Protected area development examination based on land use curve number value changes in Nusantara National Capital Region

This study examines the impact of land use changes on the intended extent of protected areas in the Nusantara National Capital Region (IKN). This study’s land use changes result from land use projections based on the IKN’s spatial plan (RTRW IKN) and land use projections considering the potential flood inundation areas for the Q100 recurrence interval, as determined by Clue analysis. Regression analysis with SPSS is used to determine the relationship between one or more independent variables (predictors) and a dependent variable (response variable) to analyze the mathematical equation used to calculate the size of protected areas. The analysis results reveal alterations in the size of protected areas within a watershed due to variations in the Curve Number (CN) values. A more excellent CN value reduces the extent of protected areas in a watershed and vice versa. This study is anticipated to provide helpful information for spatial planning decisions in the IKN, particularly in the Government Core Centre (KIPP) region.


Introduction
The study area is the dominant watershed in the Core Government Center Area (KIPP) of the new State Capital, which is in the administrative area of North Penajam Paser Regency, East Kalimantan, with the dominant watersheds in the area being the Sanggai Watershed, Trunen Watershed, and Semuntai Watershed.Figure 1 presents the watershed of the KIPP area.This study examines the impact of land use changes on the intended extent of protected areas in the Nusantara National Capital Region (IKN).This study's land use changes result from land use projections based on the IKN's spatial plan (RTRW IKN) and land use projections considering the potential flood inundation areas for the Q100 recurrence interval, as determined by CLUE-S model analysis.Regression analysis with SPSS is used to determine the relationship between one or more independent variables (predictors) and a dependent variable (response variable) to analyze the mathematical equation used to calculate the size of protected areas.The analysis results indicate changes in the extent of protected areas in a watershed caused by variations in the Curve Number (CN) values.A higher CN value reduces the size of protected areas within a IOP Publishing doi:10.1088/1755-1315/1311/1/012061 2 watershed and vice versa.This finding aligns with previous studies showing the importance of CN values in assessing hydrological impacts and land use changes [1][2][3][4][5].The implications of this research are crucial for informed decision-making in spatial planning within the IKN, particularly in the KIPP area.Understanding the dynamic relationship between land use changes and protected areas can aid policymakers in implementing sustainable development strategies [6][7][8][9][10] and enhancing the resilience of urban areas to environmental changes [11][12][13][14][15][16].
This study aims to obtain a model that can calculate the Protected Area (Ax) based on the CN value in KIPP IKN as a recommendation for changes to the RTRW to reduce the flood inundation area by recommending more protected areas to be preserved.To assert, Kurniawan et al. [17] have done similar research at the site, explaining how important it is to project land use and the likelihood of flooding in the future after the development in 2045.The study uses a new capital development scenario, which predicts 25 years after the development, resulting in a massive increase in built-up and open areas with 742% and 157%, respectively.Otherwise, forest, plantation, and fishpond will decline by 26%, 53%, and 28%, respectively, because of the massive development of the new capital.These massive replacements of natural land cover into built-up areas lead to inevitable adverse effects causing run-off increases and the likelihood of flooding, so continued research considering projected land use and hydrological response to provide accountable and proper planning.Projecting land use development about flood probability needs to be considered in the planning, especially for mega projects like new capital development, which will have many effects during and after the construction.Kurniawan et al. [17] study does not use flood factor as the projection input in the CLUE-S model, so this study aims to complement that weakness and expects better results as the spatial planning contribution to hinder adverse effects that could happen.

Analysis of Inundation Area with HEC-RAS
Flood inundation analysis was conducted using the HEC RAS 6.3.1 software.HEC RAS is a program that models water flow in rivers.HEC RAS 6.3.1 is equipped with 2D analysis capabilities, which aid in generating flood inundation areas.The run-off parameters employed in the flood inundation modeling simulation consist of the Q100 flood discharge, determined based on the R100 rainfall calculations, and land use parameters sourced from the RTRW scenario land use projections.The modeling results will provide information about potential inundation areas, which will be used as boundary parameters in the final analysis scenario for projecting land use.

Analysis of the CLUE-S Land Use Spatial Model
The CLUE-S spatial model integrates four essential variables for analyzing land use changes.These variables are land use demand, land use change likelihood (elasticity), location characteristics, and spatial policies.Additional parameters considered in the final analysis of land use projections include the flood inundation areas derived from the prior HEC-RAS simulation.Two modules comprise this model: a non-spatial module and a spatial allocation module.Four inputs are required to construct the model, which yields an optimal solution through iterative calculation of multiple conditions and possibilities.The four required inputs are (1) land-use demand, (2) geographical features, (3) a spatial planning policy that includes constraints, and (4) conversion plans for specific land-use types.In this model, the probability of land change is quantitatively calculated using the logistic regression equation.This equation predicts land use change by incorporating physical, social, economic, and policy factors.The four inputs were used to simulate the similarity of each grid cell within each land use category.It also calculates land allocation in each land use classification based on the competitive strength to fulfill the total allocation for each scenario.

Mathematical Equation Analysis for Protected Area
Regression analysis is a statistical technique used to understand the relationship between one or more independent variables (explanatory variables) and one dependent variable (response variable).After obtaining the DAS characteristic values as independent variables, a correlation analysis is conducted to identify parameters associated with the protected area's extent.From various DAS characteristic parameters, two parameters will likely be chosen for inclusion in the equation model.Multiple linear regression develops a mathematical equation for the protected area (Ax).The protected area's extent is the dependent variable (y), while DAS characteristics function as independent variables (x).

Verification and Model Validation
The generated equation model needs to be tested using Clue analysis results.Testing is conducted using different DAS regions.The verification and validation method employs Mean Absolute Error (MAE) and Root Mean Squared Error (RMSE) as metrics to validate the model against the observed baseline data.Deviations between the model and observation data are expected to be minimal.

Projection of Land Use in the RTRW IKN Scenario
The RTRW IKN scenario involves integrating the development intervention for capital city relocation into future projections.The growth of built-up areas is predicted to advance rapidly in tandem with the anticipated massive population increase.The expansion of built-up land is divided into two phases.From 2020 to 2035, the initial phase will experience intensive built-up land growth due to early-stage development and robust political interventions.Political will has been identified as a significant driving factor, similar to previous research on capital city relocation projects.This pattern was observed during Indonesia's capital city relocation process, necessitating the implementation of this initial phase.
The second phase sees continued built-up land development, mainly in supporting sectors but less intensively than the first phase.Political interventions are projected to weaken during this second phase.The growth rate of built-up land follows previous research on the evolution of land use in Jakarta as the capital city during two stages: 1950-1970 and 1970-2008.These 22% and 11% growth rates determine the built-up land growth rate in the IKN scenario in the Sanggai Watershed, as shown in Table 1.Looking at the details, Table 4 shows the inundation within each land use type in different scenarios.As inundation exposure in two different scenarios shown in Figure 2 and Figure 3, for the inundated area within the built-up area of the Sanggai watershed, the extent has reduced from 254.804 hectares to 116.166 hectares.In the Trunen watershed, it has decreased from 254.859 hectares to 80.868 hectares, and in the Semuntai watershed, it has reduced from 384.440 hectares to 101.339 hectares.The flood inundation within developed regions has been redistributed to forested areas, plantations, and open lands.

Correlation Between Protected Areas and Watershed Characteristics
Analysing statistical correlations between watershed parameters and protection area (Ax) is crucial for deriving the protection area equation model (Ax).Table 5 depicts the number of every input in the model for each watershed and sub-watershed, while Table 6    : Curve Number The requirement to do regression analysis is to do a classic assumption test.The model will be more appropriate to use and produce more accurate calculations if the classical assumptions can be met according to the standards [16].
New formula models need to be validated before their efficacy can be assessed.Finding the Root Mean Squared Error (RMSE) number is one validation test that can determine a model by analyzing the accuracy of a model's predicted outputs.Square the error (model 1-clue) and divide by the number of observations to get the root-mean-squared error (RMSE).Protected area data (Axe) for the model subwatershed and the findings of the hints analysis for 23 sub-watersheds are summarised below in Table 7.
Table 7 The validation results for the 23 sub-watersheds were 0.13, the same as those for the 20 subwatersheds in the study, namely 0.13.A low RMSE value indicates that the variation predicted by a model closely matches the observed variation.The smaller the RMSE value, the more closely the predicted value matches the analysis value of CLUE.

Comparison of Watershed Protection Formula Analysis Between RTRW Scenario and Modified/Final Scenario
Comparing the analysis results between the RTRW-based land use scenario and the modified/final scenario provides insights into the impact of land use modification on watershed protection.The model based on the modified scenario shows an inverse relationship between CN and the Watershed Protection Area for a specific watershed area.As CN increases, the Watershed Protection Area decreases, and vice versa.Looking at the details, this statement is supported by the numbers shown below in Table 8.

Conclusion
Projection of land use changes in KIPP IKN using the CLUE-S model indicates that in 2045, the builtup area will increase by 18.08 % annually, which also means an increase in the flood.Using the projection of the flood-inundated regions in 2045 as input for protection areas in the final scenario of IKN Spatial Planning will decrease flood discharge and inundation areas (For the Semuntai Watersheed, the Q100 flood discharge decreased from 321.9 m 3 /s to 305.3 m 3 /s, a decrease of 5.16% in Trunen Watershed, from 349.6 m 3 /s to 342.0 m 3 /s, a decrease of 2.18% and in the Watershed Sanggai, the Q100 flood discharge decreased by 1.88% from 502.9 m 3 /s to 493.4 m 3 /s.The area of flooded urban areas decreased from 254,804 ha to 116,166 ha in the Sanggai Watershed, from 254,859 to 80,868 ha in the Trunen watershed, and from 384,440 ha to 101,335 ha in the Semuntai Watershed).The protection area based on the IKN RTRW scenario is 55% and will be increased to 61% compared to the final scenario considering the Q100 flood.Based on the final scenario, the protected area equation reveals an inverse proportionality between the CN value in a given catchment area and the Protected Area.The greater the CN Value, the smaller the protected area, and vice versa.As this research is done and proves there is a high probability of the flood discharge increase along with the massive replacement in land use from natural land cover into built-up areas.Consequently, a continuum recommendation on preserving and enhancing current protected areas using an offsetting approach can be an ideal solution to prevent floods in the future.Besides that, further research into the actual hydrological response after every development phase is also needed to complement and strengthen our research to establish the most sustainable and proper planning characterized by nature-based and risk-based.

Figure 1 .
Figure 1.Sub-watersheds located in the KIPP area are marked with red shading.
shows the result from statistical analysis regarding driving factors of land use development.The following are the inputs into this model: S : Slope (Slope) A : Watershed Area (km 2 ) R100 : Rain design return period 100year (mm) CN : Curve Number Ia : Initial Abstraction (mm) Imp : Impervious (%) Q : Design flood discharge Q100 (m 3 /sec) Bo : Constant B1 : Coefficient of an independent variable X0 : Distance from the coast X1 : Population density X2 : Rainfall X3 : Elevation X4 : Slope X5 : Distance from the toll road X6 : Distance from the city centre X7 : Distance from the river

Table 1 .
Projection of land use in the Sanggai Watershed 2020-2045 3.2.Flood Inundation Comparison in Q100 ScenariosThe largest flooded area in the KIPP area is found in Urban Forest, covering 42.99% of the inundated area, followed by City Park with 29.30%.Built-up areas, including public infrastructure, trade and service areas, and residential zones, account for 7.87% of the inundated area.The total flooded area due to Q100 discharge in the RTRW IKN land use scenario for the 2045 projection is 646.14 hectares or 24.39% of the total KIPP area.Details of inundated areas in the KIPP area are presented in Table2.

Table 2 .
Projection of inundated areas in the KIPP Area Due to Q100 Floods in 2045 Projection of Final Scenario Land Use It is necessary to reproject land use while incorporating potential flood inundation areas as projection parameters (protected areas) to establish a secure spatial plan to mitigate flood risks.In this study, the final land use scenario is one whose parameters have been modified based on flood inundation.By projecting the ultimate land use scenario, it is possible to develop a land use plan in which developed areas, such as offices, commercial zones, and residential areas, are situated outside flood-prone zones.In addition, modifying protected areas following the final scenario is anticipated to change the projected land cover pattern by 2045.The projection results serve as a guide for spatial planning, especially in developed areas.3.4.Comparison of Q100 Flood Inundation between IKN RTRW Land Use Scenario and Final ScenarioUpon analyzing flood discharge based on the final scenario, it is evident that each major watershed within the KIPP region experiences a reduction in flood discharge compared to the IKN RTRW land use scenario.As shown in Table3, for the Semuntai watershed, the Q100 flood discharge decreased from 321.9 m 3 /s to 305.3 m 3 /s, marking a 5.16% reduction.In the Trunen watershed, the Q100 flood discharge reduced from 349.6 m 3 /s to 342.0 m 3 /s, indicating a 2.18% reduction.Lastly, in the Sanggai watershed, the Q100 flood discharge decreased by 1.88%, from 502.9 m 3 /s to 493.4 m 3 /s.Regarding the extent of flood inundation, reductions are observed across the watersheds.In the Sanggai watershed, the inundation area decreased by 5.43%, from 667.059 hectares to 630.830 hectares.In the Trunen watershed, the inundation area decreased by 5.93%, from 943.714 hectares to 887.728 hectares.Similarly, in the Semuntai watershed, the inundation area decreased by 2.58%, from 1,168.569 hectares to 1,138.374hectares.

Table 3 .
Comparison of Peak Discharge and Flood Inundation Area

Table 4 .
Comparison of Flood Inundation Area between IKN RTRW Scenario and Final Scenario Based on Land Use Types in 2045

Table 5 .
Watershed and Protected Area Characteristics Data (Ax)

Table 6 .
Regression number resulted from statistical analysis of driving factors . Validation calculations for 23 sub-Watersheds RMSE methods

Table 8 .
Comparison of protected area based on scenario three and Final scenario