF BW AL BL % DDF 905.02 558.17 11.90 60.27 3.46 1,538.81 − 16.88 41.19 51.46 30.91 20.55 SDF 155.35 2,383.81 30.47 397.08 186.60 3,153.32 − 2.55 24.40 43.77 5.03 38.74 BW 1.37 7.46 16.04 0.38 1.63 26.89 49.46 40.34 204.08 123.40 80.69 AL 0.00 0.00 0.62 408.79 0.00 409.40 45.94 0.15 113.29 112.99 0.30 BL 1.36 45.12 1.04 5.48 705.75 758.76 8.76 6.99 32.25 18.28 13.97 Total 2008(ha) 1,063.11 2,994.57 60.07 872.00 897.45 5,887.18 Gain (Gj) 10.27 19.37 163.74 113.14 25.26 DDF dense dry forest, SDF sparse dry forest, AL agricultural land, BL bare land, and BW bodies of water International Journal of Environmental Science and Technology economic activities practiced by different societies the main (v) Agricultural Land (AL) is evident from the year 2002 causes of environmental deterioration (Squeo et al. 2007). with 104.2 hectares, increasing to 409.6 hectares by 2004, The Pómac Forest Historical Sanctuary suffered a population and further expanding to 872.1 hectares by 2008. invasion for almost 10 years, which destroyed 29% of the The graphic representation of the different land uses in original vegetation cover. Since year 2009, the SERNANP the PFHS during 2000, 2002, 2004 and 2008 is shown in has implemented actions for the protection and regeneration Fig. 4. The coverage of DDF presents a decreasing vari- of plant cover. ation for the years evaluated, being its occupied areas Figure 3 shows the spatiotemporal dynamics of land use mainly for the AL and SDF cover. The changes occurred and cover Changes (LUCs) in the Pómac Forest Historical in the different vegetation covers that comprise the PFHS Sanctuary using 2000 as the base year. The following pat- are significantly related to the process of population terns are evident: (i) Dense Dry Forest (DDF) exhibits an invasion and natural phenomena (Holmgren and Schef- increase of 52% by the year 2002, followed by a decrease of fer 2001). The dense dry forest (DDF) for the year 2002 8% in 2004, and a further reduction of 44.7% in 2008 com- shows a significant increase due to the rise in precipita- pared to the year 2004, (ii) Sparse Dry Forest (SDF) shows tion in the northern coast of Peru (Squeo et al. 2007). a decrease of 10.4%, 7.9%, and 12.6% for the years 2002, From the period 2000 to 2008, a reduction in vegetation 2004, and 2008, respectively, relative to the base year, (iii) cover of the SDF and DDF is evident, with fluctuations (iv) Bare Land (BL) exhibits a similar behavior to SDF, and of increases and decreases during each analysis period. Fig. 4 Pómac forest histori- cal sanctuary land use change maps, a 2000, b 2002, c 2004 and d 2008 International Journal of Environmental Science and Technology Fig. 5 Gains and losses for period: 2000–2002, 2002–2004 and 2004–2008. LUC (land use change); AL (agricultural land), DDF (dense dry for- est), SDF (sparse dry forest), BL (bare land) and BW (bodies of water) Additionally, in the year 2008, the SDF decreased by sparse dry forest (SDF) with 17.37% (Fig. 5). These losses 430.3 ha because of population invasion, leading to log- were mainly attributed to anthropogenic activity within the ging activities for charcoal production and other logging Pómac Forest Historical Sanctuary. practices (SERNAP 2016a, b).The tabulation matrix for In the third period (2004–2008), a negative rate of change the first period (2000–2002) shows a positive rate of is observed for dense dry forest (DDF) (− 16.88%) and change for the dense dry forest (DDF) coverage (23.66%) sparse dry forest (SDF) (− 2.55%) coverages. On the other and negative rates for sparse dry forest (SDF) (− 5.34%) hand, there is a positive rate of change for (AL) (45.94%), and bare land (BL) (− 12.84%) coverages. By the year (BW) (49.46%) and (BL) (8.76%) (Table 5). The AL pre- 2002, the impacts generated by agricultural activity sented a gain of 113.14%, which is attributed to a defor- within the Pómac Forest Historical Sanctuary can already estation of the DDF and SDF vegetation covers with losses be observed, with an approximate area of 104.21 hectares of − 41.19% and − 24.40%, respectively. Additionally, during of agricultural land (SA) (Table 3). this period, the SD evidenced a gain of 25.26% (Fig. 5). The dense dry forest (DDF) is the vegetation cover that Throughout the analysis period, the deforestation of 102.6 shows the highest surface change at 68.08%, with a gain hectares was identified for DDF coverage, while for ODF of 60.5% and a loss of 7.59% (Fig. 5). The second cover a reduction of 739.9 hectares was observed, as a result of that underwent changes is the Sparse Dry Forest (SDF) with agricultural expansion in the PFHS. 33.15%, experiencing a loss of 21.77% and a gain of 11.38% relative to the base year. Calibration and validation of land use coverage In the second period (2002–2004), a positive rate of (LUC) change is observed in agricultural land (AL) with a net change of 293.41%, indicating a gain of 409.58 hectares of The degree of differentiation or similarity of the LUC agricultural coverage. During this period, there was defor- classes present a range of 0.98–2.0. The covers of DDF estation of 270.73 hectares of sparse dry forest (SDF) and and SDF present a similarity due to the scarce presence of 32.65 hectares of dense dry forest (DDF) (Table 4). Dense leaves. After that, a visual correction was performed. The dry forest (DDF) experienced a loss of 22.5%, followed by global accuracy for the classification of multispectral images International Journal of Environmental Science and Technology (Landsat 5 and Landsat 8) in the years 2000, 2002, 2004 is associated with higher precipitation due to the ENSO. and 2008 was 95.39, 95.61, 96.71 and 94.08%, respectively. In the period 2018–2022, the values remain within the The Kappa index (k) was 92.03, 92.87, 94.73 and 91.03 range of 3.5–4 for the NDVI and 2.1–2.4 for the EVI. In which indicate a high classification agreement (Hol- general, both indices show a recovery trend in deforested mgren and Scheffer 2001). In the confusion matrix for areas, revealing an improvement from 0.22 to 0.36 in NDVI, the year 2000, the BW class shows an optimum level of while for EVI 0.14–0.21 in the period 2009–2022. On the accuracy of 100%. The producer affirms that 96.6, 94.2 other hand, the EVI highlights areas with sparse vegeta- and 93.2% of the SDF, BL, and DDF classes were cor- tion and bare soil more distinctly. Both vegetation indices rectly identified. That is, the same user will find that the distinguish the year 2017 as an atypical year that signifi- DDF, SDF and BL classes correspond to field 96.5, 95.5 cantly contributed to the recovery of the tropical dry forest and 94.2%, respectively. in the PFHS. The process of human occupation has altered The confusion matrix for the year 2004 indicates that the structure of tropical dry forests, leading to a loss in the the BW and AL classes are not discussed as they exhibit diversity of forest species. Notably, NDVI shows higher an optimal accuracy level of 100%. The producer asserts values than EVI, which, according to Anand et al. (2020), that 98.3%, 97.1%, and 90.5% of the DDF, SDF, and BL is influenced by soil and leads to an increase in the irradia- and AL classes were correctly identified, respectively. tion of the near-infrared (NIR) band. Consequently, several On the other hand, a different user found that the field authors agree that the EVI index provides greater accuracy observations matched the DDF, SDF, AL, and BL classes in forest monitoring (Adán et al. 2014; Anand et al. 2020). at rates of 95.8%, 97.1%, 96.9%, and 96.6%, respectively. According to the last population census carried out by The confusion matrix for the year 2008 shows that the National Institute of Statistics and Informatics (INEI the BW class is not discussed since it exhibits an opti- for its acronym in Spanish), in 1993 the total population mal accuracy level of 100%. The producer affirms that in the district of Pitipo was 14,221 inhabitants. In 2007, 93.9%, 93.6%, 98.5%, and 89.6% of the DDF, SDF, AL, the population increased in 20,080 inhabitants; finally, and BL classes were correctly identified, respectively. in 2017, the population decreased to 19,651. This fac- On the other hand, a different user found that the field tor contributes to increased pressure on forest resources observations matched the DDF, SDF, AL, and BL classes provided by the Pómac Forest Historical Sanctuary. On at rates of 93.9%, 95.3%, 97.1%, and 85.7%, respectively. the other hand, during the period 2002–2007, the Lam- bayeque region registered 217,139 immigrants, coming Multitemporal analysis of vegetation indices from other parts of the country and abroad. This study contributes to the monitoring of forest The indices that have been used for the temporal evalu- resources in Peru, allowing to obtain thematic informa- ation of vegetation over long periods, with good results, tion which serves as a basis for formulating projects for are the NDVI and EVI (Ba et al. 2022; Roy 2021; Peng the management and conservation of forest resources et al. 2017; Li et al. 2010).The multitemporal analysis in the Pómac Forest Historical Sanctuary. In addition, of NDVI and EVI indices reveals a positive increasing this work allows to know the levels of deforestation and trend, attributed to the biomass growth resulting from the degree of conservation in an efficient and precise way. recovery process (Kubota et al. 2021). Figure 6 spatially Similar studies can be conducted in other Peruvian eco- displays the area that was occupied by human population, systems (Vargas-Sanabria and Campos-Vargas 2018). located in the Northeast of the PFHS. Through the NDVI Complementary studies are needed to monitor tropical analysis, slight recovery trend during the period from dry forests in Peru based on UAVs and multispectral 2009 to 2016 is shown, where NDVI values range from cameras that allow to collect information quickly and 0.2 to 0.3, and EVI values range from 0.14 to 0.20. In accurately with high spatial resolution. the year 2017, there is an increase in both NDVI and EVI values, reaching 0.5 and 0.4, respectively. This increase International Journal of Environmental Science and Technology Fig. 6 Annual values of: a Normalized difference vegetation index (NDVI) and b enhanced vegetation index (EVI) Conclusion sparse dry forests (SDF) were lost, which represents 14% of the total area of the sanctuary. These coverages gave up their The classification supervised by random forest showed that areas to intensive agricultural activities carried out by the there was a change in land use in the period from 2000 to illegal occupation of the Pómac Forest Historical Sanctu- 2008 due to anthropogenic activities, for example, a total of ary. After 2008, the NDV and EVI values were used on the 842.4 hectares of coverage of dense dry forests (DDF) and Google Earth Engine platform to monitor the recovery of International Journal of Environmental Science and Technology deforested areas, observing the growth of vegetation in the la provincia de Santa Elena, Ecuador: una perspectiva desde rainy years caused by ENSO. The recovery of the primary la conservación. Ind Data 22(2):117–138. https:// doi. org/ 10. forest is slow, which is why the immediate action of the 15381/i data.v 22i2. 17393Ba R, Song W, Lovallo M, Zhang H, Telesca L (2022) Informational authorities is necessary for the implementation of foresta- analysis of MODIS NDVI and EVI time series of sites affected tion programs in the affected areas determined in this study. and unaffected by wildfires. Phys Stat Mech Its Appl 604:127911. https:// doi. org/ 10. 1016/j. physa. 2022. 127911 Acknowledgements We thank Eric Rodriguez, Maria Angélica Puyo Chuvieco E (1996) Fundamentos de teledetección espacial. Ediciones and Cristina Aybar for supporting the logistic activities in our labora- RIALP S.A., Madrid tory. C.I.A. thanks Vicerrectorado de Investigación of UNTRM. Congalton RG, Mead RA (1983) A quantitative method to test for consistency and correctness in photointerpretation. Am Soc Pho- Funding This research was funded by the following project “Creación togramm 49(1):69–74 del Servicio de Agricultura de Precisión en los Departamentos de Cuello AR, Antes ME, Lois ASO (2015) Generation of spectral-tem- Lambayeque, Huancavelica, Ucayali y San Martín 4 Departamentos” poral response surfaces by combining multispectral satellite ande CUI 2449640 of the Ministry of Agrarian Development and Irrigation hyperspectral UAV imagery for precision agriculture application. (MIDAGRI) of the Peruvian Government. IEEE J Sel Top Appl Earth Obs Remote Sens 8(6):3140–3146 Espinosa C, Cruz M, Luzuriaga A, Escudero A (2012) Bosque Tropical Data availability All data generated during this study are included in Seco de la Región Pacífico Ecuatorial: 318 Diversidad, Estructura, this published article. Funcionamiento e Implicaciones Para La Conservación. Ecosis- temas 21:167–179 Declarations Gevaert CM, Suomalainen J, Tang J, Kooistra L (2015) Generation of spectral-temporal response surfaces by combining multispectral Conflict of interest The authors declare no conflict of interest. satellite and hyperspectral UAV imagery for precision agricul- ture application. IEEE J Sel Top Appl Earth Obs Remote Sens Open Access This article is licensed under a Creative Commons Attri- 8(6):3140–3146 bution 4.0 International License, which permits use, sharing, adapta- Hantson S, Chuvieco E, Pons X, Domingo-Marimon C (2011) Cadena tion, distribution and reproduction in any medium or format, as long de pre-procesamiento estándar para las imágenes Landsat del Plan as you give appropriate credit to the original author(s) and the source, Nacional de Teledetección. Revista de teledeteccion 36:51–61 provide a link to the Creative Commons licence, and indicate if changes Holmgren M, Scheffer M (2001) El Niño as a window of opportu- were made. The images or other third party material in this article are nity for the restoration of degraded arid ecosystems. Ecosystems included in the article’s Creative Commons licence, unless indicated 4(2):151–159. https:// doi. org/1 0. 1007/ s1002 100000 65 otherwise in a credit line to the material. If material is not included in Jiménez A, Macías A, Ramos M, Tapia M, Rosete S (2019) Indi- the article’s Creative Commons licence and your intended use is not cadores de sostenibilidad con énfasis en el estado de conser- permitted by statutory regulation or exceeds the permitted use, you will vación del bosque seco tropical. Rev Cubana Ciencia Forestales need to obtain permission directly from the copyright holder. To view a 7(2):197–211 copy of this licence, visit http://creativecommons.org/licenses/by/4.0/. Kubota V, Caballero R, Fernández A (2021) Variación de biomasa en un periodo de 21 años en un Bosque Atlántico del Alto Paraná (Paraguay). Colombia Forestal 24(1):60–70. https:// doi. org/ 10. 15381/r pb.v 29i4. 21613 References Li Zhongfei, Li Xiaobing, Dandan Wei XuXu, Wang Hong (2010) An assessment of correlation on MODIS-NDVI and EVI with natural vegetation coverage in Northern Hebei Province, China. Adán E, Mutanga O, Abdel-Rahman E, Ismail R (2014) Estimación de Procedia Environ Sci 2:964–969. https://d oi.o rg/1 0.1 016/j.p roenv. la biomasa en pie en el pantano de papiro (Cyperus papyrus L.): 2010. 10. 108 exploración de índices hiperes-pectrales in situ y regresión alea- Maass Manuel, Burgos Ana (2011) Water dynamics at the ecosystem toria de bosques En t. J Remote Sens. https:// doi. org/ 10. 1016/j. level in seasonally dry tropical forests. Seasonally dry tropical cub. 2018.0 3. 014 forests. Island Press/Center for Resource Economics, Washington, Alejandra M, Romero C (2016) Introducción conservación y rentabi- DC, pp 141–156. https://d oi. org/ 10. 5822/ 978-1- 61091- 021-7_9 lidad de la transformación artesanal de los recursos del bosque Miles L, Newton AC, DeFries RS, Ravilious C, May I, Blyth S, Kapos secos. Investig Territ 2:106–118 V, Gordon JE (2006) A global overview of the conservation status Álvarez-Dávila E, Cayuela L, González-Caro S, Aldana AM, Steven- of tropical dry forests. J Biogeogr 33(3):491–505. https://d oi.o rg/ son PR, Phillips O, Cogollo Á, Peñuela MC, Von Hildebrand P, 10. 1111/j. 1365-2 699. 2005. 01424.x Jiménez E, Melo O, Londoño-Vega AC, Mendoza I, Velásquez O, MINAM (2022) Mapa nacional de ecosistemas del Perú. Lima, Perú Fernández F, Serna M, Velázquez-Rua C, Benítez D, Rey-Benayas Moreno MV, Chuvieco E (2009) Validation of global land cover prod- JM (2017) Forest biomass density across large climate gradients ucts for the Spanish Peninsular area. Revista de Teledetección in northern South America is related to water availability but not 1(31):31–22 with temperature. PLoS ONE. https:// doi. org/ 10. 1371/ journ al. Peng D, Wu C, Li C, Zhang X, Liu Z, Ye H, Luo S, Liu X, Hu Y, pone.0 17107 2 Fang B (2017) Spring green-up phenology products derived from Anand A, Pandey P, Petropoulos G, Pavlidés A, Srivastava P, Sharma MODIS NDVI and EVI: intercomparison, interpretation and vali- J, Malhi R (2020) Uso de Hyperion para la evaluación de las dation using national phenology network and ameriflux observa- reservas de carbono de los bosques de manglares en la reserva tions. Ecol Indic 77:323–336. https:// doi. org/ 10. 1016/j. ecoli nd. fores-tal de Bhitarkanika: una contribución a la iniciativa del 2017. 02.0 24 carbono azul. Sens Remotos 12:58. https:// doi. org/ 10. 3390/ Pontius RG, Shusas E, McEachern M (2004) Detecting important rs1204 0597 categorical land changes while accounting for persistence. Agric Astudillo-Sánchez E, Pérez Flor J, Medina G, Medina A (2020) Ecosyst Environ 101:251–268. https:// doi. org/ 10. 1016/j. agee. Gestión de los bosques tropicales estacionalmente secos de 2003.0 9. 008 International Journal of Environmental Science and Technology Quiroga J, Rosa H, Melo O, Fernández F (2019) Estructura de frag- in the Atacama desert. J Veg. Sci 18:95. https://d oi.o rg/1 0.1 8845/ mentos de Bosque seco Tropical en el departamento de Tolima. tm.v 33i4. 4528 Bol Cient Mus Hist Nat 23(1):31–51 Stan K, Sanchez-Azofeifa A (2019) Tropical dry forest diversity, cli- Ramírez M, Martinez L, Montilla M, Sarmiento O, Lasso J, Díaz matic response, and resilience in a changing climate. Forests. S (2020) Obtención de coberturas del suelo agropecuario en https://d oi.o rg/1 0.3 390/ f10050 443 imagenes satelitales Sentinel-2 con la inyección de imágenes de Strahler AH, Boschetti L, Foody GM, Friedl MA, Hansen MC, Herold dron usando random forest en google earth engine. Rev Telede- M, Mayaux P, Morisette JT, Stehman SV, Woodcock CE (2006) teccion 56:49–68 Global land cover validation: recommendations for evaluation Renzo P (2003) Propuesta de conservación del bosque seco en el Sur and accuracy assessment of global land cover maps. Ecosistemas. de Ecuador. Lyonia 4(2):183–186 https:// doi.o rg/ 10.7 818/E COS. 2016.2 5-2. 06 Roy B (2021) Optimum machine learning algorithm selection for fore- Vargas-Sanabria D, Campos-Vargas C (2018) Sistema Multi-Algoritmo casting vegetation indices: MODIS NDVI & EVI. Remote Sens Para La Clasificación de Coberturas de La Tierra En El Bosque Appl Soc Environ 23:100582. https:// doi. org/ 10. 1016/j. rsase. Seco Tropical Del Área de Conservación Guanacaste. Costa Rica 2021.1 00582 Rev Tecnol 32:58. https://d oi.o rg/ 10. 18845/ tm. v31i1. 3497 Santana MV, Elsa J, Rosales M, Luis E and Solís RM (2014) Aportes Zhao L, Shi Y, Liu B, Hovis C, Duan Y, Shi Z (2019) Finer classifica- de la Geografía a ciudades saludables: caso Zona Metropolitana tion of crops by fusing uav images and sentinel-2A data. Remote de Toluca Sens 11(21):3012. https:// doi.o rg/1 0. 3390/f 8040 108 SERNAP (2016a) Plan Maestro Santuario Histórico Bosque de Zhu Z, Wulder MA, Roy DP, Woodcock CE, Hansen MC, Radelo VC, Pómac-2011 Scambos TA (2019) Benefits of the free and open landsat data SERNAP (2016b) Plan Maestro Santuario Histórico Bosque de Pómac policy. Remote Sens Environ 224:382–385. https:// doi. org/ 10. 2011–2016. Lima-Perú. https:// legis lacio nanp. org. pe/ aprue 3390/ f13050 698 ban- el- plan- maest ro- perio do- 2011- 2016- del- santu ario- histo rico- bosque- de- pomac Squeo FA, Holmgren M, Jiménez M, Albán L, Reyes J, Gutiérrez JR (2007) Tree establishment along an ENSO experimental gradient