Field data on diversity and vegetation structure of natural regeneration in a chronosequence of abandoned gold-mining lands in a tropical Amazon forest

Abstract

Anthropogenic activities (e.g., logging, gold-mining, agriculture, and uncontrolled urban expansion) threaten the forests in the southeast of the Peruvian Amazon, one of the most diverse ecosystems worldwide. However, gold-mining generates the most severe impacts on ecosystems and limits its resilience. The natural regeneration of degraded areas in the southeastern Peruvian Amazon have not been studied deeply. The dataset contains floristic inventories of previously uncharacterized or poorly studied secondary forests degraded and abandoned by goldmining activities and an intact forest in the Tres Islas indigenous community, Madre de Dios region, in southeastern Peru. The data presented was obtained from 12 plots (20 m × 60 m) established in three successional forests abandoned by gold mining and an intact forest (without mining impacts), where all trees with a stem diameter at breast height greater than 1 cm were inventoried. To the best of our knowledge, this is the only dataset in the southwest of the Peruvian Amazon that compares the natural colonization after gold-mining and intact forests. This dataset can be useful for long-term study and monitoring of structure and tree diversity in relatively understudied yet important secondary forests after gold-mining abandonment. Also, this dataset could be used to analyze the successional trajectory process of vegetation and the recovery of aboveground biomass. Furthermore, the data could be used to investigate the effects of functional traits and types of mining on vegetation recovery. Hence, understanding the successional processes will help to improve restoration, reforestation, or reclamation strategies for the recovery of degraded lands in the Amazon.

Specifications Table

Subject	Environmental Science.
Specific subject area	Ecology, plant diversity, botany, floristics studies.
Type of data	Table, Raw, Processed.
Data collection	Field survey of plants. Vegetation data were collected in field using 20 m × 60 m plots, divided into three 20 m × 20 m sub-plots, where all individuals with a stem diameter at breast height (DBH) greater than 1 cm were inventoried. Total height and DBH of all individuals were measured using a clinometer (Suunto PM 5/360PC) and a diameter tape (Forestry Suppliers – 283D/10M), respectively. Four sites were selected for the sampling: three successional forests after gold mining and an intact forest (reference forest, without mining impacts). At each site, three plots were established, a total of 12 plots (36 sub-plots) [1].
Data source location	Tres Islas indigenous community, Madre de Dios region, in southeastern Peru. Latitude: 12° 31′ 34.22″ & Longitude: 69°23′ 45.77″ W
Data accessibility	Repository name: Mendeley Data Data identification number: 10.17632/58hc2ttj39.2 Direct URL to data: https://data.mendeley.com/datasets/58hc2ttj39/2 Instructions for accessing these data: File 1: Nat_regen_Mining_vegetation v6.xlsx contains the list of tree plant species sampled in three successional forests after gold mining and an intact forest
Related research article	Garate-Quispe, J., Canahuire-Robles, R., Alarcón-Aguirre, G., Dueñas-Linares, H., F. Roman-Dañobeytia, Changes in floristic and vegetation structure in a chronosequence of abandoned gold-mining lands in a tropical Amazon forest, Heliyon, 10(9): E29908. 10.1016/j.heliyon.2024.e29908

1. Value of the Data

• •The dataset contains floristic inventories of previously uncharacterized or poorly studied. This emphasizes the potential to identify potential tree species for restoration, reforestation or reclamation, thus improving applications for the recovery of areas degraded by gold mining in the Amazon.
• •The data contribute to characterizing the natural regeneration in Amazon forests after mining abandonment and allow comparative studies with reference forests.
• •Can be used by researchers to investigate the effects of functional traits and types of mining vegetation recovery.
• •The data on floristic composition can be combined with allometric equations to obtain estimates of biomass or carbon recovery after mining and compare them with reference forest carbon stocks.
• •To the best of our knowledge, this is the only dataset in the southwest of the Peruvian Amazon that compares the natural colonization after gold-mining and intact forests.
• •The methodology (size and design plot) used for vegetation sampling can be applied in studies of natural regeneration in degraded and abandoned areas of the Peruvian Amazon in a standardized way.

2. Background

Madre de Dios represents one of the most diverse ecosystems in the Amazon [2]. However, this Amazonian biodiversity hotspot is threatened by anthropogenic activities [3,4]. However, mining generates the most severe impacts on ecosystems, thus limiting the establishment of natural regeneration [5,6]. The natural regeneration from degraded areas in the southeastern Peruvian Amazon have not been studied deeply in terms of species diversity, floristic composition, biomass, forest structure, seed dispersal and successional trajectories. The dataset represents a follow-up to a published article [7] where the objective was to compare the forest recovery after gold mining with a reference forest in the southeastern Peruvian Amazon.

The dataset aimed to provide a baseline for long-term study and monitoring of structure and tree diversity in a relatively understudied yet important secondary forests after gold-mining abandonment. Also, this dataset could be used to analyze the successional trajectory process of vegetation and the recovery of aboveground biomass. Furthermore, the data could be used to investigate the effects of functional traits and types of mining on vegetation recovery. Hence, it will help to understand the successional processes and thus improve restoration, reforestation or reclamation strategies for the recovery of degraded lands in the Amazon.

3. Data Description

Data presented was obtained from 12 plots established in three successional forests abandoned by gold mining and an intact forest (reference forest, without mining impacts), in the southwestern Peruvian Amazon. The datasets on species occurring in the evaluated forests can be found in the repository (10.25381/10.17632/58hc2ttj39.2). The file includes information for 13 variables that are explained in Table 1.

Table 1.

Variables of vegetation.

N	Variable	Explanation	Summary information
1	N	Standardized number used for the dataset, consisting of the registration number assigned to each identified tree.	The total is 3371 trees with a DBH > 1 cm.
2	Forest vegetation type	Type of site where the vegetation inventory was carried out: (1) successional forests abandoned by gold mining, or (2) intact forest (reference forest, without mining impacts).	Of the total (3371 trees), 74 % are from successional forests abandoned by gold mining, and 26 % from intact forest.
3	Age (years)	Time since abandonment by mining (years) in successional forests. For reference forest without impacts caused by mining activities the code was “intact forest”.	The time since abandonment by mining ranged from 5 to 23 years.
4	Plot	Sampling plot identifier (ranged 1 to 12).	The trees were found on 12 plots, between 181 and 366 trees by plot. The mean number of trees per plot is 283 and the median is 273.
5	Sub-plot	Sampling sub-plot identifier, ranged from 1.1 to 12.3 (3 subplots per plot), where the first number represents the parcel followed by a dot and then the subplot number.	The trees were found on 36 sub-plots, between 46 and 126 trees per sub-plot. The mean number of trees per sub-plot is 94 and the median is 93.
6	Species	The tree species were taxonomically identified according to the Angiosperm Phylogeny Group IV [8]. The Taxonomic Name Resolution Service (TNRS) app was used to review the accepted scientific names of the trees [9].	The total is 129 species; the mean is 26 trees per species, but the median is 10; 103 species are in successional forest after mining and 66 species in intact forest. The most abundant tree species are Inga marginata (299 trees), Cecropia membranacea (253), Inga sertulifera (227), Tessaria integrifolia (163), Margaritaria nobilis (154), Senegalia polyphylla (137) and Sapium marmieri (136).
7	Author	The abbreviated author(s) of the tree species name. TNRS app was used to review the accepted author(s) name [9]	There are 93 different authors or author combinations.
8	Genus	The corresponding genus of the tree species was identified according to the Angiosperm Phylogeny Group IV [8]. The Taxonomic Name Resolution Service (TNRS) app was used to review the accepted botanical genus of the trees [9].	The trees are from 94 botanical genera; 78 genera are in successional forest after mining and 56 in intact forest. The most abundant tree genera are Inga (17.7 % of the trees), Cecropia (7.7 %), Tessaria (4.8 %), Sapium (4.7 %), Margaritaria (4.6 %), and Senegalia (4.1 %).
9	Family	The corresponding family of tree species was identified according to the Angiosperm Phylogeny Group IV [8]. The Taxonomic Name Resolution Service (TNRS) app was used to review the accepted botanical family of the trees [9].	The trees are from 45 botanical families; 40 families are in successional forest after mining and 32 in intact forest. The most abundant families are Fabaceae (26.3 % of the trees), Urticaceae (8.8 %), Euphorbiaceae (7.7 %), Asteraceae (5.2 %), and Moraceae (4.8 %).
10	Height (m)	Total height (m) measured for all trees using a clinometer (Suunto PM 5/360 PC).	The total height ranged from 1 to 37 m. The mean tree height is 7.9 m, and the median is 6 m.
11	Diameter (cm)	The stem diameter at breast height (DBH) was measured for all trees using a diameter tape (Forestry Suppliers – 283D/10 M).	DBH ranged from 1.1 to 115.9 cm. The mean DBH of trees is 11.1 cm, and the median is 6.7 cm.
12	Height Class	Trees were divided into height classes from <5 m to < 40 m at an interval of 5 m.	There are 8 height classes. The most abundant height classes are 5 to < 10 m (51.9 % of the trees), 0 to < 5 m (24.8 %), and 10 to < 15 m (11.5 %).
13	Diameter Class	Trees were divided into diameter classes ranging from <10 cm to 120 cm at a 10 cm interval.	There are 11 diameter classes. The most abundant diameter classes are 1 to < 10 cm (69.1 % of the trees), 10 to < 20 cm (16.5 %), and 20 to < 30 cm (8.2 %).
14	Stage of development	All trees were classified according to the stage of development [10], as saplings (1 cm to < 5 cm DBH), poles (5 cm to < 10 cm DBH) and trees (>10 cm DBH).	The most abundant stages of development for the tree species are poles (38.21 % trees), saplings (30.94 %), and trees (30.85 %).

4. Experimental Design, Materials and Methods

4.1. Site location

This study focuses on secondary forests degraded and abandoned by goldmining activities and an intact forest in the Tres Islas indigenous community, Madre de Dios region, in southeastern Peru (latitude: 12° 31′ 34.22″ and longitude: 69°23′ 45.77″ W). Its elevation range varies between 195 and 215 m above sea level. The study area has a dry season from June to September when the rainfall is less than 100 mm per month [11]. The mean annual rainfall and mean annual temperature are 2120 mm and 25 °C, respectively. Four sites were selected for the sampling, three successional forests after gold mining and an intact forest (reference forest, without mining impacts). At each site, three plots were established, a total of 12 plots (36 sub-plots).

4.2. Sampling vegetation

Data presented was obtained from 12 plots (20 m × 60 m) established in three successional forests abandoned by gold mining and an intact forest (reference forest, without mining impacts), in the southwestern Peruvian Amazon. Vegetation data were collected in the field using 20 m × 60 m plots, divided into three 20 m × 20 m sub-plots, where all trees with a stem diameter at breast height (DBH) greater than 1 cm were inventoried. Total height and DBH of all individuals were measured using a clinometer (Suunto PM 5/360PC) and a diameter tape (Forestry Suppliers – 283D/10M), respectively. In the field, all trees were identified by botanists and a parataxonomist. Specimens of unidentified species were collected for identification at the Alwyn Gentry Herbarium. All species were taxonomically identified according to the Angiosperm Phylogeny Group IV [8]. The Taxonomic Name Resolution Service (TNRS) app was used to review the accepted scientific names [9] (Fig. 1).

Fig. 1.

Location of the study area in the Southeastern Peruvian Amazon (a), Madre de Dios (b), showing the four sites where the field work took place (c). (c) Landsat-8 images (July 2023) over the gold mining corridor of Madre de Dios.

Limitations

There are some limitations in our dataset. First, the study focused solely on degraded areas in a floodplain ecosystem, which affects the generalizability of the results because the areas impacted by gold mining in the southeast Peruvian Amazon reaches to the uplands and piedmont areas. Second, due to the inaccessibility of many areas, a small sample size of the forest regenerated after mining activities was used for plant sampling. However, the total raw data was 3371 trees with DBH > 1 cm (including reference forest plots), which should compensate for the limitation described earlier and the data could provide new insights for a deeper understanding of natural regeneration in areas degraded by gold mining in the Peruvian Amazon.

Ethics Statement

This research did not involve human subjects, animals, or any species requiring ethical approval.

CRediT Author Statement

Jorge Garate-Quispe: Conceptualization, Investigation, Data curation, Formal analysis, Methodology and Writing - original draft. Ramiro Canahuire-Robles: Conceptualization, Methodology, Investigation and Data curation. Marx Herrera-Machaca: Data curation software, validation and Writing - review & editing. Sufer Baez-Quispe: Data curation and Writing - review & editing. Gabriel Alarcón-Aguirre: Formal analysis and Writing - review & editing.

Data Availability

• Mendeley DataData of natural regeneration in a forest chronosequence after mining lands in Southeast Peruvian Amazon (Original data).