Fluctation moisture properties in the application of soil dielectric measurement in the North Bengkulu District, Indonesia

. Information about soil moisture in the root layer is very important in evaluating water availability for plants, erosion rates and various other hydrological characteristics. Previous research has found a technology for measuring soil dielectric properties as a soil moisture estimator that can be carried out quickly in the field. The purpose of this study was to apply dielectricity technology in estimating the characteristics of soil moisture in situ in the field at the North Bengkulu in Bengkulu Province. Five land units in North Bengkulu Regency, Indonesia, have been selected as experimental sites. At each location, the soil profile was excavated to a depth of 100 cm, the basic characteristics of the soil in each layer were described in the field and analyzed in detail in the laboratory. The variables measured in the field are the measurement of the dielectric properties of the soil, the effective depth of the soil, the horizon and the characteristics of each horizon. The variables analyzed in the laboratory were texture, organic matter in each layer, aggregate stability at 0-10 cm layer, field soil moisture at the time of measurement, and groundwater characteristics. The relationship between the data on the basic characteristics of the soil with the soil moisture balance variable in the laboratory analysis and measurements of dielectric properties in the field, was then determined using regression analysis. The results showed that the five locations studied had different profile characteristics and groundwater fluctuations. Of the four independent variables tested (percentage of sand, silt, clay and organic carbon) then the content of sand and clay can be used as predictors of the water content of field capacity. Because the four independent variables have a very weak correlation with permanent wilting point water, the available water content can only be predicted from the sand and clay content.


Introduction
Information about soil moisture or moisture content in the root layer is very important in evaluating water availability for plants, erosion rates and various other hydrological characteristics.Groundwater is one of the physical properties that directly affect plant growth and other aspects of human life because water is the highest component of plants and other living things.The ability of the soil to store water is an indicator related to the criticality of the land in supplying water for plant growth, as well as absorbing rainwater that falls to the surface so as not to cause erosion and flooding.On a laboratory scale, the research team has managed to find a close relationship between several basic soil properties, such as texture and organic matter, and soil moisture content available to plants.
Soil moisture prediction models in the field at the district scale can be developed by looking for the relationship between water balance and basic soil properties such as texture and organic matter.However, when carried out on a large area scale, studies of the correlation between components of groundwater balance, such as moisture at saturation and permanent wilting points, and the basic characteristics of these soils require a very large number of samples.Our previous research has obtained technology for measuring soil dielectric properties as a soil moisture estimator that can be carried out quickly in the field even though it is only on a limited scale in one plantation area.
The specific purpose of this research is to apply dielectric technology in estimating soil moisture characteristics in situ in the field at the district scale in Bengkulu Province.The urgency of this research is the need for a model to predict soil moisture characteristics that have been tested at the district scale.The benefit of the research is the availability of information on soil moisture characteristics for the benefit of land management related to land suitability for certain commodities, as well as irrigation and erosion control measures that need to be carried out on agricultural land.

Tools and materials
The tools that will be used in this research are as follows: • Geographical information system (GPS), to determine the geographical location of the soil sampling point so that the data from the groundwater profile analysis can be mapped digitally.• Pressure plate apparatus, used to measure soil moisture content at a certain pressure, which describes the amount of water available to plants under conditions of environmental stress.• Laboratory equipment for texture analysis (pipette, soil solvent tube, electric stirrer, etc.) and soil organic matter content (Walkley & Black method).• Measuring device for soil water content (moisture meter) from previous research (Figure 1).Materials needed sensor cable to measure soil moisture content to a depth of 100 cm, hoe, fork and spade (for soil sampling), plastic bags to store soil samples, chemicals to be used for soil analysis (texture and materials) organic) in the laboratory, as well as other consumables that will be used for sampling and soil analysis.

Research location
The location of this research includes five areas assisted by the Agricultural Extension Center (BPP), namely BPP Air Napal, BPP Kerkap, BPP Hulu Palik, BPP Armajaya and BPP Argamakmur (Figure 2).The five selected locations represent the dominant land uses in North Bengkulu Regency (rubber, oil palm and horticulture), as well as the soil types that dominate the district (Ultisol and Inceptisol).Determination of the research location is carried out in stages, starting from coordination with the Regional Apparatus Organization (OPD) of North Bengkulu Regency in charge of food crops, horticulture and plantations, as well as counseling.Furthermore, the OPD in charge of counseling contacted the head of the BPP who was the candidate for the research location.The result of the discussion between the research team and each BPP leader was an agreement on sampling points and observations from April to November 2019.Fig. 2. Dielectrometer instrument from previous research modified to measure the availability of ground water for plants.

Soil profile observation
Soil profile observations were carried out in five selected locations, each on the land of farmers assisted by BPP Air Napal (old oil palm), BPP Kerkap (melon), BPP Hulu Palik (immature oil palm), BPP Armajaya (20 years rubber) and BPP Argamakmur (10 years rubber) North Bengkulu Regency (Figure 3).The soil was excavated with dimensions (length, width, depth) of 1 x 1 x 1 m, the vertical section of the soil was described quantitatively and qualitatively.The profile description includes the boundary and horizon thickness, soil color, texture, structure, consistency and plant roots in each layer.Characteristics of vegetation cover (such as staple and understory plants) were also observed around the excavated profiles.

Soil sampling
After observing the profile, the soil surface on the side of the profile was leveled, intact soil samples were first taken using a stainless ring with a diameter of 7.4 cm and a height of 4.0 cm at a depth of 0-10 cm, the soil on the surface and bottom of the ring was leveled with a knife, then both sides are covered with a plastic cap.Since the soil at the top and bottom of the ring was removed, the intact soil samples in the ring represented a depth of about 3 to 7 cm.At the same depth, about 500 g of disturbed soil samples were taken compositely at a depth of 0-10 cm and then put in a plastic bag and tied so that the water in the soil sample did not evaporate.The ex-sampling soil on the profile side was excavated and leveled to a depth of 10 cm, the process of sampling intact and disturbed soil was repeated.Soil sampling as described above was carried out to a depth of 90-100 cm (Figure 4).

Soil analysis
Two days after arriving at the Soil Science Laboratory, Bengkulu University, about 50 g of disturbed soil sub-samples were taken compositely, weighed and then dried in an oven at 1050 for 24 hours and weighed again to determine the soil moisture content under field conditions when sampling was carried out.The whole soil sample in the ring was weighed, after deducting the weight of the ring, then the weight of the soil in the ring was converted to an oven-dry equivalent weight of soil using the moisture content of the field soil, then divided by the volume of the ring to get the value of the weight of the soil volume.A total of 50 soil samples in the ring were then sent to the Soil Physics Laboratory of the Soil Research Institute in Bogor for determination of specific gravity, groundwater characteristics and pore size distribution.
Disturbed soil samples were dried, grown and sieved with a 2.0 mm mesh sieve for texture analysis (three fractions) and organic carbon in each layer.For samples at a depth of 0-10 and 10-20 cm, the ground that has been ground was sieved with an 8.0 mm sieve and accommodated with a 4.76 mm sieve for the purposes of analyzing the stability of the aggregate.Approximately 100 g (oven dry equivalent) of 4.76 to 8 mm aggregate are placed on 4.76 mm sieves which are stacked on 2, 1, and 0.25 mm sieves.The soil sample was sifted in water for three minutes at a speed of 35 swings per minute and a swing height of 7.5 cm.Aggregate stability is calculated based on the proportion of aggregate weight that is accommodated in each sieve after sieving.

Measurement of soil moisture in the field
Twenty pairs of wires were inserted into the ground adjacent to the described profiles of 5, 10, ... 100 cm respectively, the lower end of 5 cm long was stripped first so that the copper was in contact with the ground.The process of inserting the sensor cable into the ground is shown in Figure 5A.This pair of cables is used for measuring soil moisture content at a depth of 0-5 cm.The following pairs of cables are each inserted at intervals of 5 cm to a depth of 100 cm so that there are 20 pairs of wires in each soil profile (Figure 5B).Dielectric properties in the form of ground electrical impedance (Z, units of kilo ohms) are measured by connecting the cable section that appears at the ground surface with a Dielectrometer tool developed from previous research (Figure 5b).The Z value is then changed to soil water content (θ, unit g/g) using the model, namely: where the constants a and b were determined by performing a regression analysis between the values of Z and of disturbed soil samples on the same measurement day.Z measurements were carried out twice a week to describe the redistribution of precipitation water in the soil profile during the drying period (discharge phase, April-July 2019) and wetting period (recharge phase, August-November 2019).

Conversion of electrical impedance value to ground moisture content in the field
Soil profile observations and soil sampling (whole and disturbed) were carried out at 10 cm depth intervals, and simultaneously measured electrical impedance at 5 cm depth intervals.These data are used to calibrate the Dielectrometer in converting the impedance Z value to the water content.The results of tool calibration are presented in Figure 7 so that Equation (1) becomes.
E3S Web of Conferences 373, 06002 (2023) https://doi.org/10.1051/e3sconf/202337306002ISEPROLOCAL 2022 Fig. 6.Dielectrometer calibration results to obtain constants a and b in Equation ( 1) using the results of measurements of electrical impedance and water content in the field gravimetric method in the laboratory.

Description of research location
The field research locations in 2019 covered 5 coastal areas and tropical forest buffers spread across the North Bengkulu Regency.All research areas have been described in detail which include geographical position, village, sub-district, altitude above sea level, and land units as presented in Table 1.The land observed and sampled was agricultural land located in cultivation areas, such as land owned by Agricultural Extension Center (BPP) and community-owned land.Administratively, the land under study covers 21 village areas, 18 sub-districts, and 3 districts, stretching from the north to the Bengkulu City area.The research location included in the coastal area is located at an altitude of about 20 to 50 m above sea level, while which includes tropical forest buffer areas occupy an altitude of 50 to 200 m above sea level.
Based on the aspect of land use, the research location is agricultural land used for plantation farming (rubber and oil palm), rice fields, food crops (palawija), and BPP's pilot garden.The research sites also represent various types of soil, both mineral soils and peat soils.In coastal areas, the type of soil studied generally belongs to the order Entisol and Ultisol, while in tropical forest buffer areas it belongs to the order Ultisol and Inceptisol.There are two locations that are categorized as organic soils with organic matter content above 30 percent, namely in the Mukomuko area (MM-4 code) and Central Bengkulu (BT-7 code).Thus, the results of this study can later be used to predict groundwater characteristics for most soil types in Bengkulu Province.
The research location covers 17 different land units so that it has a relatively wide distribution of basic soil characteristics for data extrapolation needs in the Bengkulu Province area.Of the 21 locations studied, only 4 land units appeared in two sub-district locations, namely Pq 8.2 (in Ulok Kupai and Talang Empat sub-districts), Hq 1.1.1(Giri Mulya and Penarik), Pa 5.2 (IV Koto and Pondok Kelapa), and Tf 3.2 (Lais and Pondok Kubang).While the other 13 locations each represent a separate unit of land so that this study is sufficient to have a wide range of soil types.This description of the location shows that the results of this study have a level of regional representation that can be accounted for.This is very important because one of the research outputs is as a reference in estimating the characteristics of groundwater in other locations not studied at this time.

Soil profile description
The five locations studied represent the typology of land in North Bengkulu Regency.The Air Napal location represents a coastal area with an altitude below 100 m above sea level (asl), Armajaya and Argamakmur locations represent an altitude of 100-300 m above sea level, while the Kerkap and Hulu Palik locations represent areas at an altitude of 300-500 m above sea level.A visual description of the soil profile at the 5 research sites is presented in Figure 8, while the detailed description results are presented in Table 2.The Kerkap location has the thickest A Horizon (0-50 cm depth) with a high organic matter content, while the thinnest is found in the Upstream.Palik (10 cm depth).The location of Argamakmur is a former rice field which was converted into rubber since 10 years ago so that it has an impermeable layer (Horizon E) at a depth of 25-32 cm.As a result, the land in this location is easily flooded when it rains.On the other hand, rubber plants often fall when the rain does not fall for a relatively long time.

Groundwater profile
The groundwater profile analysis begins with the presentation of the electrical impedance profile data which is measured directly and periodically at the five research locations, because the soil water content detector used in this study was designed through the conversion of the ground electrical impedance value measured by the device.Based on the measurement results presented in Figure 8, the electrical impedance at the five research locations has various fluctuations related to the occurrence of rain before the measurement.The value of electrical impedance at the Napal and Kerkap Air Locations was relatively less volatile for six measurements during the period 26 May to 12 June 2019 especially in layers less than 50 cm below the ground surface.For all locations, the highest electrical impedance value occurred on the measurement on May 30, 2019, which is after about a week of never raining at the observation location.On the other hand, the lowest electrical impedance value was obtained on the measurement on June 4, 2019 when there was a rather heavy rain about 12 hours before the measurement was made.Based on the description above, the value of electrical impedance increases when there is no rain and drops significantly following the occurrence of rain at the research location.
The electrical impedance response to weather changes, which is indicated by the distribution of values between measurement times, is seen to be highest in the Upper Palik and Argamakmur locations.The maximum range of impedance values at the two locations ranged from 5 k-Ω when measurements were made immediately after the rain to 40 k-Ω during conditions without rain.The high diversity of electrical impedance values between measurement times in Hulu Palik and Argamakmur looks consistent in most of the observed profile layers.State of the art of this research is the fact that soil moisture characteristics in agricultural land, including soil moisture content at field capacity conditions and permanent wilting point, is one indicator of sustainable land productivity.The characteristic of instantaneous soil water content is the product of the amount of water entering and leaving the soil profile.Water that enters the soil profile is generally influenced by climatic factors, especially the distribution and intensity of rainfall [1].Loss of water from the soil profile occurs through percolation due to gravity and evapotranspiration through the soil surface and plants.Both processes of groundwater loss are influenced by topography and soil properties so that the rate of water loss varies from one place to another [2][3].Thus, mapping the distribution of soil moisture characteristics requires a variety of land variables that are closely related to soil moisture content in the field.The process of water loss from the soil profile is influenced by evapotranspiration (evaporation from the soil surface and plant canopy) and drainage due to gravity.Loss of water through plant transpiration contributes 85% while evaporation from the soil surface is only 15% of the overall evapotranspiration [4].This phenomenon shows the dependence of physiological processes on plants on the availability of water in the soil.
There are two factors that determine the rate of water loss from the soil, namely external factors in the form of weather conditions, especially temperature, solar radiation, relative humidity and wind speed, and internal factors such as the basic characteristics of the soil itself.Calculation of evapotranspiration based on weather conditions requires accurate data and is measured directly because weather variables obtained based on estimates will result in a measurement error of 30% [5].Two of the internal soil factors that affect groundwater loss through evapotranspiration and drainage processes are soil texture and organic matter content.Both of these soil properties regulate the availability of groundwater through the ability of the soil to hold and pass water from the soil profile.Coarse-textured (sandy) soils can pass water quickly so they are less able to store water for a long time, while fine-textured (clay) soils have greater water-holding capacity but are less available to plants [6].Medium textured soil (clay texture, where the content of sand, silt and clay fractions in a balanced condition) has the maximum ability to provide water for plants.Therefore, the ability to store water in sandy soils can be achieved by removing the clay layer on the subsoil and mixing it with the sandy layer on the topsoil [7].
Organic matter content has long been known as an agent that can increase the capacity to store and provide groundwater for plants as reported by Hudson [8,9].Organic matter in the soil is able to store water weighing several times the weight of the organic matter itself so that its presence can overcome water shortages in coarse-textured soils.However, the accumulation of organic matter in the soil is usually limited to a depth of less than 20 cm, while the root zone of agricultural crops is in the 0-20 cm layer [10].Research from 1998 to 2000 [11] found that soil moisture content has a very close relationship with the electrical impedance representing the dielectric properties of the soil and the results were first reported in 2004 [12].The close relationship between the two variables was followed up by research in 2016 and 2017 on the design of instruments that can be brought into the field for direct soil moisture measurement [13].In the past year, the instrument, which we call the handheld dielectric, has been used for in-situ soil moisture estimation research in the field by measuring the dielectric properties of the soil [14].However, the scope of application of this technology is only on a laboratory scale, while the soil samples used only partially represent the locations of agriculture and plantations so that they do not represent the district as a whole.Therefore, this study is a follow-up to 2018 research on a more detailed scale at the district scale and carried out directly in the field.

Conclusions
The five soil types studied had different temporal fluctuations in groundwater profiles during the period from the end of the rainy season to the beginning of the dry season in 2019.The highest temporal fluctuations were found in the 0-40 cm layer, while the groundwater profile at a depth of 40-100 cm was relatively higher.stable.
Spatial fluctuations of soil water content according to depth are largely determined by the character of the soil in each layer.Layers containing high organic matter fluctuated the least during measurement, possibly due to their high water holding capacity.

Fig. 1 .
Fig. 1.Dielectrometer instrument from previous research modified to measure the availability of ground water for plants.

Fig. 4 .
Fig. 4. Sampling process of intact and disturbed soil at 10 depths at each profile location.

Fig. 5 .
Fig. 5. A. Implantation of sensor cables at intervals of 5 cm to a depth of 100 cm (left).B. Illustration of an embedded sensor for measuring soil moisture content (right).

Fig. 8 .
Fig. 8. Temporal changes in ground electrical impedance profile at 5 study sites.

Table 1 .
Description of soil sampling locations in North Bengkulu Regency as survey results at the end of July April 2019.

Table 2 .
Soil profile description.