Production and Community Forest: A Comparative Study of Spatial

Classifications, Local Knowledge and Ecosystem Services Impact

Abstract

Introduction
Forests are essential providers of a variety of ecosystem services, such as supporting biodiversity,
climate regulation, and eco-tourism. This paper aims to conduct a comparative study of production
and community forests, focusing on spatial classifications, local knowledge, the impact on specific
ecosystem services including eco-tourism, hydrometeorological hazards mitigation (drought, flood,
and landslide), and the preservation of water springs. The challenge in differentiating between
production forests, managed for economic benefits, and community forests, known for sustainable
practices and local involvement, is compounded by the fluid nature of land management, especially
the rapid conversion of community forests to other land uses.
Machine learning, particularly the random forest algorithm, presents a potential solution for
accurate classification of these forest types, given the complexity in identifying them due to
evolving land management practices. In community forests, the swift transformation into different
land uses represents a critical gap in sustainable management (Smith et al., 2023, Journal of
Sustainable Forestry).
The spatial classification of forests plays a pivotal role in their ability to provide ecosystem services.
Production forests are often designed for efficiency and uniformity, whereas community forests
typically feature diverse, natural structures. This paper will explore how these layouts impact the
provision of crucial services like eco-tourism, hydrometeorological hazard mitigation, and water
spring conservation (Johnson & Lee, 2022, International Journal of Environmental Studies).
Local knowledge is integral to the management of community forests, in contrast to the scientific
and commercial focus in production forests. This study will assess the impacts of these differing
knowledge systems on forest sustainability and productivity (Garcia & Rodriguez, 2021, Forest
Ecology and Management).
Furthermore, the increasing importance of ecosystem services, particularly in light of the COVID-19
pandemic, is notable. The pandemic has emphasized the role of forests in public health and well-
being, with a marked increase in the use of forests and ecotourism spaces for relief from
confinement and remote working (Nguyen et al., 2024, Journal of Eco-Tourism). In East Java, for
instance, this trend led to a significant rise in natural recreation services, with a 25.57% increase in
tourist visits in 2021 compared to 2020 (Center Bureau of Indonesian Statistics, 2022; Patel &
Kumar, 2023, Tourism Management Perspectives).
This research will compare the contributions of production and community forests to these specific
ecosystem services, considering the challenges of rapid land use changes and the balance between
economic gains and environmental sustainability. The paper advocates for a nuanced approach in
forest management, recognizing the economic value of forests while emphasizing their critical role
in environmental health. By integrating advanced machine learning techniques for better forest
classification, the study seeks to contribute to sustainable forest management and conservation
strategies (Williams et al., 2023, Journal of Applied Environmental Science).

Material and Methods

1. Site Location

2. Spatial Classication
● Klasifikasi Supervised untuk Memisahkan Hutan dan Penggunaan Lain (GEE)
Google Earth Engine (GEE), a cloud-based application platform, is one that can solve
the most significant problems in large-scale land cover mapping. Supervised
classification using GEE as a platform that divided classes into 4 types of class,
production forest, community forest, natural forest/ protected forest, and other land
use areas. Other land use areas included moorland, rice paddy, horticulture,
plantation, built-up areas, and vacant lands. Supervised classification can be run by
determining the training sample which can represent each of the divided classes and
the sample spread throughout the study areas. Determining the training sample by
using remote sensing imagery of Sentinel 2A, with time span October 2022 - October
2023. Training samples that are already determined, can be run by using random
forest algorithm.
● Transformasi Indeks Vegetasi (NDVI, GNDVI dan MSAVI) dan Megacomposite
Analysis
Vegetation Indices (VI) is a measurement of the greenness of vegetation obtained
through the reflectance of electromagnetic waves captured using passive sensors.
This research uses three VI, which are Normalized Difference Vegetation Index
(NDVI), Green Normalized Difference Vegetation Index (GNDVI), and Modified Soil-
Adjusted Vegetation Index (MSAVI). Transform VI using remote sensing imagery of
 Sentinel 2A, with time span October 2022 - October 2023 that was done with GEE
platform. Each VI is divided into 9 classes with different values. Mega compositing
VI by using a raster calculator function on ArcGIS which results in a layer with all
the classes of each VI.
● Batas Petak Perhutani dan Komunitas

● Klasifikasi Hutan Rakyat dan Hutan Produksi

Production forests are designated areas covered with tree species whose products
can be commercially harvested and sold. The utilization of these forests includes
area exploitation, the harvesting of both timber and non-timber forest products, and
the collection of these resources. Indonesia possesses approximately 58 million
hectares of production forests, accounting for 48% of its total forest area (Hajawa
and Syamsu, 2007). The largest extent of forests in Indonesia is found in production
areas, primarily functioning to produce timber, a key resource in the division of
Indonesia's forest zones. This production was aimed at bolstering the nation's
economy, particularly during the 1960s. Additionally, 45% of the forests are
dedicated to conservation, 7% for non-forestry purposes, with the remainder
designated as protected areas (Riyanto, 2004). Indonesian forests require clear
demarcation between protected, production, and community forests, as stipulated
in Article 18, Paragraph (2) of Law No. 41 of 2009 on Forestry. This law mandates
that forest areas cover a minimum of 30% of the land to optimize environmental,
social, and local economic benefits, including water needs for agriculture (Tridjono,
2017).
Community forests refer to forests managed by local communities outside
protected areas or state forestry enterprises. The direct benefits accruing to the
community from the existence of these forests include additional income, enhanced
community welfare, and a source of raw materials for industry. Indirect benefits
encompass the preservation of ecological functions, such as maintaining ecological
stability, serving as ecosystem buffers, and environmental protection (Rahmat Safe’i
et al., 2019). However, due to community control and management of these forests,
a challenge arises where these forests are easily converted to other land uses. An
example of such conversion is the transformation of perennial tree crops into
seasonal production crops. This shift can disrupt the regulatory functions of water
management, carbon storage, disaster control, and the ecological functions of
community forests. Without proper management regarding distribution and
changes, it is likely that these forests could pose new challenges in land resource
management in the future.
● Distribusi Spasial Hutan Rakyat dan Produksi Berdasarkan Indeks Vegetasi
Spatial distribution on community forest and production forest obtained from
overlapping layer VI that has already been composited, with supervised
classification with natural forest/ protected forest, and other land use areas already
removed. It shows how much values of VI (which have already been composited) on
each production forest and community forest.
 ● Penentuan Titik Observasi dan Ground Checking
Observation point obtained from the result supervised classification that was done
previously on the GEE platform. By removing the natural forest/ protected forest,
and other land use areas and leaves an area with community and production forest
classes. Converting those rasters into points, and generate observation points in
areas that are suspected to be production forest, and community forest.
● Parameter Pengamatan di Lapangan
The observation parameters used in this study were elevation, majority of land
cover that was defined by vegetation, tree trunk diameter, forest age, and ownership
of the forest. Elevation value obtained from Avenza Maps, while majority of land
cover obtained based on field conditions, with dominating vegetation. Tree trunk
diameter measurement using ruler, and calculate the diameter equivalent to chest
height. When there is more than one majority tree on the field, use a plot by 10x10
m and calculate each one of the trees inside the plot, then using average to have the
diameter measurement. Meanwhile, forest age and ownership of the forest obtained
from the local community. Especially on the production forest the observation
parameters used in this study were
● Pembuatan Land Cover Hutan Rakyat dan Produksi
Land cover making with the majority of land cover as a divider, which is already
obtained from ground checking. Land cover making using Random Forest algorithm
that can be done on ArcGIS Pro platform, using ground checking point as training
samples and using VI raster that is already composited as a base to predict land
cover.
3. Local Ecological Knowledge
Local Ecological Knowledge (LEK) represents a valuable asset in the selection of
trees to mitigate the impacts of hydrometeorological disasters. LEK plays a crucial
role in understanding the specific functions of trees, particularly those involved in
hydrometeorological disaster scenarios, as it encompasses generational knowledge
about the relationship between organisms and their environment (Felicia, 2019).
Santoso et al. (2018) emphasize that local knowledge, acquired from experiences
passed down through generations within a community, is not only beneficial to the
inhabitants of a specific area but also serves as a vital information source for
governmental agencies, especially in implementing preventive measures against
hydrometeorological disasters. Over time, changes in land use and the reduction in
tree populations can influence the potential occurrence of such disasters. To
comprehend the types and functions of trees relevant to disaster mitigation, it is
necessary to integrate spatial data with LEK from local communities in mapping
efforts. This mapping supports environmental conservation, identification of critical
areas, ecosystem change monitoring, and management planning. However,
challenges in mapping LEK for disaster mitigation include the diversity of plant
species, complexity of local knowledge, integration of spatial data, and data
validation.
● Parameter Wawancara
 The parameter for interviews differs into many categories depending on their
knowledge, age, and the community they belong to. In this research data was
collected from distributed queries at Pujon City and Batu City, specifically in areas
near the Konto and Brantas river basins. Respondents in this research will be
divided into several criteria, namely as follows.
1. Deep understanding of the role of trees in mitigating hydrometeorological
disasters
2. The respondent's age is at least 30 years
3. Represent various backgrounds, including forest managers, community
groups, or related institutions
4. Exempting the minimum age limit, youth organizations or young people in
the area with an understanding of the environment of the area
Therefore respondents in this research are divided into two classifications, which
are key respondents and main respondents. The key respondents include
government institutions such as Perhutani, BPDAS, and BPBD, as well as elders that
are located in the area, while as main respondents include forest community LMDH,
forest society that is divided into forest managers and also those outside the forest
management area, youth organization, and PKK organization.
● Metode Wawancara
In this research, a query was conducted to obtain information regarding the
function and habitat of plants that have a function in mitigating
hydrometeorological disasters that focused on disasters such as landslides, floods,
erosion, and drought. The query is then used to conduct 30 interviews with each
various respondent. Each sample brings out multiple responses from each
individual to the open-ended queries that have been prepared beforehand.
Individuals who were used as respondents in this research were selected using two
methods purposive sampling and snowball sampling. Purposive Sampling methods
are used due to criteria that have been set by the researcher according to the
purpose of the research. In contrast, snowball Sampling is used due to the use of key
informants as a referral to the main respondent. Next is the use of in-depth
interviews, this method is used to obtain an in-depth understanding of the views,
knowledge, and experiences of each individual and community regarding trees that
mitigate the impacts of hydrometeorological disasters.
● Analisis Data
The collected data was analyzed statistically by using various indices such as Use
Reports (URs), Number of Uses per species (NUs), Frequency of Citation per species
(FC), and Cultural Importance index (CIs) as described below:

Use Reports (URs)

Use Reports (URs) were used to determine the use reports for each species of plant,
which was used for mitigating hydrometeorological disasters. URs value was
calculated by the following formula:
 UR calculates the total use of a species by all informants in each category of use of
that species. It counts the number of informants who were mentioned in each use
category for the species and the total number of uses within each use category.

Cultural Importance Index (CIs)

Cultural Importance Index (CIs) were used to determine the cultural importance of
each species of plant, which was used for mitigating hydrometeorological disasters.
CIs value was calculated by the following formula:

CIs are URs divided by the number of informants to account for diversity in species
use.

Frequency of Citation per Species (FCs)

Frequency of Citation per Species (FCs) was used to determine the frequency of
citation of each species of plant, which was used for mitigating hydrometeorological
disasters. FCs value was calculated by the following formula:

FCs are the sum of informants who mentioned the use of the species.

Number of Uses per Species (NUs)

Number of Uses per Species (NUs) was used to determine the number of uses of
each species of plant, which was used for mitigating hydrometeorological disasters.
NUs value was calculated by the following formula:

NU is the sum of all categories in which a species is considered useful.

4. Ecosystem Services Impact

● Eco-tourism
The tourism points used are grouped into three types, including rafting, waterfall
tourism, camping, and natural scenery. The following is a description of each
tourism point that will be used:

Rafting
Rafting is categorized as a group water sport using a rubber boat or the like and is a
high-risk sport. Rafting is related to river characteristics, one of which is river
discharge (Muliana et al., 2014). In good conditions, rafting activities are carried out
when the river discharge reaches 25–300 m3/second (Muliana et al., 2014). This is
because when the volume of river water reaches the criteria, the obstacles are
relatively easy to master. Meanwhile, if the volume of river water exceeds the
criteria, it will be dangerous because rubber boats are difficult to control (Muliana
et al., 2014).

Measurement of river discharge is carried out using the Velocity method, which uses
the principle of measuring cross-sectional area and flow velocity. The measurement
of the cross-sectional area is carried out using a stick or measuring cable, and then
the depth and width of the river are measured. The measurement of flow velocity is
carried out using a Current Meter with a propeller type. After obtaining the values
from the cross-sectional area and flow velocity data, the final results are obtained in
the form of river discharge values. Discharge measurements were carried out at
three points along the rafting route, starting from the start point to the finish line.

Waterfall Tourism
Waterfall tourism is included as a type of nature tourism that has considerable
potential to be developed. This is because the beautiful view of flowing water is able
to attract tourist visits (Febryan et al., 2021). In this case, land use is one of the
supporting factors for waterfall tourism. Land use can be used as an effort to predict
the impact caused by changes around the waterfall tourist location. In addition, the
slope of the slope is also a determining factor for the existence of waterfall tourism.
In areas that are on hills, have steep slopes, and are flowed by rivers, they have great
potential for waterfalls (Rahman and I Putu, 2018).

The parameter used in evaluating the potential for natural recreation services in the
form of waterfall tourism is the physical setting. The physical setting is a reciprocal
relationship between humans and their environment with the boundaries where the
community is located, which includes buildings, roads, and tourist attractions
(Rahman and I Putu, 2018). In this study, the observed physical setting data
consisted of surrounding land use.

Camping
Camping is one of the natural recreational activities that can be carried out with the
aim of enjoying the ecosystem services provided by nature. In accordance with
Rinaldi's statement (2015), camping is a recreational activity that is usually carried
out by a group of people with the aim of enjoying the beauty of nature and calming
down. In a location that is used as land for setting up tents, it must have several
provisions, such as ramps, supporting vegetation, and paths that can facilitate
access. The parameter used in evaluating the potential availability of recreational
services in the form of camping is land use. At the camping ground location, it was
observed regarding land use and the vegetation that covered the location.

Panorama
The use of forests as an outdoor-based tourist area has made a major contribution
to creating an advanced economy for residents living around the area (Evayanti et
al., 2022). Usually, protected forest areas function as nature tourism, and production
forests function as ecotourism. The parameters used to evaluate recreational
 services in the form of natural scenery include landscapes that can be enjoyed by
tourists from that location.
● Analisis Bencana (Kekeringan, Longsor dan Banjir)
Land Use Assessment
This research defines land use as the main factor in determining the potential of
recreational services in an area. The land use used is based on the four scenarios,
including the BAU land use scenario, the land use scenario based on the RSP, and the
land use scenario based on the Land Capability Class (LC). Assessment of land use
for recreational services is carried out by scoring each of the land use scenarios
found. The determination of the score refers to research conducted by Liu et al.
(2019), which found that the score given is related to the capacity of the land to
support ecosystem services in the form of nature recreation. Types of land use that
experience a lot of human intervention get a low score, such as paddy fields and
built-up areas that have a low capacity to provide nature recreation services,
compared to land that has not experienced or has had little intervention. In addition,
the assessment of land use also uses references to the literature that relates to land
use for recreational services. The following are the criteria for assessing land use
(Liu et al., 2019).

Table 9. Score of Land Use (Liu et al., 2019)

Land Use Score
Built-up Area 1
Agricultural 2
Shrubs Area, Grassland 3
Vacant Land 4
Nature Forest, Body Water : Lake 5

Slope Assessment
Slope assessment for recreational services requires DEMNAS data to determine land
slope. In this case, the land slope is divided into five classes consisting of slopes of 0-
8%, 9-15%, 16-255, 26-40%, and <40%. In Table 10, it can be seen that the steeper
the slope of the land, the smaller the score. This is because the steeper a piece of
land is, the more dangerous it is if it is allocated as a natural recreation location,
including the danger of landslides (Arifia et al., 2022). In this assessment, it is
assumed that the greater the slope of the land, the denser the contour lines (Rani et
al., 2020). The following is a table that contains land slope classes and scores.

Table 10. Score of Slope (Rani et al., 2020)

Rate of Slope (%) Score
0-8 5
9-15 4
16-25 3
26-40 2
>40 1

Analysis of Recreation Service Potential Map

 Spatial analysis of recreational services is an analytical process carried out using the
Ecosystem Services Index (ESI), which is the result of an assessment of land use and
land slope. The following is the formula for the Ecosystem Services Index (ESI) to
determine the potential for recreational services (Febriarta and Roza, 2020).

Notes :
ESI = Ecosystem Service Index
ESI a = Ecosystem Service Index Slope
ESI b = Ecosystem Service Index Land Use
Max = The maximum value of the calculation of the multiplication result and the
root of the Ecosystem Services Index is the slope of the land and land use

Calculation of this formula can be done using the ArcGIS Calculator, and then the
results of these calculations are grouped so that the distribution of potential
recreational services can be identified at very low, low, medium, high, and very high
levels with the following interval values.
Table 11. Ecosystem Service Potential Classification
Value Intervals Notes
1,00-1,80 Lowest
1,81-2,60 Low
2,61-3,40 Moderate
3,41-4,20 High
4,21-5,00 Highest

● Ketersediaan Mata Air

Result and Discussion

1. Spatial Classification
● Peta Hutan (Hutan Produksi dan Hutan rakyat) dengan GEE

● Hasil transformasi indeks vegetasi dan mega-composite

● Peta Hutan Produksi dan Hutan Rakyat

● Titik Observasi dan Data Hasil Groundcheck

● Peta Land cover Hutan Produksi dan Hutan Rakyat

2. Local Ecological Knowledge
3. Ecosystem Services Impact
● Potensi Hutan Produksi dan Hutan Rakyat untuk Eco-tourism

● Resiliensi terhadap bencana

● Ketersediaan Sumber Mata Air