By comparing past recordings to the temporal data acquired, remote sensing can determine regions affected by the lava slide. This data includes land dynamics. Relief extraction and geomorphology are two more analyses that make use of spatial topographical data in the form of a raster data model known as the Digital Elevation Model (DEM). Additionally, the application of DEM data in the field of numerical and static topographic analysis is expanded by the raster data model (distributed model) on altitude data in several studies ((Boyong et al., 2019); (Jenson & Dominque, 1988)). One of these studies is the prediction of rain-triggered lahar slides. Geological maps are another source of spatial information that can be used to ascertain the lithological conditions in the research area.

1. Sentinel-2 data

Sentinel-2 data is optical remote sensing data that describes the condition of the land before and after the eruption. The condition of the land before the eruption was obtained from the recording on April 20, 2021, while the condition of the post-eruption land was obtained from the recording on April 5, 2022. Besides being used as an identification of the impact of the lava slide, the Sentinel-2 image after the eruption was also used for the identification of morphoprocesses and morpho-arrangements for the manufacture of geomorphological information. The product used is the Sentinel-2B Multispectral Instrument Image with a spatial resolution of 10 meters, which was acquired temporally. The Level 2-A product was chosen as research data with geometrically corrected specifications and has a Bottom-of-Atmosphere (BoA) spectral correction level. Spectral correction in optical remote sensing images can improve display quality, especially when using two different recording images (Danoedoro, 2012).

2. National Digital Elevation Model data

National Digital Elevation Model Data (DEMNAS) is an Indonesian DEM product that has high spatial resolution. This data is classified as Digital Surface Model (DSM) data, which still includes objects above the ground, such as vegetation and buildings. The National DEM was generated from a combination of several SAR datasets, including IFSAR data (5m resolution), TERRASAR-X (5m resolution), and ALOS PALSAR (11.25m resolution), supplemented with mass point data obtained through stereo-plotting. This processing resulted in a DEM with a spatial resolution of 0.27 arc-seconds or 8.1 meters, provided in GeoTIFF format with a brightness level of 32-bit float. In this study, DEM data is used without any filtering or conversion into Digital Terrain Model (DTM), considering that the research area consists of bareland-specifically, extensive lahar flow zones that are larger than the spatial resolution of the DEM. The DEM data were used for watershed extraction, classification of morphology and relief for geomorphological maps, and identification of flow accumulation.

3. Geological Map and Digital Earth Map of Indonesia

Lithological information was used as the basis for the creation of the geomorphological map, specifically in the form of morphochronological data. The geological map utilized was based on remote sensing image interpretation, produced by the Geological Survey Center of the Ministry of Energy and Mineral Resources (ESDM) in 2013. The maps used in the process of identifying geological materials in the Rejali Watershed area consisted of sheets with a scale of 1:50,000, including Ampelgading, Lumajang, and Tanjungan. The base map information was obtained from the Indonesian Digital Elevation Model Map at a 1:25,000 scale, provided by the National Geospatial Information Agency of Indonesia in 2021.

The field survey aimed to present an overview of actual field conditions, guided by a tentative hybrid map developed before fieldwork. This map was constructed by considering slope classifications, material types, and process identification, which were heuristically grouped through the interpretation of temporal imagery and validated by expert judgment. Additional information regarding the study area could be easily gathered during the field survey activities ((Otto & Smith, 2013);(Smith et al., 2011)). Another purpose of the field survey was to collect physical data on lahar material, validate and conduct ground-checking, and identify the impact of lahars on disasterrisk elements, both physical and other environmental components.

The field survey was conducted using purposive sampling based on the geomorphological units of the distal zone affected by lahars in the Rejali Watershed area. Geomorphological units were chosen as the unit of analysis because they represent the surface features and processes occurring within them, offering a useful framework for analysis and visualization. Geomorphological units classify the morphological features of the region, materials, and processes (time), and are widely used as analytical units in various studies related to lahar flow identification (Lavigne et al., 2016)(Pavlopoulos et al., 2009)(Procter et al., 2021).

1. Ground check and update geomorphology condition

Validation and ground-checking were carried out based on the previously created tentative geomorphological map to update the boundaries of land units. The identification of rain-triggered lahar impacts was conducted to assess the damage caused by the lahar flows. The identification activities carried out included: (a) morphological identification based on slope gradient information obtained through direct measurements using a distometer, (b) material identification, particularly in areas affected by rain-triggered lahars indicated by the presence of newly deposited materials, and (c) identification of the dominant geomorphic processes occurring in the field.

2. Identification of existing rain-triggered lahar characteristics

As a key parameter related to lahar flow, the characteristics and properties of the lahar were identified in detail in the field. The field survey of lahar characteristics included: (a) identifying existing lahar flow paths using cross-sectional measurements at two affected locations, utilizing GPS plots and a distometer, (b) measuring lahar thickness using stratified random sampling at the research site, (c) qualitatively identifying and measuring the percentage of lahar material size in the field, and (d) collecting lahar deposit samples for sediment property and granulometric analysis through purposive sampling based on geomorphological units.

3. Identification of rain-triggered lahar impacts

Field observations also examined the extent of damage caused by the rain-triggered lahar flow. The impacts were categorized into: (a) effects on physiographic and geomorphological conditions, and (b) effects on exposed elements such as settlements, road access, plantations, and others. Observations were conducted by plotting the affected locations and qualitatively defining the resulting damage.

The geomorphological approach is a comprehensive classification of land units. In this study, the mapping of geomorphological land units was identified using spatial data, remote sensing, and validated through ground checks and field measurements. According to (Pavlopoulos et al., 2009), geomorphology is an approach that classifies land based on four main categories: morphology obtained from DEM data and relief, morphochronology, derived from geological data, morphoarrangement, and morphoprocesses, based on heuristic remote sensing interpretation. The geomorphological boundaries resulting from heuristic interpretation possess unique characteristics, with tangible boundaries in the form of relief. These boundaries are not the result of the direct overlaying of several parameters, thereby producing representative data that is easy to use (Sartohadi et al., 2014).

1. Morphological condition extraction

The analysis used to obtain morphological information includes the extraction of slope steepness and hillshade. Digital Elevation Model (DEM) was used for this analysis and classified based on Desaunetes’s classification system (Desaunettes, 1977). This slope classification system is representative of actual field conditions Table 1. The On-Screen Image Interpretation (OSII) digitization process, based on slope and hillshade extraction results, is used to generalize morphological conditions to obtain representative boundaries based on terrain relief.

2. Heuristic visual interpretation

All spatial information used to identify geomorphological units was interpreted visually using the On-Screen Image Interpretation (OSII) method. This method accommodates the geomorphological concept heuristically by integrating morphological data (slope gradient and hillshade), morpho-arrangement and morphoprocess (satellite imagery), and morpho-chronology (lithological information based on geological maps). These data layers were analyzed simultaneously based on expert judgment in the field of geomorphology. Furthermore, manual digitization allows for the delineation of boundaries between land units based on topographic conditions that are easily recognizable in the field.

3. Re-interpretation and updating of geomorphological boundaries

Information obtained from field validation conducted during the field survey stage was subsequently used to update the boundaries of geomorphological units. In addition, supplementary information collected during field data acquisition was also utilized to update the attributes of each geomorphological unit. The output of this stage is a draft of the final geomorphological map of the Rejali watershed area following the 2021 eruption of Mount Semeru.

1. Mapping the existing slide zone

The lahar flow zones in the Rejali Watershed area were spatially identified using temporal remote sensing data from Sentinel-2B, which acquired conditions before and after the Semeru eruption in 2021. The delineation of boundaries was carried out visually and digitally by observing changes in land cover around rivers and floodplains in the Rejali Watershed. The lahar flow zones identified through remote sensing provided tentative spatial information that was subsequently validated in the field. Cross-sectional profiles, derived from field measurements of the affected areas, were used to correct the delineation of the lahar flow boundaries on the tentative map. This correction was necessary due to the time difference between image acquisition and the field survey.

2. Lahar thickness point interpolation

The lahar thickness data obtained from the field survey consists of point data with reference coordinates. The requirement for thickness information in this study pertains to the thickness conditions across the entire affected area (distal zone) of the Rejali Watershed. Based on this need, spatial analysis through interpolation was performed to determine the thickness values at surrounding locations. Previous studies ((Pasaribu & Haryani, 2012);(Pramono, 2008)) recommended the Inverse Distance Weighted (IDW) interpolation method due to its ability to provide good accuracy for topographic and sedimentation studies. This method assumes that each elevation point exerts a local influence that diminishes as the distance increases (Azpurua & Ramos, 2010). Equation 1for IDW interpolation, as proposed by (Azpurua & Ramos, 2010), is as follows:

(1)

Where Z^* represents the estimated value (thickness), Z_i is the known elevation value at point i among N known points, and ω_i is the weight value determined by the following equation 2:

(2)

where p is the power parameter that determines the influence of each point (this study uses a power value of 2), and h_j is the distance between the interpolated point and the known elevation point, which is calculated using Equation 3:

(3)

where (x, y) are the coordinates of the point to be interpolated, and (x_i, y_i) are the coordinates of the known thickness location. A total of 196 thickness points obtained from field measurements (direct measurements, measurements of lahar-affected objects, and scaled field photographs) were used as input for the interpolation process to generate the distribution of eruptive material thickness across the entire study area. The interpolated data were then trimmed according to the lahar flow boundaries identified from the existing flow zone analysis.

3. Lahar granulometry

Granulometric analysis of lahars was conducted to determine the grain size distribution, processed statistically, and to reflect the properties of the lahar. Laboratory analysis was performed to obtain the texture size of lahar deposits from six lahar sediment samples collected during the field data gathering. A 100-gram sample was taken through splitting to reduce subjectivity in the samples used for sieving, which was performed using a series of sieves with the sizes presented in Table 2.

The weight of the sand in each sieve fraction was then converted into a PHI scale using the cumulative curve Figure 2 for statistical calculations of the sediment. This stage was carried out to determine the sediment properties at each observation site, focusing on average grain size (mean), grain size uniformity (sortation), skewness of grain size distribution, and the kurtosis or peakedness of the grain size distribution. The weight line derived from the cumulative curve was used to identify the sediment transport mechanisms, which are classified into: (a) traction (materials transported by rolling along the riverbed), (b) saltation (materials transported by bouncing along the riverbed), and (c) suspension (materials transported while floating in the water column above the riverbed).

The calculation of granulometric statistical parameters is based on the intersection between the cumulative curve and particle size (Φ). The statistical parameters are presented as Equations 4, 5, 6, & 7.

(4)

(5)

(6)

(7)

1. Updating DEM data based on the thickness of the lahar slide

The changes in geomorphology in this study focus on the topographic alterations caused by the covering of land by lahar flow material. The results of the lahar thickness interpolation from the previous identification stage were used as the topographic change values in the DEM data. This was done using spatial analysis in the form of a raster mosaic with the "sum of values" option, resulting in a DEM with lahar thickness information.

2. Flow Accumulation as glide prediction

Flow accumulation is a raster-based spatial analysis that models the potential flow based on the accumulation/number of pixels influenced by surrounding pixels. The use of flow accumulation information in the modeling of lahar flows serves as the basis for determining the flow direction, as it is related to the identification of river channels as lahar transport pathways ((Aisandy & Sukojo, 2016); (Cando-Jácome & Martínez-Graña, 2019); (Lee & Lee, 2015)). The input data for this analysis, in the form of DEM, is first extracted to provide flow direction information through spatial flow direction analysis. The flow concentration generated from each pixel will be directed toward a specific pixel and will accumulate as the number of contributing pixels increases. Figure 3 and Figure 4 illustrate the schematic modeling of DEM data, flow direction, and flow accumulation.

A higher flow accumulation value indicates that the location/pixel has a larger accumulation of flow. In relation to the research activities, the potential for lahar flows can be detected using the flow accumulation model, which directs the flow out of the identified existing lahar flow zone. The threshold value used for predicting lahar flows in this study is a flow accumulation value greater than 300.

The distribution of lahar materials, identified through remote sensing observations and field surveys, was categorized into two types: eroded zones and deposited/sedimented zones. Eroded zones are characterized by horizontal widening of riverbanks and increased depth of river valleys vertically, due to the impact of lahar materials (Hadmoko et al., 2018). These zones are marked by locations with higher elevations (riverbanks and bottoms with significant height differences) than the peak of the lahar flow, leading to erosion of riverbanks by lahar materials (Kurnianto, 2019). Sedimented/deposited zones represent areas covered by lahar materials, characterized by locations with lower elevations than the peak of the lahar flow ((Kurniawan et al., 2019); (Masitoh et al., 2019)). Figure 8 illustrates the different types of impacts from rain-triggered lahars validated through field profiling.

Field profiling was also used as quality control for the identification results of lahar boundaries using remote sensing (Mutaqin et al., 2022). Based on the results of cross-section mapping of the lahar-affected areas at two locations, there was a difference in length/horizontal distance between field data collection and boundary delineation in the imagery Table 6.

Based on the zoning identified using remote sensing, there was a difference in the area of lahar material distribution before and after the eruption, both in the erosion zone and the deposition zone. Based on the area calculation of pre-eruption imagery, the lahar material in the pre-eruption condition was identified to cover an area of 1640.01 ha. Whereas post-eruption imagery shows that the distribution of lahar material increased to 2305.15 ha (40.56%) Figure 9. Besides its temporal acquisition capability, remote sensing was able to identify areas that were inaccessible, such as steep slope regions ((Joyce et al., 2009); (Rijal, 2020)).

The distribution of the lahar deposition zone in the Rejali watershed, which had a massive impact, was located in the central area of the watershed, bordering the bottleneck formation of the Mandalika formation, characterized by the landform "Heavily Eroded Mahameru Volcano Lahar Valley". The Mandalika formation, with its retention properties and resistant rock structure, caused the concentration of lahar material transportation during peak discharge events (Widodo et al., 2018). Additionally, another phenomenon that caused the lahar flow was the increasing perpendicular meanders of the river to the flow direction, leading to the lahar material diverting and overflowing from the river channel (Lestari et al., 2019).

The distribution of the eroded lahar zone in the Rejali watershed, which had a massive impact, was located in the upstream area of the watershed before the bottleneck formation. The massive damage to the morphology included the erosion of the river bars with the landform unit "Heavily Eroded Foot Slope of Volcanic Lahar Fan Deposits 4 and 5 of Mount Mahameru" within the floodplain zone. This condition was caused by the accumulation of lahar material during peak discharge in the zone before passing through the bottleneck channel. The locations of the eroded and sedimented lahar material zones in the Rejali watershed are shown in Figure 9.

The lahar-affected zone in the Rejali watershed predominantly occurred in areas with relatively gentle slopes, where the surface material consisted of previous lahar flow deposits. This phenomenon demonstrates that the processes occurring in a landscape at present are indicative of events that took place in the past ((Cooke & Doornkamp, 1990); (Hazbavi, 2018)). In relation to disaster analysis, geomorphological landforms can be used as mapping units because they provide comprehensive information about a land area, including the dominant processes taking place.

The identification of lahar material thickness in the field was observed using several methods, including direct measurement, estimation of affected objects, and profiling measurements Figure 10. However, thickness information in the proximal zone, such as the upper and middle slopes of Mount Semeru, was not identified due to the difficult-to-access terrain conditions.

There were 196 measurement points obtained in the field, with the maximum lava thickness reaching 18.5 meters and the minimum thickness being 0.1 meters. Based on all the measurement data, the average thickness in the Rejali Watershed flow zone reached 4.1 meters. The thickness mapping process was continued with interpolation to obtain the spatial distribution of thickness using spatial analysis in the form of digital interpolation ((Adedapo & Zurqani, 2024); (Akbarurrasyid & Kristiana, 2020); (Ohashi & Torgo, 2012)).

The IDW interpolation method was used to produce the distribution of material thickness by considering the values between elevation points that decreased with increasing distance ((Adedapo & Zurqani, 2024); (Pasaribu & Haryani, 2012)). The IDW method was widely used to model topographic conditions with measured points because it could accurately predict values between points. The spatial distribution of lahar material in the flow zone is presented in Figure 11.

The results of the IDW interpolation produced a distribution of lahar material thickness ranging from 0 to 18.5 meters. Material thickness less than 3 meters was located throughout the lower middle zone to the downstream floodplain of the Rejali River, where flow energy began to decrease and carried little material. Lahar thickness less than 3 meters was also found in the bottleneck zone, which was the transition zone from the upstream to the downstream of the watershed. The minimal thickness of material in that zone was due to the watertight channel conditions, allowing all material to be carried to different zones. Material thickness between 3 and 12 meters was spread across the upper middle slope deposition area, bordering the bottleneck. This was caused by the concentration of material and energy that had passed through the narrow channel, resulting in a massive overflow phenomenon. Additionally, locations with similar thicknesses were sandbar areas eroded and buried by lahar material in the upstream region. Lahar material with thicknesses between 12 and 18 meters was identified at several points in the upstream area, marking the beginning of the distal zone. This condition was caused by a significant change in slope, leading to the deposition of lahar material before it was carried to the lower zones.

Six geomorphological unit locations were selected as observation zones, consisting of the upstream watershed area in unit (F/3/FL/4/S), bottleneck formation (F/11/VL/4/EB), middle watershed area (F/2/FL/4/S and F/1/FL/4/S), and downstream area (F/1/AP/14/S and M/1/AP/15/S). The qualitative observation results of the lahar material were presented in Figure 12.

Six geomorphological unit locations were selected as observation zones, consisting of the upstream watershed area in unit (F/3/FL/4/S), bottleneck formation (F/11/VL/4/EB), middle watershed area (F/2/FL/4/S and F/1/FL/4/S), and downstream area (F/1/AP/14/S and M/1/AP/15/S). The qualitative observation results of the lahar material were presented in Figure 12. The condition of lahar material in the "Wavy Floodplain Sedimentation of Mahameru Volcano Lahar Flow" zone (F/3/FL/4/S) was dominated sequentially by sandy material, gravel, pebbles, and bombs. This zone was a buildup of material before passing through the bottleneck formation. Bombs/large rocks with diameters greater than two meters were frequently found in this location, due to the landform being in the upstream area with wavy floodplain morphology, causing a reduction in flow energy. Additionally, the movement of bomb material was limited due to the bottleneck formation. Based on the condition and type of material size, zone F/3/FL/4/S had a relatively high hazard potential. This was because bomb-sized material had significant destructive power on geophysical aspects and human activity areas (Gomez & Lavigne, 2010).

The "Heavily Eroded Mahameru Volcano Lahar Flow Valley" zone /F/11/VL/4/EB, which formed the bottleneck, was sequentially dominated by lahar material such as pebbles, gravel, sand, and bombs. No sand material accumulation occurred at this location due to the concentrated flow that carried lighter materials and left larger and heavier materials like pebbles and bombs with diameters of 1-2 meters. It was classified as a landform with heavy erosion processes, as evidenced by the numerous landslide materials from the valley slopes (having a different color than the lahar material, which was Mandalika formation material) due to the high energy of the lahar flow. Based on the material conditions at location F/11/VL/4/EB, the site had high destructive power due to the concentrated lahar flow in the bottleneck zone.

Zone F/2/FL/4/S or "Wavy Floodplain Sedimentation of Mahameru Volcano Lahar Flow" was a floodplain zone adjacent to the lahar deposition/flow zone. Located in the middle watershed area after the bottleneck formation, it caused sandy material to dominate and overflow into the surrounding area as a flow zone. Gravel and pebbles sized 20-60 cm were frequently found near the river flow, while bombs had a lower percentage. Fine sand material began to be found at this location due to the morphology, which was classified as an undulating plain. The dominant hazard potential at location F/2/FL/4/S was the massive overflow of sandy material in the surrounding floodplain area.

Zone F/1/FL/4/S or "Floodplain Sedimentation of Mahameru Volcano Lahar Flow" was a geomorphological unit similar to zone F/2/FL/4/S with a flat plain morphology. Located in the middle watershed area bordering the downstream region, the lahar material that dominated this area sequentially was sand and fine sand, gravel, pebbles, and bombs. The pebbles found were 15-30 cm in size with rounded and blunt shapes. Additionally, there were ancient bomb materials (not from the 2021 Semeru eruption lahar), marked by their rounded and blunt shapes, pushed by new lahar material and concentrated at the edge of the floodplain in several locations with flat morphology. The material shapes became more rounded and blunt downstream as the transportation distance increased (Dana et al., 2016). Based on the type and size of material in zone F/1/FL/4/S, it had a low damage potential, supported by the decreasing flow speed and lahar material transport speed due to the gentle slope conditions.

The landform zone F/1/AP/14/S or "Aluvial Deposit Plain Sedimentation" was a landform in the downstream Rejali Watershed area. The lahar material that dominated this location sequentially included sand, fine sand deposits, gravel, pebbles, and bombs. Fine sand deposits had a larger percentage than the landforms above due to the downstream morphology, classified as a flat plain (Suriani et al., 2024). Gravel and pebbles were frequently found in sizes of 5-20 cm mixed/bound by fine sand deposits. The damage potential from lahar material in this zone was low based on the identified material conditions.

Zone M/1/AP/15/S or "Fluviomarine Deposit Plain Sedimentation" was located at the river mouth directly bordering the sea. The morphology was classified as flat, with sedimentation as the dominant process due to wave flow opposing the river flow. The most common materials were fine sand deposits, sand, and gravel with diameters of 5-10 cm. Pebbles were found in small percentages, while bombs were not found at this location. Based on the material conditions and flow speed, the damage potential from lahar flow at this location was small. The identified lahar material widening in the image at location M/1/AP/15/S was not horizontal riverbank erosion but an increase in sediment material in the river body and floodplain.

As a complement to the qualitative description of lahar material properties, laboratory-based quantitative analysis was conducted to illustrate the variation in sediment grain size across zones representing the upper, middle, and lower parts of the study area. The upper zone, defined by F/3/FL/4/S, was dominated by medium sand-sized material (mean = 1.40 Φ), formed due to flow deceleration and intense material collisions before entering the bottleneck area. In the middle zone, represented by F/2/FL/4/S and F/1/FL/4/S, the average grain size was categorized as coarse sand (mean = 0.95-0.90 Φ). This condition indicates a mixture of saltation-and suspension-transported materials from the upper zone, with flat morphology enabling semi-fluid deposition. Meanwhile, the lower zone, represented by F/1/AP/14/S and M/1/AP/15/S, was characterized by coarse to very coarse sand-sized material (mean = 0.23-0.21 Φ). The larger grain sizes in the lower zone resulted from sedimentation on flat terrain with low flow energy, along with the dominance of fine fractions such as very fine sand and silt.

Granulometry analysis was used to measure the grain size of sediment to obtain its properties and characteristics quantitatively ((Htun et al., 2020);(Sayudi et al., 2014)). The material was described based on the mean grain size value to identify the grain size representing the entire sample (Boggs, 2012). Sortation was used to determine the uniformity of sediment grain size in rock sorting, with values obtained from the standard deviation (Mir & Jeelani, 2015). The standard deviation value influenced the sorting value of a rock (Sasmito et al., 2018). The larger the standard deviation obtained, the poorer the sorting of the rock, and vice versa.

The following parameter used was skewness, indicating the skewness level of a curve that described the coarseness/sharpness of the sediment material. Positive skewness indicated smooth-surfaced grains ((Htun et al., 2020); (Putra & Nugroho, 2017)), while negative skewness indicated grains dominated by rough surfaces (Rozamuri & Hidayat, 2016). Kurtosis describes the frequency distribution and grain size distribution in the analyzed sediment material. Kurtosis classifications included platykurtic, where sediment size distribution was evenly spread, mesokurtic, where sediment material was dominated, and leptokurtic, where significant material dominance existed (Suriani et al., 2024). The cumulative percentage curve from the granulometry analysis could also classify the sediment transport process: traction (material rolling on the river surface), saltation (material deposited by hopping), and suspension (material carried by water currents). Granulometry analysis results were presented in Table 7, while cumulative percentage curves for each location were presented in Figure 13.

Legend: (a) mean = CS: coarse sand, MS: medium sand, (b) sorting = MdS: moderately sorted, VWS: very well sorted, (c) skewness = VF: very fine, S: symmetrical, (d) kurtosis = VP: very platykurtic, M: mesokurtic.

The sand at F/3/FL/4/S zone was categorized as medium sand with an average size of 1.40 Φ. The current speed caused this condition (Rachman et al., 2021) Intensive collisions between materials passing through the bottleneck result in the accumulation of finer material. The mixing of materials in the upstream area indicated non-uniform material conditions, described by the moderately sorted sorting value. Intensive collisions between materials resulted in uneven material surfaces, as indicated by the symmetrical skewness value, and the dominance of material in a deposit was not uniform (mesokurtic). Based on the cumulative percentage curve, the dominant transport process in the F/3/FL/4/S landform was suspension (material dissolved by water), indicated by strong current speeds and high destructive power.

The sand deposits at F/11/VL/4/EB zone were categorized as coarse sand based on the obtained mean value of 0.85 Φ, consistent with field observations where the dominant material was coarse. The concentration of flow in the Mandalika formation valley area resulted in a material size distribution that was not entirely uniform but evenly spread (platykurtic). The material surface was smooth due to erosion passing through the massive river walls. The material transport process was still dominated by suspension, where water discharge and current speed in the bottleneck zone were high.

Sand deposits at location F/2/FL/4/S were categorized as coarse-textured sand with a size of 0.95 Φ. The grain size was not entirely uniform but evenly spread (platykurtic) in the study location. The material surface was very smooth, giving it high sliding power. The dominant transport process was suspension, making the material semi-fluid and easily overflowing from the floodplain. The saltation process at location F/2/FL/4/S started to increase due to relatively flat slope conditions and reduced flow velocity. Material characteristics at zone F/1/FL/4/S were almost the same as those of the landform above it, F/2/FL/4/S. It had coarse sand texture material with a size of 0.90 Φ, not entirely uniform. The study location material was evenly spread with slightly rounded (smooth-surfaced) material. This was due to the increasing saltation process, which accounted for approximately 47%. However, the suspension process still dominated in the F/2/FL/4/S area, which was the middle part of the watershed area.

Landform F/1/AP/14/S is located in the downstream watershed area with a sloping plain. The material condition in this location was coarse-textured sand with uniform grain proportions. The deposit material properties were evenly spread in the observation area, characterized by smooth material surfaces. The dominant transport process in the downstream area was saltation due to the relatively calm and non-destructive river current (Gomez et al., 2018).

Lahar material conditions in the M/1/AP/15/S landform were categorized as coarse sand deposits with uniform grain size proportions evenly spread in the M/1/AP/15/S location. Characterized by very smooth grain surfaces with a dominant saltation process of 92%, lahar material damage was found in the downstream zone of the Rejali watershed due to very flat morphology conditions with low water flow energy.

The terrain slope gradually decreases (as indicated by the decreasing morphometric values in the geomorphological unit), particularly after passing through the bottleneck area of the Mandalika Formation. This slope reduction leads to a significant decrease in flow energy, thereby promoting the deposition of fine particles such as sand in flat areas. Gentle slopes below 8% are present in the downstream areas (for example, in land units F/1/AP/14/S and M/1/AP/15/S). Thus, the tendency for fine layering in the downstream zone is closely related to hydrodynamic deceleration and reduced transport capacity, reaffirming that slope gradient and flow energy are also key factors in the distribution of lahar sediments.

Based on the characteristics of the rain-triggered lahar after the eruption of Mount Semeru in 2021, it provided an overview of the processes, mechanisms, properties, and spatial distribution of deposition zones. The rain-triggered lahar phenomenon that occurred in the Rejali watershed also intersected with human activities, causing losses (Anwar et al., 2021). The encounter of the hazard phenomenon with vulnerability factors produces losses of property, assets, and lives, which defines it as a disaster (Schneiderbauer et al., 2017). In this context, the disaster is the rain lahars. Several objects affected by the rain lahars disaster in the Rejali watershed, based on field data collection, could be categorized as damage to infrastructure, industry, and settlements.

The impact of rain-triggered lahar disaster damage on infrastructure in the Rejali watershed included the destruction of the Sabo DAM, which was used to restrain the lahar flow, and the disruption of transportation facilities such as roads and bridges. The Sabo DAM and the disrupted road network were located in the same area, namely Supit Urang Village, Pronojiwo District (Wahyuningtyas et al., 2021). The condition of the damaged objects is shown in Figure 14.

The Sabo DAM and the disrupted road network were located in the upstream area of the Rejali watershed, bordered by a geomorphological bottleneck formation in the landform unit F/3/FL/4/S. The damage to the Sabo DAM was characterized by the destruction of the lahar material retaining embankment due to the concentration of lahar flow energy passing through the Mandalika rock formation valley. The damage to the highway access involved the disruption of the road crossing the floodplain of the Rejali River, which was one of the alternative routes connecting Malang Regency and Lumajang Regency (Thouret et al., 2022).

The location that experienced massive lahar flow impacts was Sumberwuluh Village, Candipuro District, Lumajang Regency. The impacts included the disruption of the bridge transport network and the burial of settlements and industries in areas adjacent to the Rejali watershed floodplain (Bachri et al., 2023). The damage in Sumberwuluh Village is presented in Figure 15.

The high concentration of lahar in the upstream area was transported to the middle area of the watershed through the Lahar Flow Valley landform zone (F/11/VL/4/EB) with material characteristics of impermeable rocks. This location was the only channel connecting the upstream area to the middle area of the watershed, forming a bottleneck. During the peak discharge of lahar flow, the concentration of flow caused damage to the "Gladak Perak" bridge. This bridge was the main connecting route between Malang Regency and Lumajang Regency (Putra et al., 2022).

The sandy material passing through the lahar flow valley zone had high energy and frequency, causing overflow in flatter zones located in the eastern part of the floodplain (landform unit V/5/FP/6/S). The impacts included the burial of settlements and industries with an average lahar material thickness of 6-8 meters. The impact of the rain lahars in the middle to downstream areas of the watershed resulted in material deposits that did not cause damage. This condition was due to the low energy of the water flow, attributed to relatively flat relief and morphology classes.

Based on the thickness and grain characteristics of the lahar material collected in the field, it was found that the infrastructure damage zone, such as the Sabo DAM and bridge in Supiturang, was buried under lahar deposits ranging from 5 to 10 meters in thickness. The dominant material in this area consists of large boulders (bombs) and coarse gravel, as identified in the landform units F/3/FL/4/S and F/11/VL/4/EB. Both zones are characterized by an average grain size ranging from coarse to very coarse sand (Φ 0.85-1.40) and a significant proportion of solid materials with diameters exceeding 1 meter. The combination of high material thickness and large grain size indicates a highly destructive energy, which led to structural failure of the dam and bridge.