**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

*Proposed Refinements to Design Procedures for Geosynthetic Reinforced Soil (GRS) Structures in AASHTO LRFD Bridge Design Specifications*. Washington, DC: The National Academies Press. doi: 10.17226/25416.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

*Proposed Refinements to Design Procedures for Geosynthetic Reinforced Soil (GRS) Structures in AASHTO LRFD Bridge Design Specifications*. Washington, DC: The National Academies Press. doi: 10.17226/25416.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

*Proposed Refinements to Design Procedures for Geosynthetic Reinforced Soil (GRS) Structures in AASHTO LRFD Bridge Design Specifications*. Washington, DC: The National Academies Press. doi: 10.17226/25416.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

**Suggested Citation:**"Chapter 3 Proposed Revisions to Incorporate Sv into Current AASHTO Design Procedures." National Academies of Sciences, Engineering, and Medicine. 2019.

Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

14 Chapter 3 PROPOSED REVISIONS TO INCORPORATE SV INTO CURRENT AASHTO DESIGN PROCEDURES 3.1 OVERVIEW The qualitative and quantitative outcomes of this research, stemming from laboratory testing, field instrumentation data, numerical analysis, and comprehensive interpretation of previous work by others, was used to develop the design recommendations proposed herein. This section summarizes these recommendations and provides guidance for modification of appropriate sections of AASHTO to recognize the benefits of using closely-spaced reinforcements in both GMSE and load-carrying GMSE bridge abutments (including those currently identified as GRS-IBS structures). The outcomes of this research study, as detailed in the previous sections of this report, led to a number of design recommendations presented herein to evaluate the effect of closely-spaced reinforcements on the design of geosynthetic-reinforced soil structures. The proposed revisions to incorporate the effect of Sv into current AASHTO design procedures are grouped into five design aspects, as follows: 1) Effect of vertical spacing Sv on Tmax magnitude and distribution 2) Effect of vertical spacing Sv on To magnitude and distribution 3) Effect of vertical spacing Sv on stress distribution and the design of the bearing seat, including the design of reinforcement under bearing seat 4) Effect of vertical spacing Sv on the structureâs vertical and lateral deformation 5) Effect of vertical spacing Sv on the bump at the end of the bridge In the execution of the NCHRP Project 24-41study, a distinct influence of closely-spaced reinforcement was identified on the performance of geosynthetic-reinforced soil structures. Accordingly, the boundary vertical spacing below which such improvement is significant was identified. The major benefits of adopting a comparatively closely-spaced vertical reinforcement spacing Sv are described in the results of the research conducted as part of this project (see Sections 5 through 7 of theNCHRP Project 24-41 Final Report) as well as in the results of research by others (see Section 4 of the NCHRP Project 24-41 Final Report). As indicated in Chapter 1, the specific basis and research outcomes relevant to the five identified design aspects will be documented in Chapter 4. A combination of experimental results and numerical parametric evaluations of reinforced fill materials having different shear strength properties showed little to no effect on the extent of the soilâgeosynthetic interaction influence zone. On the other hand, soil-reinforcement interface shear strength properties were found to have a significant impact on the extent of the influence zone. Specifically, increasing soil-reinforcement interface strength was found to result in larger load transfer from the reinforcement to the surrounding soil. Considering the minor influence of soil shear strength properties, a zone of influence ranging from 0.1 to 0.2 m (4 to 8 in.) on each side of the geosynthetic is recommended for cases where the coefficient of interaction exceeds

15 0.8 (i.e., the reinforcement pullout factor F* > 0.8 tan Ï) for free draining reinforced fill materials that currently meet AASHTO 2017 specifications. Additional testing would be required to define the extent of the zone of influence for soilâgeosynthetic interfaces characterized by comparatively lower interface shear strength values. As will be discussed throughout the rest of this document, the primary structural advantages of closely-spaced reinforcement include the following: â¢ A uniform distribution of reinforcement unit tension with depth, which has cost- effectiveness implications for the case of structures designed using the same vertical reinforcement spacing and reinforcement design strength with depth â¢ A uniform distribution of reinforcement connection load with depth, which also has cost- effectiveness implications for the case of structures designed using the same vertical reinforcement spacing and reinforcement design strength â¢ Improvement of structural stiffness as it relates to comparatively decreased lateral displacements of the overall structure â¢ Improved performance of the bearing seat for the case of load-carrying GMSE bridge abutments (including GRS-IBS structures) â¢ Decreased bump at the end of the bridge for the case of load-carrying GMSE bridge abutments (including GRS-IBS structures) â¢ Reduction in the potential for pullout failure, both at the back of the reinforced soil mass and at the face when using frictional connections â¢ Comparatively more consistent and uniform placement and compaction of backfill soil due to enhanced distribution of compaction stresses and expected improved field quality control during construction The increased number of reinforcement layers was observed to decrease the lateral movement and rotation of the facing elements. As a result, the wall facing unit actually carries more load as observed through increased lateral stress measurements on facing elements that have modest interface strength between units, thereby reducing the stress in the reinforcements, especially at the facing connection T0. Similar reductions in T0 were also observed when using secondary reinforcements at the facing with wider spaced primary reinforcements. These improved responses appear to be especially significant for support of external surcharges such as bridge structures. With regard to the discrepancy between design equations currently available to predict tension in the reinforcement Tmax, a major conundrum is the reliability of the current methods of design. For stand-alone roadway walls, the current AASHTO (2017)/FHWA (2009) simplified method has been reported to over predict the required maximum Tmax for geosynthetics based on instrumented structures, with an overestimation often reported to be a factor of two (Allen et al. 2001, Allen et al. 2002a, Allen and Bathurst 2002b, Miyata and Bathurst 2007a, 2007b, and Bathurst et al. 2008). Such overestimation has often been explained by acknowledging that instrumentation results provide working stress conditions rather than limit state conditions. The GRS-IBS design guide (FHWA 2012, 2018) produces both a limit state design value for the failure strength Tf and a required working stress value Treq, somewhat related to service conditions.

16 However, the ultimate strength value was defined based on 3D column tests, which may not have boundary conditions that are representative of walls supporting bridges (e.g., confinement of the loading plate and 3D vs 2D loading). These results have been reported to be even more conservative for roadway walls, requiring almost twice the reinforcement strength obtained using the AASHTO/FHWA design method (Phillips et al. 2015). For bridge-supported structures, the two methods have been reported to result in reasonably similar tensile capacity requirements, but are both still overly conservative based on results from instrumented structures (Phillips et al. 2015). The two new design approaches for geosynthetic reinforcements, limit equilibrium method at a limit state (Leshchinsky et al. 2016) and simplified stiffness method at working load conditions (Allen and Bathurst 2018), produce less conservative, although seemingly more accurate results based on results from instrumented structures. As a result of the findings from the different research components conducted in NCHRP Project 24-41, the most appropriate approach is to incorporate the design of structures with closely- spaced reinforcements directly into the current AASHTO Section 11 (AASHTO 2017) design specifications based on the Simplified method, but adopting special considerations for the distribution of Tmax with depth and the deformation response of GMSE systems with closely- spaced reinforcement. Accordingly, changes are proposed to Article 11.10 of the existing AASHTO code that incorporates the benefits of closely-spaced reinforcements. In addition, the method could also be incorporated directly into the newly proposed simplified stiffness method, which accounts for the stiffness of the reinforced soil mass and facing stiffness. Finally, the newly proposed limit equilibrium method (Leshchinsky et al. 2016) could also be appropriately modified based on the same approach. However, the modifiers for the design approaches other than the Simplified method would need to be method-specific due to differences in the reliability of the predictions. Additionally, modifications to the current Article 11.10.10.1, Concentrated Load Conditions, are proposed for the case of load-carrying GMSE bridge abutments in general and for the case of GRS-IBS structures in particular, due to the uniqueness of the approach and documented performance. The selected design approach can be directly incorporated into the current AASHTO LRFD design approach using the current nominal and factored reinforcement strength requirements for internal stability as the method does not modify the global determination of strength demand. The synthesis of data generated by others, as documented in Sections 2 to 4 of the NCHRP Project 24-41 Final Report provide additional support to the proposed revisions in the design methodology. The use of existing resistance factors is recommended as they are inherently conservative for closely-spaced reinforcement. Although sufficient data may not be available for full calibration of the approach, the use of closely-spaced reinforcements leads to several inherent benefits that improve the overall conservatism of the structure and should balance out any uncertainties resulting from the proposed modifications. These include increased redundancy in the system and better control of lift thickness and correspondingly compaction control. Different resistance factors could and should be developed in the future to account for these improvements in uncertainty.

17 In summary, adopting closely-spaced reinforcement in design was found to have significant impact on design on a number of aspects, including the distribution of maximum reinforcement tension with depth (a comparatively uniform profile was identified), as well as the magnitude of the horizontal structure deformations (a comparatively smaller magnitude), with these impacts being particularly evident after a vertical surcharge has been applied. Adoption of a uniform distribution of reinforcement tension with depth may have significant beneficial implications for the cost of structures constructed using a single reinforcement type and vertical reinforcement spacing with depth. However, the recommendation is to maintain the same overall reinforcement tensile capacity of the wall, independent of the reinforcement vertical spacing, when using reinforcements of the same modulus/strength. This is because a further reduction in the tension carried by the reinforcements that can be directly attributed to closely-spaced conditions was found to be difficult to quantify. Specific revisions to account for the impact of Sv on current AASHTO design procedures are provided next, with revisions grouped according to the previously identified five design aspects. 3.2 EFFECT OF SV ON TMAX MAGNITUDE AND DISTRIBUTION A number of revisions to the current AASHTO design procedures are recommended to account for the impact of reinforcement vertical spacing, Sv, and particularly the effect of closely-spaced reinforcements on the magnitude and distribution of Tmax. The specific revisions recommended for incorporation into AASHTO are: â¢ For the purposes of defining Tmax, a vertical spacing Sv,nc = 0.4 m (16 in.) shall be adopted as the vertical spacing above which prediction of Tmax does not account for any benefits of âclosely-spacedâ reinforcement when considering free draining granular materials. â¢ For the purposes of defining Tmax, a vertical spacing Sv,c = 0.2 m (8 in.) shall be adopted as the vertical spacing below which prediction of Tmax accounts for the benefits of âclosely- spacedâ reinforcement when considering free draining granular materials. â¢ When adopting intermediate vertical spacing, ranging from Sv,c = 0.2 m (8 in.) to Sv,nc = 0.4 m (16 in.), determination of Tmax partially benefits from âclosely-spacedâ reinforcement. â¢ The total tensile capacity Ts (i.e. the summation of maximum tension in the entire structure) shall be the same, independent of the reinforcement vertical spacing, and equal to that currently adopted by AASHTO for the Simplified method. That is, for surcharge-free geosynthetic-reinforced structures, the total tensile capacity remains as (see Figure 3.1): Ts = â , = â â [Equation 3.1] The total horizontal load resulting from the increased horizontal stresses due to surcharge loads, ÎÏH , should be added to the total tensile capacity Ts defined using Equation 3.1.

18 Figure 3.1. Total tensile capacity (shaded area) for all MSE walls, independent of the reinforcement vertical spacing, Sv. â¢ Even though the effect of closely-spaced vertical spacing does not affect the calculated total tensile capacity, it is considered to affect the distribution with depth of Tmax. Specifically, while the distribution remains linear with depth for widely-spaced reinforcement, it should be considered uniform with depth for closely-spaced reinforcement, and should involve an intermediate distribution with depth for the case of intermediate vertical reinforcement spacing, as follows (see Figure 3.2): , = â â â + â â for Sv â¥ Sv,nc = 16â [Equation 3.2] , = â â â + â â for â¤ , = 8" [Equation 3.3] , = â â â + , , â + âÏ â S for 8" = , â¤ â¤ , = 16â [Equation 3.4] â¢ The resistance factors could be different (higher) for structures involving closely-spaced reinforcements than for structures involving comparatively widely-spaced reinforcements due to the inherent improved reliability of structures with closely-spaced reinforcement. â¢ Until sufficient data for calibration is generated, the same resistance factors shall be adopted for closely-spaced and widely-spaced reinforcements (even though use of different notations for the two resistance factors are recommended). â¢ The reinforcement strength and layout selected under working load conditions must also be evaluated under limit state conditions. Global stability using limit equilibrium analysis, as currently recommended by AASHTO, should confirm the adequacy of the adopted resistance factor on soil strength at a limit state of the wall. The reciprocal value of the resistance factor on soil strength is the conventional factor of safety, recommended by

19 AASHTO as 1.3 or 1.5, depending on the criticality of the structure. The importance of such assessment is demonstrated by Leshchinsky (2009) and Leshchinsky et al. (2017). â¢ While the distribution of Tmax with depth is supported by field data, it is assumed that its locus coincides with a planar slip surface inclined at (45+Ï/2). Such an assumption is consistent with AASHTO for vertical walls. Subsequently, knowledge of Tmax and its location for each layer enables assessment of the adequacy of the available pullout resistance. Overall, and while no differences are recommended at this point regarding the calculated total tensile capacity in structures involving closely- and widely-spaced reinforcements, the distribution of Tmax with depth for the case of closely-spaced reinforced structures under working stress conditions should be considered uniform. The practical implication is that the Tmax for structures involving closely-spaced reinforcement will be approximately half the maximum value of Tmax for structures with the same total tensile capacity but widely-spaced reinforcements (this is for the case in which the reinforcement is uniformly spaced and its type does not vary along the height of the wall). (a) (b) (c) Figure 3.2. Tmax,i value for reinforcement layer at a depth zi: (a) for cases involving Sv â¥ 16â; (b) for cases involving Sv â¤ 8â; (c) for cases involving 8â â¤ Sv â¤ 16â.

20 the NCHRP Project 24-41 study concluded that close vertical spacing in extensible geosynthetic- reinforced walls, having coverage of 100%, affects the distribution of Tmax. over the depth of a wall. Specifically, while the distribution remains linear with depth for widely-spaced reinforcements, it is considered uniform with depth for closely-spaced reinforcements. Specific recommendations to incorporate this approach into AASHTO would be in Article 11.10.6.2.1 (Maximum Reinforcement Loads), as follows: Edit paragraph 4 of 11.10.6.2.1 as follows: For the Simplified Method, factored horizontal stress, ÏH, at each reinforcement level shall be determined as: ( )H P v r HkÏ = Î³ Ï + ÎÏ (11.10.6.2.1-1) where: Î³P = the load factor for vertical earth pressure EV from Table 3.4.1-2 kr = horizontal pressure coefficient (dim.) Ïv = pressure due to gravity forces from soil self- weight within and immediately above the reinforced wall backfill, and any surcharge loads present (ksf) ÎÏH = horizontal stress at reinforcement level resulting from any applicable concentrated horizontal surcharge load as specified in Article 11.10.10.1 (ksf) For design with geosynthetic reinforcement, the beneficial effect of closely-spaced reinforcements can be accounted for in the Simplified Method by using the following equation, instead of 11.10.6.2.1-1, when the reinforcement vertical spacing is less than or equal to 8 inches: = (0.5 + â ) (11.10.6.2.1-2) and the following equation, instead of 11.10.6.2.1- 1, when the reinforcement vertical spacing is between 8 and 16 inches: = + + â (11.10.6.2.1-3) where: Sv is geosynthetic reinforcement vertical spacing (inches) Commentary: Add following to C11.10.6.2.1 NCHRP Project 24-41 found that closely-spaced vertical spaced reinforcements in walls reinforced using extensible geosynthetic reinforcements affects the distribution of Tmax over the depth of a wall. Specifically, while the distribution remains linear with depth for structures with widely-spaced reinforcement, it is considered uniform with depth for closely-spaced reinforcement, as shown in the following figure. Equations 11.10.6.2.1-2 and 11.10.6.2.1-3 account for this benefit. The equations are valid for continuous geogrid and geotextile reinforcements. Research has not been performed to verify the use of these equations for the case of geosynthetic strip or inextensible steel-reinforced walls.

21 3.3 EFFECT OF SV ON T0 MAGNITUDE AND DISTRIBUTION A number of revisions are recommended herein to account for the impact of reinforcement vertical spacing, Sv, and the effect of closely-spaced reinforcements on the magnitude and distribution of T0 in particular. The specific revisions recommended for incorporation into the current AASHTO design procedures are: â¢ Consistent with the limits defined for determining Tmax, For the purposes of defining T0, a vertical spacing Sv,nc = 0.4 m (16 in.) can be adopted as the vertical spacing above which prediction of T0 would not account for any benefits of âclosely-spacedâ vertical spacing when considering free draining granular materials. â¢ For the purposes of defining T0, a vertical spacing Sv,c = 0.2 m (8 in.) can be adopted as the vertical spacing below which prediction of T0 accounts for the benefits of âclosely-spacedâ vertical spacing when considering free draining granular materials. â¢ The adoption of an intermediate vertical spacing, between Sv,c = 0.2 m (8 in.) and Sv,nc = 0.4 m (16 in.) would partly account for the benefits of âclosely-spacedâ vertical spacing in T0 when considering free draining granular materials. â¢ The magnitude of the connection load, T0 is recommended as follows: , = . , [Equation 3.5] where Î» is the connection load coefficient (dimensionless) While it is anticipated that the value of Î» may be a function of the vertical spacing Sv, sufficient data is not available to establish such differentiation. Consequently, no revision to the AASHTO Design Guidelines is recommended at this point on the relation between T0 and Tmax, which effectively implies a connection load coefficient Î» = 1 for any value of Sv. â¢ The resistance factors should be different (higher) for structures involving closely-spaced reinforcements than for structures involving comparatively widely-spaced reinforcements. â¢ Until sufficient data is generated for calibration, the resistance factors for closely-spaced reinforcement shall be adopted as being equal to that for widely-spaced reinforcements (even though different notations for the two resistance factors are recommended). Since the distribution of Tmax with depth for the case of closely-spaced reinforced structures should be considered uniform, even the conservative assumption of T0 = Tmax will lead to comparatively lower requirements of T0 for the case of closely-spaced reinforced structures. The bulleted list of recommendations above should be incorporated into the commentary section of AASHTO to identify the effect of Sv on T0 magnitude and distribution in Article C11.10.6.2.2, as follows:

22 C11.10.6.2.2 For closely-spaced geosynthetic reinforcement, T0 will be equal to the maximum nominal reinforcement tension as calculated and the vertical spacing defined as in 11.10.6.2.2. Based on research in NCHRP Project 24-41 the magnitude of the connection load, T0 is recommended as follows: , = . , C11.10.2.2-1 where Î» is the connection load coefficient (dimensionless) While it is anticipated that the value of Î» is a function of the vertical spacing Sv and is anticipated to be < 1. However, sufficient data is not currently available to establish such differentiation. Consequently, Î» = 1 for any value of Sv. 3.4 EFFECT OF SV ON STRESS DISTRIBUTION AND DESIGN OF BEARING SEATS Modest revisions are also recommended to account for the effect of geosynthetic reinforcement vertical spacing, Sv, and the effect of closely-spaced reinforcement on the bearing seat in particular. These recommendations impact the design of the reinforcement under the bearing seat. Research findings documented in Chapter 7 of this report strongly supports the use of the truncated 2:1 distribution currently in AASHTO Article 3.11.6.3 and in FHWA (2009) for stress distribution when using closely-spaced reinforcement. The primary recommendation is to add commentary to Article 3.11.6.3 as well as Articles 11.10.10.1âConcentrated Dead Loads and 11.10.11-âMSE Abutments on the benefits of a bearing reinforced zone including: â¢ A bearing reinforced zone, involving geosynthetic reinforcements with a vertical spacing of 10 cm (4 in). is recommended for placement under the bearing seat of the bridge structure. â¢ Increasing the density of geosynthetic reinforcement immediately below the bearing seat or bridge wing wall is recommended, as it has a significant impact in minimizing the bump at the end of the bridge and reducing the lateral deformation of the wall. â¢ While the benefits of closely-spaced reinforcement may be applicable to steel reinforcements, research has not been conducted to verify the effect on the performance of steel-reinforced reinforced soil walls. The increase in vertical stress, ÎÏv, due to bearing seat loading affects the lateral deformations of load-carrying GMSE walls. It is also directly incorporated in the prediction of Tmax and T0 at each depth. The value of ÎÏv is added to the soil overburden pressure which, upon multiplication

23 by appropriate lateral earth pressure coefficient, leads to the horizontal stress used to predict the reinforcement load. The rate at which the stress dissipates with depth in the proposed approach is defined by a 2D pyramid with side slopes of 2(v) to 1(h). This approach is consistent with AASHTO (2017) and FHWA (2001, 2009). Extensive experience gained through the construction and operation of numerous GRS-IBS structures has indicated that very close reinforcement, spaced 0.1 m (4 in.) apart, will enhance performance when placed under the bearing seat. This bearing reinforcement zone may be composed of primary, full length layers spaced every 0.2 m (8 in.) whereas intermediate length layers are placed mid-span between the primary layers. Consequently, a reinforced mass under the seat bed is formed at a spacing of 0.1 m (4 in.). For the purpose of enhancing the load-carrying performance, the intermediate layers should be placed to a depth equal to at least the width of seat bed and extend at least 1.5 m (5 ft.) beyond the edge of the seat. Again, the specific commentary on the recommendations to incorporate into AASHTO regarding design of the bearing seat would be in Articles 3.11.6.3, 11.10.10, and 11.10.11 as follows: Commentary: Add the following to C3.11.6.3, C11.10.10, and C11.10.11 For geosynthetic-reinforced MSE walls supporting strip loads (e.g., bridge structures), a bearing reinforced zone, involving geosynthetic reinforcements with a vertical spacing of 4 in. is recommended for placement under the bearing seat of the bridge structure based on research performed under NCHRP Project 24-41 and field data developed under other FHWA research studies (FHWA-HRT-17-080 Adams and Nicks, 2009). Increasing the density of geosynthetic reinforcement immediately below the bearing seat or bridge wing wall is recommended, as it has a significant impact in minimizing the bump at the end of the bridge and reducing the lateral deformation of the wall. Design guidelines for an alternate approach using geosynthetic-reinforced soil walls to support bridge structures and other strip loads is provided in (FHWA-HRT-17-080 Adams and Nicks, 2009).

24 3.5 EFFECT OF SV ON VERTICAL AND LATERAL DEFORMATIONS The results of the parametric numerical evaluation (Section 7 of the NCHRP Project 21-41 Final Report) indicated a significant influence of reinforcement vertical spacing on the lateral deformation response of GMSE structures in general, and load-carrying GMSE bridge abutments in particular. The revisions recommended in this section are those to account for the impact of reinforcement vertical spacing, Sv, and the effect of closely-spaced reinforcement on the lateral deformations of GMSE structures in particular. While improvement could also be identified in the structure vertical response, the magnitude was minor. The specific revisions recommended for incorporation into the current AASHTO design procedures when using closely-spaced reinforcement and high quality aggregates (which should always be used for load-carrying GMSE bridge abutments and other GMSE structures supporting footings) are: â¢ Vertical displacements in load-carrying GMSE bridge abutments should be predicted using empirical data from similar structures (i.e., based on load test or similar instrumented structures) or analytical methods such as that proposed by Schmertmann as documented in FHWA (2006). â¢ In cases where the vertical settlement is known, or can be accurately estimated, the maximum lateral displacement and maximum lateral strain shall be estimated in agreement with the method documented in FHWA (2018), as follows: = 2 , [Equation 3.6] = , = = 2 [Equation 3.7] where DL = the maximum lateral displacement, bq,vol = the width of the load along the top of the wall including the setback, DV = the vertical settlement of the load, H is the height of the structure, ÎµL = the maximum lateral strain, and ÎµV = the vertical strain. The calculated lateral strain is typically limited to 2% (FHWA, 2018). â¢ In cases where the vertical settlement is unknown, a first order estimation of the lateral movement for free standing walls as well as walls supporting surcharge loads with closely- spaced reinforcement can predicted using a model established considering AASHTO (2017) Figure C11.10.4.2-1 / FHWA (2009), as follows: = 11.81 â 42.25 + 57.16 â 35.45 + 9.471 [Equation 3.8] = . 1 + 1.25 ( ) [Equation 3.9] where Î´R = an empirically derived relative displacement coefficient (dim.),

25 Î´max = the maximum displacement (units of H), H = the height of the wall (ft or m), J = the reinforcement tensile stiffness defined by the secant modulus at 2% strain, Sv = the reinforcement vertical spacing (units of H), q = the surcharge magnitude, and po = atmospheric pressure introduced in the equation for normalization purposes. Specific recommendations to incorporate into AASHTO the approach proposed herein regarding prediction of vertical and lateral deformations would be in Articles 11.10.4.1 and 11.10.4.2, and a specific commentary should be added to Article 11.10.11 outlining these approaches and referring to Article 11.10.4, as follows: 11.10.4.1 (add the following) For vertical compression within the reinforced zone in load-carrying GMSE bridge abutments shall be predicted using empirical data from similar structures (i.e., based on load test or similar instrumented structures) or analytical methods such as that proposed by Schmertmann as documented in FHWA (2006). 11.10.4.2 C11.10.4.2- Add under Figure C11.10.4.2-1 The curve in the figure can be modeled using the following equation: = 11.81 â 42.25 +57.16 â 35.45 + 9.471 [Equation C11.10.4.2-1] INSERT THE FOLLOWING AT THE END OF THE CURRENT COMMENTARY: For closely-spaced reinforcement (< 16 in.) and high quality aggregates (which should always be used for load-carrying GMSE bridge abutments and other GMSE structures supporting footings), lateral wall displacement can be predicted base on one of the following two methods. Where vertical settlement is known, or can be accurately estimated using the methods in Article 11.10.4.1, [INSERT 2ND BULLET ABOVE]

26 In cases where the vertical settlement is unknown, [INSERT 3RD BULLET ABOVE] C11.10.11 Lateral displacements shall be evaluated following the recommendations in Article 11.10.4.2. The calculated lateral strain ÎµL shall be limited to 2% (FHWA-HRT-17-080 Adams and Nicks, 2018). 3.6 EFFECT OF SV ON BUMP AT THE END OF THE BRIDGE Revisions are recommended herein to account for the impact of reinforcement vertical spacing, Sv, and particularly the effect of closely-spaced reinforcements on the abutment transition (i.e. âbump at the end of the bridgeâ) for the case of GMSE structures associated with bridge abutments (either load-carrying GMSE structures or structures with bridge loads transferred through piles to the foundation soils). The specific revisions recommended for incorporation into the current AASHTO design procedures are: â¢ A prediction of the total foundation settlements, St,f, for the abutment and embankment behind the abutment should be produced as part of the design of a GMSE bridge abutment involving the use of deep foundations to transfer the bridge loads in accordance with AASHTO Articles 10.5.2.4 and 10.6.2.4. â¢ A prediction of vertical reinforced fill elastic compression, Se,r, should be produced as part of the design of a GMSE abutment (whether a load-carrying GMSE abutment or a GMSE abutment involving deep foundations) in accordance with AASHTO Article 10.6.2.4. â¢ The bump at the end-of-the-bridge, BEB, for the case of a GMSE bridge abutment involving deep foundations is based on the settlement of the fill and foundation soils behind the deep foundations and shall be predicted as BEB = St,f + Se,r. â¢ The bump at the end-of-the-bridge, BEB, for the case of a load-carrying GMSE bridge abutment (including GRS-IBS walls) shall be predicted as BEB = Se,r â¢ Decreasing the reinforcement vertical spacing immediately below the bearing seat or bridge wing wall is recommended, as it has a significant impact in minimizing the bump at the end of the bridge. Specific recommendations to incorporate into AASHTO the approach proposed herein regarding prediction of the bump at the end of the bridge would include a paragraph added to Article 11.10.4.1, as follows, with commentary added to 11.11.1:

27 11.10.4.1 Add the following after last paragraph in current article: Differential settlement shall also be part of the analysis of abutment transitions in accordance with Article 10.5.2.4. For MSE walls with the bridge abutment supported by deep foundations, the analysis shall include the total settlement of the embankment behind the bridge including the compression of the fill following Article 10.6.2.4, as compared to the settlement of the footing supported by the deep foundation based on Articles 10.7.2.3 -10.7.2.5 for piles or Articles 10.8.2.2 â 10.8.2.4 for drilled shafts. For load- carrying MSE bridge abutments (i.e. cases in which the bridge is supported directly by the reinforced fill), the analysis shall include the compression of the reinforced fill under the bridge load following Article 10.6.2.4 and assume there is negligible compression of the fill behind the end of the bridge. C11.11.1. Add the following: Differential settlement shall also include an analysis of the abutment transition in accordance with Article 10.5.2.4 and Article 11.10.4. Design guidelines for an alternate approach for mitigating differential settlement (e.g., the bump at the end of the bridge) by using closely-spaced geosynthetic soil reinforcement and an integrated bridge structure is provided in (FHWA-HRT-17-080 Adams and Nicks, 2009).