Flood Standards Development
Input to Draft Meteorological/Hydrological Flood Draft Standards Dated 11-17-14
December 5, 2014
/ 2CONFIDENTIAL
AIR Input to Draft Flood Standards
/ 2
CONFIDENTIAL
AIR Input to Draft Flood Standards
Table of Contents
1 Introduction 7
2 Standard MHF-1 Flood Event Data Sources 8
2.1 MHF-1 Standards 8
2.1.1 Standard MHF-1.B 8
2.1.2 Standard MHF-1.C, first suggestion 8
2.1.3 Standard MHF-1.D 8
2.2 MHF-1 Purpose Statement 9
2.2.1 MHF-1 Purpose, First Paragraph 9
2.3 MHF-1 Disclosures 10
2.3.1 MHF-1 Disclosure 1, 2 and 4 10
2.3.2 MHF-1 Disclosure 5 10
2.4 MHF-1 Audit Items 11
2.4.1 MHF-1 Audit Item 1 11
2.4.2 MHF-1 Audit Item 5 12
2.4.3 MHF-1 Audit Item 6 12
2.4.4 (Moved/New) MHF-1 Audit Items 12
3 Standard MHF- 2 Flood Parameters and Characteristics 14
3.1 MHF-2 Standards 14
3.1.1 Standard MHF-2.B 14
3.1.2 Standard MHF-2.C 15
3.2 MHF-2 Purpose Statement 15
3.2.1 MHF-2 Purpose, First Paragraph 15
3.2.2 MHF-2 Purpose, Third and Fourth Paragraphs 16
3.3 MHF-2 Disclosures 16
3.3.1 MHF-2 Disclosure 2 16
3.3.2 MHF-2 Disclosures 8, 9 and 10 17
3.3.3 MHF-2 Disclosures 13, 14 and 15 17
3.3.4 MHF-2 Disclosure 17 18
3.3.5 MHF-2 Disclosure 19 18
3.3.6 MHF-2 Disclosure 20 19
3.3.7 MHF-2 Disclosure 21 19
3.3.8 (new) MHF-2 Disclosures moved from MHF-5 Disclosures 4 & 5 19
3.4 MHF-2 Audit Items 20
3.4.1 MHF-2 Disclosures 8, 9 and 10 to become new MHF-2 Audit Items 20
3.4.2 MHF-2 Disclosure 19 to become (new) Audit Item 21
3.4.3 MHF-2 Disclosure 20 to become (new) Audit Item 21
3.4.4 MHF-2 Disclosure 21 to become (new) Audit Item 21
3.4.5 (new) MHF-2 Audit Items moved from MHF-5 Audit Items 2 & 3 22
4 Standard MHF-3 Flood Probabilities 22
4.1 MHF-3 Standards 22
4.1.1 Standard MHF-3.B 23
4.1.2 Standard MHF-3.C 23
4.2 MHF-3 Purpose Statement 23
4.3 MHF-3 Disclosures 23
4.3.1 MHF-2 Disclosure 20 23
4.4 MHF-3 Audit Items 24
4.4.1 MHF-3 Audit Item 2 24
4.4.2 (new) MHF-3 Audit Item (moved from MHF-1 Audit Item 5) 24
4.4.3 (new) MHF-3 Audit Item (moved from MHF-1 Audit Item 6) 24
5 Standard MHF-4 Flood Profile Structure 26
5.1 MHF-4 Standards 26
5.1.1 Standard MHF-4 title 26
5.1.2 Standard MHF-4.A 26
5.1.3 Standard MHF-4.D 26
5.1.4 Standard MHF-4.E 27
5.2 MHF-4 Purpose Statement 27
5.3 MHF-4 Disclosures 27
5.3.1 MHF-4 Disclosure 1 27
5.3.2 MHF-4 Disclosure 2 28
5.4 MHF-4 Audit Items 29
5.4.1 MHF-4 Audit Item 4 29
6 Standard MHF-5 Modeling of Natural and Man-Made Flood Mitigation and Prevention Measures 30
6.1 MHF-5 Standards 30
6.1.1 Standard MHF-5 title 30
6.1.2 Standard MHF-5.A 30
6.1.3 Standard MHF-5.B and C moved to Standard MHF-2 31
6.1.4 Standard MHF-5.D 31
6.1.5 (new) Standard MHF-5 brought in from Standard MHF-6 (modeling failure of flood mitigation and prevention measures) 32
6.2 MHF-5 Purpose Statement 32
6.2.1 MHF-5 Purpose 33
6.3 MHF-5 Disclosures 33
6.3.1 MHF-5 Disclosure 1 33
6.3.2 MHF-5 Disclosure 2 33
6.3.3 MHF-5 Disclosure 3 34
6.3.4 MHF-5 Disclosures 4 & 5 34
6.3.5 MHF-5 Disclosure 6 35
6.3.6 (new) MHF-5 Disclosures (brought up from Standard MHF-6 Disclosures 1-4) 35
6.4 MHF-5 Audit Items 36
6.4.1 MHF-5 Audit Item 1 36
6.4.2 MHF-5 Audit Items 2 & 3 (moved to MHF-2) 36
6.4.3 MHF-5 Audit Item 4 37
6.4.4 (new) MHF-5 Audit Items (brought up from MHF-6) 37
7 Standard MHF-6 Modeling for the Failure of Flood Mitigation or Prevention Measures 39
7.1 MHF-6 Standards, Purpose, Disclosures and Audit Items 39
7.1.1 Standard MHF-6 39
8 Standard MHF-7 Logical Relationships of Flood Characteristics 40
8.1 MHF-7 Standards 40
8.1.1 Standard MFH-7.A 40
8.1.2 Standard MFH-7.B 40
8.1.3 Standard MFH-7.C & D 40
8.1.4 Standard MFH-7.F 41
8.1.5 Standard MFH-7.H 41
8.1.6 Standard MFH-7.I 41
8.1.7 (new) Standard MFH-7 42
8.1.8 All MHF-7 Standards – ordering of 42
8.2 MHF-7 Purpose Statement 43
8.3 MHF-7 Disclosures 43
8.4 MHF-7 Audit Items 43
9 Form MHF-1 Annual Flood Occurrence Rates by County 44
9.1 Items in Form MHF-1 44
9.1.1 Form MHF-1 Item A 44
9.1.2 Form MHF-1 Item C 45
9.1.3 Form MHF-1 Item D 45
10 Form MHF-2 Maps of Flood Profiles by Return Period 46
10.1 Items in Form MHF-2 46
10.1.1 Form MHF-2 Item x 46
10.1.2 Form MHF-2 Item x 46
10.1.3 Form MHF-2 Items C., D. and E. 47
10.1.4 Form MHF-2 Item x 48
11 Attachment with Track Changes 49
1 Introduction
AIR appreciates the opportunity to provide input into the development of the Florida Commission on Hurricane Loss Projection Methodology (FCHLPM or Commission). This document contains AIR’s input into the draft Meteorological/Hydrological Standards published on November 17, 2014.
At the outset of the development of flood standards, we are in a different position than we were in 1995 at the outset of the development of wind standards primarily because most modelers already had existing wind models in 1995. In 2014, while some modelers have flood models, no modeler has a comprehensive flood model (as the Commission currently defines it in working usage). In 1995, the wind standard development could reference the existing modeling methodologies, and over time both the methodologies and the standards evolved together.
The November 17, 2014 draft of the Meteorological/Hydrological standards makes many presumptions about specific details that are assumed to be present in a given flood model. Since the mandate of the Commission is to evaluate the acceptability of a broad range of possible models rather than to dictate how such models must be developed, the standards as written are overly prescriptive on the whole. Rather, it would be more beneficial to word the standards in language that is as generic as possible and require the modeler to fully disclose and justify the flood parameters and characteristics that are specific to their model. To this end, we have tried to identify elements of each standard that fall into this category, which we refer to as “PRESCRIPTIVE”.
AIR believes the modeling firms should develop flood models based on their own scientific expertise and the historical and claims data available. It is true that there are existing flood models in the market that capture certain features of the flood peril, but these models are not necessarily insurance flood loss projection models. Insurance loss projection modelers are always constrained by the amount and detail of insurance claims data available to support different facets of modeling. If the data is not available to properly account for a certain aspect of flood modeling, then requiring it in the standards will only encourage poor modeling practice, which goes against the mission and mandate of the Commission. To this end, we have tried to identify elements of each standard that fall into this category, which we refer to as “POTENTIALLY NO FLOOD DATA”.
Mention has been made that we have over three years until the first submissions are due, so the modelers should be able to create or update their existing models to fit the standards developed. Three years may not necessarily be enough time with flood models. In any case we are confident the Commission would agree that models built hastily or without supporting data in order to meet a set of standards that overreach the current state of the insurance flood modeling science is not in anyone’s best interest.
The remainder of this document presents AIR’s suggestions for edits to the draft Meteorological/Hydrological Standards using the required format.
2 Standard MHF-1 Flood Event Data Sources
AIR has identified several items in this Standard that we suggest be edited. The Problem Statement and Explanation sections below discuss suggested edits to the various sections (Standards, Purpose, Disclosures, Audit).
2.1 MHF-1 Standards
AIR’s suggested edits to the MHF-1 Standards are described, explained and justified below.
2.1.1 Standard MHF-1.B
· Problem Statement: In standard MHF-1.B, the word “all” in the first line of the sentence is not necessary and may be overreaching. Requiring the model to incorporate relevant data sources should be sufficient.
· Explanation: This is a general edit. Requiring the model to incorporate relevant data sources is sufficient. There is no similar use of the word “all” in the hurricane standards.
· Amendatory Language:
2.1.2 Standard MHF-1.C, first suggestion
· Problem Statement: In standard MHF-1.C, the phrase “based upon”is not consistent with similar phraseology used in the hurricane standards, so we suggest making them consistent by using “consistent with”.
· Explanation: This is an editorial change and should not change the overall meaning. This change will be consistent with similar wording in the hurricane standards.
· Ammendatory Language:
2.1.3 Standard MHF-1.D
· Problem Statement:
o In standard MHF-1.D, the first line, the word “all” is not necessary and may be overreaching.
o Also, the terms “flood plain flooding” and “non-flood plain flooding” are not standard terminology. We suggest instead the terms “fluvial flooding” and “pluvial flooding” which are more common in different parts of the world. This comment applies generally to the rest of the standards but has not always been pointed out.
· Explanation:
o Removing the word “all” is a general edit; requiring the model to encompass relevant flood event data sources is sufficient. There is no similar use of the word “all” in the hurricane standards.
o Changing the terms is a technical edit. The more common terms are fluvial and pluvial; they are also more concise and have standard definitions.
· Amendatory Language:
.
2.2 MHF-1 Purpose Statement
AIR’s suggested edits to the MHF-1 Disclosures are described, explained and justified below.
2.2.1 MHF-1 Purpose, First Paragraph
· Problem Statement: The first paragraph of the Purpose Statement is problematic because it could be interpreted as dictating more frequent model updates than is feasible. Some types of flood data get updated every year or more frequently. Incorporating new data after flood events into a model update requires more than a year.
· Explanation: Removing this first paragraph is a general edit. We agree that models should be kept up to date, but it may be too early in the flood modeling lifecycle to dictate what the update frequency should be. At minimum for now we recommend the Commission should not require less than a 5 year update frequency.
· Amendatory Language:
2.3 MHF-1 Disclosures
AIR’s suggested edits to the MHF-1 Purpose Statement are described, explained and justified below.
2.3.1 MHF-1 Disclosure 1, 2 and 4
· Problem Statement: The term “Flood Event Data Sources” (a parallel to the Base Hurricane Storm Set), if used, should be defined in a very general way, so as to not dictate particular data sources. Given that, it may not be necessary or ideal to use a defined term, when something like “relevant data sources” will suffice.
· Explanation: Removing the defined term “Flood Event Data Sources” is an editorial change. It is appealing to use a defined term because it simplifies references to it, but in this case there is probably no easy way to define the Flood Event Data Sources. The Commission should not look to the modelers to assist with such a definition now because not every modeler has created flood models yet. It migh be best to wait until the 2019 ROA revision or later to create a defined term if necessary.
· Amendatory Language:
2.3.2 MHF-1 Disclosure 5
· Problem Statement: There are several problems with this disclosure.
o First, the flood occurrence rates may differ for the different types of flooding.
o Second, the concept of an occurrence rate for a county does not make sense. The data used for understanding flooding are from gauging stations. It makes more sense to calculate flood occurrence rates in terms of specific gauging stations, or perhaps in terms of a collection of gauging stations, though aggregating across gauging stations may not produce a very meaningful result..
o Finally, the request to show “data for the most recent 5, 25, 50, and 100 years” does not make sense from the perspective of occurrence rates. We believe the Commission might instead be referring to various probabilities of exceedance corresponding to specific return periods.
· Explanation: These suggested edits are technical in nature.
o The first edit suggested edit is general. We suggest adding in “storm surge, fluvial and fluvial flooding” to allow for the reporting of these rates separately if necessary.
o Second suggested edit is technical in nature. We suggest changing the wording to allow the modeler to report the occurrence rates for identified locations (i.e. where there are gauging stations) in a region or for the region as a whole.
o Finally the third suggested edit is technical in nature. The wording “Provide annual historical and modeled flood occurrence rates for the most current 5, 25, 50, and 100 years” is not precise or clear. We suggest clearer wording to be “Provide annual flood occurrence rates. . . which include data for various probabilities of exceedance (corresponding to 5, 25, 50, and 100 year return periods).”
· Amendatory Language:
Note: these same edits have been suggested in Form MHF-1.
2.4 MHF-1 Audit Items
AIR’s suggested edits to the MHF-1 Audit Items are described, explained and justified below.
2.4.1 MHF-1 Audit Item 1
· Problem Statement: See section 2.3.1 for the problem statement relating to “Flood Event Data Sources”.
· Explanation: See section 2.3.2 for the explanation relating to “Flood Event Data Sources”.
· Amendatory Language:
2.4.2 MHF-1 Audit Item 5
· Problem Statement: This audit item deals with modeled flood probabilities; Standard MHF-1 is about flood event data sources so it doesn’t seem to be in the right place.
· Explanation: This is an editorial change. We suggest moving this audit item, dealing with modeled flood probabilities, to MHF-3 which addresses flood probabilities.