Winkler v. Great American Ins. Co., 447 F. Supp. 135 (E.D.N.Y. 1978)

Lester E. Fetell, Brooklyn, N. Y. (Sergi & Fetell, Brooklyn, N. Y., of counsel), for plaintiffs.

John M. Speyer, New York City (Greenhill & Speyer and Michael E. Gorelick, New York City, of counsel), for defendants.

DOOLING, District Judge.

Plaintiffs, owners of a summer property on the oceanfront at Robins Rest, Fire Island, Town of Islip, sued the defendants under a flood insurance policy issued pursuant to the National Food Insurance Act of 1968 for the direct loss by "flood," as defined in the policy, sustained by their property. See 42 U.S.C. §§ 4001, et seq. Jurisdiction of actions to enforce claims for loss under such policies of insurance is granted to the district court for the district in which the insured property is located. 42 U.S.C. § 4053.

In April 1972 plaintiffs bought for about $55,000 a two-story summer cottage located at Robins Rest. Robins Rest is the area just west of Ocean Beach. The evidence is that in the late winter or early spring of 1972 a stone jetty or groin had been built just east of plaintiffs' house at the west end of Ocean Beach. As measured off on the map, Exhibit 2, the jetty would appear to have been in the order of 1800 feet to the east of plaintiffs' house. The house was not the nearest house to the ocean. The Looney house, was just east and wholly south of plaintiffs' house, and, thus, nearer to the ocean by the length of the beachfront building lot.

The Winkler house rested on ten foot locust posts or piles, and it stood on a dune that rose 11 feet above sea level. The first floor elevation of the structure itself was 1½ feet above the dune level. Between the house and the wholly sandy part of the beach there was a stretch about 30 to 40 feet wide of sand dune covered with salt *136 grass and perhaps scrub pine. The plot on which the house stood measured 112½ feet east and west along the ocean front and 50 feet in depth. The house was not a new house; it was some years old; it is shown as having been in place in the September 1, 1968 survey map, Exhibit 3.

Plaintiffs applied for flood insurance under the Act on September 12, 1972 and insurance was granted to them on or about September 22, 1972 in the amount of $17,500 on the dwelling and $5,000 on the contents. (Later the amount of coverage was apparently doubled.) The policy was effective for one year from September 27, 1972, and the premium for the year's coverage was $84. The estimated actual cash value of the dwelling exclusive of the contents was given as $45,000 and the estimated actual cash value of the contents was given as $20,000.

In or just before October 6, 1972, plaintiffs decided to build an extension at the west end of the existing two-story building to contain on the first floor a bedroom, laundry, bathroom and store room and, on the second story, a dining room about 11½ feet by 15 feet; a second story sun deck was to extend toward the ocean leading out from the dining room. A walk along the west edge of the new extension would extend beyond its front wall, and stairs were planned to go from that point to the sun deck, which would extend toward the ocean beyond the front line of the rest of the structure. As originally planned the building would have rested on poured concrete piers about 5½ feet in overall length resting in the dune sand, about 4 feet being below grade and a foot and a half above grade. Except at the very center of the planned addition the piers would have been on 4 foot 8 inch centers. However, the front row of piers, those nearest the ocean were 6 feet 6 inches in front of the next in-shore set of piers. The next further in-shore set of piers was 2 feet 4 inches from the next row, which was the row planned to support the front wall of the addition. The two front rows of piers were to support a walk at the first floor level along the front wall of the addition, and to support posts which would support the sun deck above. The outermost set of piers had no other evident function beyond supporting the outer edge of the elevated sun deck.

Plaintiffs applied to the Town of Islip for a zoning permit on October 6, 1972, and it was granted on October 27, 1972. The estimated value of the work was given at $4,800. The contractor for the work was Tom Hinderlang, Jr., who lived in Robins Rest not far from plaintiffs, had been in the construction business for 20 years and had been living on Fire Island for 7 years.

By mid-December the concrete piers were all in place. The second story of the addition was enclosed and most of the first story was enclosed except for the front wall at the east end nearest the old construction where only the studding was in place. The window frames had been installed, with glass. No part of the sun deck was in place, nor was any part of the walkway that would have extended across the first floor of the addition.

Plaintiffs visited the building site frequently as construction went forward. They were at the property on Sunday December 23d. They found that the Looney house, east and south of theirs, had disappeared, and that a part of the porch of the Karr house, next east of them, had been damaged and was sagging. Of the 30 or 40 feet of dune that had been in front of their own house little was left except an irregular section that plaintiff Irving Winkler recalled as being as narrow as 2 feet in some places and as wide as 7 feet in others but which, on all of the evidence, seems to have been somewhat wider than that; apparently its width was in the order of a dozen feet at points. There was still salt grass and what looks like scrub pine on top of the dune.

What, if anything, had happened to plaintiffs' house is controversial in the extreme. It is plaintiffs' insistence that the ocean water had somehow reached the front line of piers and had toppled one of them, displaced to some extent three others, and left one standing but apparently not sound. It *137 is claimed that by December 23d the front piers had been integrated into the rest of the building (although the plans, Exhibit 8, did not call for that) in accordance with a change in plan which visualized, in effect, an extension of the walk adjacent to the front wall of the addition all the way out to the front line of piers.

There is no question that the piers were put in place to be an integral supporting part of the completed structure whether they were to support a first floor extended walk or porch, or simply to help to carry the weight of the second floor sun deck. On all the evidence, however, it cannot be found as a fact that the front line of piers had been in any way framed into or tied into the rest of the building fabric. There is no concrete physical evidence of a tie-in of the piers to the rest of the addition by timbers or braces. There is no indication on the visible side of the addition, to the extent shown in the photographs in evidence, that there was anything into which timbers reaching the piers could have been tied. Three supporting timbers protrude from the front of the addition but they are not of a length sufficient to have reached the piers in question. It is, finally, difficult to visualize any kind of wave or water action that would have stripped away the framing and left the piers in the condition in which they are seen in defendants' Exhibit A, a photograph taken on February 6.

What had happened to the property presents a difficult question. There appears to be no question that enough of the dune was removed by water action of some kind so that the piers were exposed, although the photograph of February 6, 1973 shows an exposure of them which, in the light of some of the descriptions of the events of December 23, might have been expected to be very different. As they appear in that picture and it appears to be the only picture definitely dated to the period before the house was moved one was thrown down, three are visible, one can be conjecturally supplied at the west end of the picture, and only one appears to be exposed almost to its foot. The picture does show the ocean-ward edge of the dune as a low scarp of the kind caused when wave water undermines the foot of a dune, causing the sand above to collapse into the water, to be transported away, leaving a steep, receding scarp.

There is no evidence that water washed over the dune to reach the piers of the house. The meagre evidence afforded by the picture of February 6th, Exhibit A, is inconsistent with the sea water's having washed over the dune material. Both the expert evidence and common knowledge are persuasive that, had the wave water gone over the dune, it would not have left the well-defined scarp visible in the picture; there is, indeed, explicit testimony that water was seen under the house and in the area around it, but this evidence can only have meant that the mark of water was visible, for the sand could hardly have retained the water so long if any had intruded. The evidence, so interpreted, is of doubtful significance, and remains, in any event of dubious credibility.

It is clear that plaintiffs very sensibly wasted no time in taking steps to protect their property. Some little time at least before December 31st, plaintiffs had been in touch with Hughes Contracting Company, and Hughes under date of December 31st furnished an estimate of $7500 to install thirty-nine 8 inch by 10 inch 35 foot long piles, the tops of the piles to be about 5 or 10 feet above the concrete walk, and to construct a 3 inch by 12 inch bridge at an elevation specified in an accompanying plan. As to the date of the work, the estimate said,

"the above work to be done as soon as house is moved from present location."

The copy of the estimate produced in evidence is approved by plaintiff Irving Winkler and has a note on it to the effect that it was mailed to Hughes on January 9th.

The credible evidence is that plaintiffs, consulting with Hinderlang and with Davis Bros. Engineering Corp., decided to move the house off the ten foot locust posts, replace them with 35 foot piles, and then return the house to its original site. Plaintiff *138 Irving Winkler anticipated that if he did not do so, the house would be lost, that he could not hope to keep it at the present site unless he had piers driven into the ground deep enough and extending high enough above ground so that the house would be beyond the reach of the water even if it inundated the entire building site.

In the circumstances his decision was a sensible one. It was general knowledge, or at least the general belief, that the winter storms were worst in February and March, and that the winter season was the time of year in which the Fire Island beaches lost sand to the sea, sand which was ordinarily replaced by the sea in the summer. Most of the dune in front of the Winkler house had already been lost, and, of the two houses to the east of it, one had been lost and the other damaged. The evidence warrants the conclusion that the local people regarded the season as a bad one from the point of view of storms, although the scientific data do not support that belief.

There is no evidence that at the time immediately in question there was a storm of extraordinary severity. Neither does the evidence indicate that there were any unanticipatably severe storms in any part of the period in question. However, it was a period in which, no matter how active the storms may have been, there was loss of beach and dune in this part of the south shore of Fire Island.

The sequence of photographs shows both the narrowness of the beach itself and the characteristic scarp effect produced by the undermining of the sand dunes resulting from the classic type of collapse into the water of the upper portions of the dunes, leaving a low sand cliff.

The wind and water data in evidence are not altogether easy to interpret. It appears to be more or less agreed that Exhibit 7 does not give wind and wave direction and height for the ocean front but, possibly, for the Great South Bay. The wind data and direction, however, would, as the map of the island and the bay north of it indicates, be generally common to Great South Bay and to Fire Island. The 1971 Fire Island data for December (Exhibit 7) show west northwest wind as high as 32 knots on one day. Wind at 30 knots was recorded on December 31st, and winds in the 20s were not infrequent. But in general the winds were prevailingly less than 20 knots and, for a good part of the month, less than 10 knots. In November 1972 winds reaching 19 and 20 knots were recorded on the 4th and 5th of November and on November 8, in rain, winds at 45 and 40 knots are recorded and on the following day winds of 25 and 20 knots were noted. On November 14, in rain, winds at 30 and 35 knots were recorded, and on November 20 winds at 25 knots, in rain. On November 24 winds were recorded as high as 24 knots, on November 26 winds reached 28 knots, and on the 27th 30 knots; on the 28th winds were at 22 knots and on the 30th reached 28 knots.

The wind speed data for December 1972 are as follows (wind speeds are in knots):

            Highest             Highest
  Date    Wind Speed   Date   Wind Speed
   1          28        16        32
   2          20        17        38
   3          20        18        23
   4          20        19        11
   5          14        20        15
   6          25        21        21
   7          25        22        25
   8          15        23        18
   9          20        24        15
  10          12        25         9
  11          20        26        10
  12          12        27        20
  13          25        28        21
  14          10        29        13
  15          40        30        10

There appears to be general agreement that removal of sand from Fire Island beaches is brought about by varying conjunctions of winds, waves and tides. In general, Fire Island beaches tend to lose sand to the sea in winter and to regain it in summer. Characteristically, the drift along the south shore of Long Island is from east to west; sand picked up at one point along Fire Island is generally transported westward and deposited to the west. At any one place on Fire Island the removal and deposition of sand would depend along with the other factors on the total topography of the particular place. The winter storms, *139 that is those which can be expected from November to April, are factors which accelerate the removal of sand from the beaches; in the summer time it is the circumstance that the waves are longer and slower that accounts for the deposit of sand on the beaches. Building a rock jetty or groin jutting out into the ocean has the effect of collecting sand from the passing current of water on the easterly side of the jetty. As the drift along the shore toward the west takes place and the sand is deposited at the east side of the jetty a compensating effect is produced on the west side of the jetty, which surrenders sand to the eastward current.

A distinction is drawn between the terms erosion and overwash. Erosion includes any removal of sand to the water which takes place by undermining the dune structure, creating the typical scarp or sand cliff. An overwash occurs when the waves of water wash over the dune and carry the water to a point back of the original dune line. Overwashes often occur during hurricanes; an overwash can create an overwash terrace and deposit rather than withdraw sand.

Both expert witnesses were of the opinion that the pictures in evidence depict classic beach erosion through wave action. Neither of them saw any evidence of overwash effect.

Tidal effect is not significant at all in the loss of sand since tides respond to constant astronomical forces. Only storm tides are significant in loss of beach sand. The storm tide is the excess blown up by the wind on top of the basic astronomically determined tide, and its height depends on the direction of the wind, whether it is on-shore or off-shore. On December 20 the wind was west southwest, then southwest, then northeast, then north-northeast and then northeast. The wind stayed in the northeast until the 22d. The wind directions and speeds as recorded at three hour intervals during the 21st, 22nd and 23rd of December are shown in Annex 1.

Sand dunes are by their nature temporary. An effort is made to hold them in place through grass plantings, sand fences and by other means, but they are inevitably subject to wind erosion to some extent and, mainly, to being undermined by waves. Flooding of the dunes, that is, in the sense of completely overwashing them, is unusual except during hurricanes since most strong winds are not strong enough to produce an overwash effect.

From an overall long-term point of view the December 1972 period was not exceptional. In the five years through December 1971 there were eighteen cases of storms as great as or greater than that of December 23d and the historical summary of storm evidence in the North Atlantic book indicates that the principal December storm track was well off shore of Fire Island. Exhibit U, Chart 107 summarizes the wind direction and velocity data. While there are some problems of legibility, Annex 2 attached tabulates the data from the chart, using the concentric pattern that is centered on the new Jersey shore just southwesterly of Long Island.

The flood insurance policy, expressly issued pursuant to the National Flood Insurance Act of 1968, states that the insurance Company, a member of the National Flood Insurers Association,

". . . DOES INSURE the Insured . . . to the extent of the actual cash value of the property at the time of loss, but not exceeding the amount which it would cost to repair or replace the property with material of like kind and quality within a reasonable time after such loss . . . against all DIRECT LOSS BY `FLOOD' as defined herein, to the property described while located or contained as described in the application and declarations . . . or pro rata for 30 days at each proper place to which any of the property shall necessarily be removed for preservation from the peril of `Flood', but not elsewhere."

The policy defines the term "flood" as meaning

". . . a general and temporary condition of partial or complete inundation of normally dry land areas from (1) the *140 overflow of inland or tidal waters, (2) the unusual and rapid accumulation or runoff of surface waters from any source, or (3) mudslides which are caused or precipitated by accumulations of water on or under the ground."

The term flood is also defined in the statute, 42 U.S.C. § 4121(a) (1) in the following language

"(1) the term `flood' shall have such meaning as may be prescribed in regulations of the Secretary, and may include inundation from rising waters or from the overflow of streams, rivers, or other bodies of water, or from tidal surges, abnormally high tidal water, tidal waves, tsunamis, hurricanes, or other severe storms or deluge . . ."

Shortly after the Act became law it was extended to cover inundation from mudslides, which were included within the term flood, and by a still later amendment, that of December 31, 1973, a new subdivision (c) to Section 4121, added the following extended definition of flood.

"The term `flood' shall also include the collapse or subsidence of land along the shore of a lake or other body of water as a result of erosion or undermining caused by waves or currents of water exceeding anticipated cyclical levels, and all of the provisions of this chapter shall apply with respect to such collapse or subsidence in the same manner and to the same extent as with respect to floods described in paragraph (1), subject to and in accordance with such regulations, modifying the provisions of this chapter . . . to the extent necessary to insure that they can be effectively so applied, as the Secretary may prescribe to achieve (with respect to such collapse or subsidence) the purposes of this chapter and the objectives of the program."

Under the policy the excluded perils, among others, are losses caused by (1) rain, snow, sleet, hail or water spray, or (2) by freezing, thawing or by the pressure or weight of ice or water except where the property covered has been simultaneously damaged by flood, or (3) water, moisture or mudslide damage of any kind resulting primarily from conditions solely related to the described premises or which are within the control of the insured. By exclusion (d) it is provided that the company shall not be liable for loss

"by fire, windstorm, explosion, erosion, earthquake, landslide or any other earth movement except such mudslides as are covered under the peril of flood, or by theft."

The policy grants permission to make alterations, additions and repairs and then continues:

"In the event of loss hereunder, the Insured is permitted to make reasonable repairs, temporary or permanent, provided such repairs are confined solely to the protection of the property from further damage and provided further that the Insured shall keep an accurate record of such repair expenditures. The cost of any such repairs directly attributable to damage by the peril insured against shall be included in determining the amount of loss hereunder. Nothing herein contained is intended to modify . . . the requirement that in case loss occurs the Insured shall protect the property from further damage."

In the policy headed "Requirements In Case of Loss" it is provided that the insured shall give written notice of loss and

"The insured shall . . . protect the property from further damage, forthwith separate the damaged and undamaged property and put it in the best possible order."

In the part of the policy headed "Property Covered" the term "dwelling" is defined and it is then said that when the Insurance under the policy covers a dwelling

". . . such insurance shall include additions in contact therewith;"

Under the heading "Deductibles" it is provided that "With respect to loss to the dwelling . . . this Company shall be liable for only the amount of loss in any one occurrence which is in excess of . . . (b) 2% of the amount of insurance hereunder applying to the dwelling . . ." It *141 is agreed that the loss or damage directly related to the row of piers which were displaced did not exceed the deductible amount, which was taken as being $350.

Plaintiffs argue that the damage and threat to their house on Fire Island was the most likely and characteristic damage and loss from the action of sea water that was to be anticipated for Fire Island and similar ocean-front properties, and that, in consequence, it is absurd to suppose that the statutory insurance was not intended to cover exactly the risk of damage and loss to which their and all similar property on the beach front was exposed. They argue that the one word "erosion," buried in a list of exclusions having nothing to do with water or wave action, cannot be used effectually to nullify coverage for the most typical of the oceanfront wave-action risk by the legerdemain of distinguishing between the ways in which the typical oceanfront catastrophe occurs, that is, by confining coverage to the cases in which the ocean waters flow over the dune grass and denying coverage if the ocean waters produce the same effect by an undermining action technically called erosion. Cf. Pfeiffer v. General Ins. Co., N.D.Cal.1960, 185 F. Supp. 605.

The argument fails because the policy and the statute are addressed specifically to flood and not to other events which may be even more frequent and equally damaging. The modest premium suggests that the coverage is against a risk of infrequent occurrence and not a risk that is almost a condition of Fire Island existence. The term "flood" as defined in the policy has as its critical words "partial or complete inundation". That too, occurs on such beachfronts as Fire Island's, but, as the testimony indicated, usually in connection with hurricanes. The evidence in the present case does not show that any flood occurred, or that flood occasioned the loss that did occur to plaintiff's house.

The case, thus, does not turn on the force and extent of the word "erosion" in the exclusions. The word erosion in the exclusion, however, reinforces the conclusion that there is no coverage. It is perhaps unnecessary, therefore, to seek to decide whether, if it stood alone, it would suffice to exclude liability for the loss. It must be concluded that it would suffice.

Plaintiffs argue that the word erosion is distorted out of its usual meaning when it is applied to the kind of massive erosion that can occur along Fire Island's oceanfront. Plaintiffs point to evidence of a storm having deposited 45,000 yards of beach material while eroding 285,000 yards of such materials. But the word erosion when it is used as an exclusion in a flood policy cannot be denied its intrinsic meaning, expressive of a contrast to inundation. Its operation is from moment to moment gradual and not massive, but its cumulative effect can, as is well known, change the geography of a state, or threaten a house with collapse. Indeed, the sharp distinction between erosion and inundation marks exactly the circumstance of gradualness that gave plaintiffs the opportunity to save their house by moving it, providing it with a new supporting substructure, and then restoring it to its original site. An ocean flood inundating the property would have afforded no such opportunity.

Plaintiffs argue that it is surely plain that they at least suffered a covered loss in the damage to the piers forming the front rank of the foundation for the extension. Having sustained that loss, they argue, they were at least permitted to make reasonable repairs confined to the protection of the property from further damage, and moreover were under the more general duty under the policy to protect the property from further damage. Since there was no flood damage the argument cannot be supported.

Even if there had been a flood damage covered by the policy, defendant argues that the piers were not part of the dwelling covered by the language of the insurance policy. To adopt that argument would appear to defeat the intention to cover the dwelling property in a normally inclusive sense. Other language in the same clause visualizes that a covered property may be a structure in course of construction, and it *142 would be unreal to think that only such parts of the construction as had been tied together by courses of brick or beams would be within the coverage. To answer defendant in a purely semantic way, the piers were an integral part of the building structure from the moment they were set in place notwithstanding that they had not yet been linked to the sun porch which they were meant to support. But the point does not advance plaintiffs case because the cost of restoring the piers does not exceed the $350 amount set out in the deductible clause of the policy.

The real loss was that incurred by plaintiffs in moving the building and moving it back again, and, quite separately, the cost of furnishing an adequate substructure to support the building safely in future. What plaintiffs did was sensible and practical. Plaintiffs' duty to protect the property from further damage, however, does not under the express language of the policy entitle them to recover the cost of that protection from the insurer. The policy, oddly, is clear that the property continued to be covered for at least 30 days at each place to which it was removed for preservation from the perils anticipated. But only in the alterations and repairs clause does there appear an authority to make repairs solely for the protection of the property from further damage; such repair costs, if directly attributable to damage by the peril insured against, are recoverable expressly from the insurer.

Is it, nevertheless, possible to say that the duty to protect the property from further damage implies a responsibility on the insurer's part to pay for the cost of reasonable protective measures to the extent of the insurers interest in the risk? Such a duty has been implied, and it would have appeared reasonable to have done so here if there had been assurance that the property was moved to avert further damage from a covered peril. See Slay Warehousing Co. v. Reliance Ins. Co., 8th Cir. 1973, 471 F.2d 1364, 1367-1368; Ross v. Employers' Liability Assurance Corp., N.D.Miss.1968, 290 F. Supp. 569, 576-577; Harper v. Pelican Trucking Co., La.App.1965, 176 So. 2d 767, 772-774; Hatley v. Truck Insurance Exchange, 1972, 261 Ore. 606, 494 P.2d 426, 433; cf. Thornwell v. Indiana Lumberman's Mutual Ins. Co., 33 Wis.2d 344, 147 N.W.2d 317. But see Port Murray Dairy Co. v. Providence Washington Ins. Co., 1958, 52 N.J.Super. 350, 145 A.2d 504, 508-509. Even if the implied duty were accepted in this case, such an implied undertaking to compensate for protection costs would extend only to imposing on the insurer responsibility for its share of the risk, for plaintiffs were coinsurers. The values involved in the present case would appear to have been (assuming that the additional insurance is effective despite the date of the application) in the ratio of 35 to 45. The cost of moving the house from the original locust posts to the new locations and returning the house to the original site was, on the evidence, $5100. There was evidence allocating $4,595.28 of the total construction, reconstruction and repair costs to repairs necessitated by the moving of the house. It would not under any circumstances be proper to include any part of the plaintiffs' costs of the new sub-structure, or any of the other related expenses, to the "protection from further damage" category.

But again the difficulty with plaintiffs' case is that the house was not moved to protect it from an imminent risk of the peril insured against, that is, flood damage, but from the familiar and much less immediate risk of the long winter's wear on the beach (cf. Young's Market Co. v. American Home Assurance Co., 1971, 4 Cal. 3d 309, 93 Cal. Rptr. 449, 454, 481 P.2d 817, 822); the erosion was evidently encroaching farther and farther on plaintiffs' fragile dune. Had plaintiff been taking steps to protect against a flood, their procedure could not have been so slow paced; indeed it would almost certainly not have been feasible, for floods, characteristically, do not give a warning sufficient for the removal of buildings. It follows that in this case there can be no recovery.

It is accordingly

*143 ORDERED that the Clerk enter judgment that the plaintiffs take nothing and that the action is dismissed on the merits.

Date and             Wind            Wind
  hour             Direction        Speed
Dec. 21
00:30                 NE              18
03:30                 NE              18
06:30                 NE              17
09:30                 NE              21
12:30                 NE              20
15:30                 NE              15
18:30(?)              NE              18
21:30                 NE              17
Dec. 22
00:30                 NE              20
03:30                 NE              20
06:30                 NNE             25
09:30                 NNE             20
12:30                 N               20
15:30                 N               14
18:30                 N               14
21:30                 N               12
Dec. 23
00:35                 NNE             12
03:35                 NNE             14
06:30                 NNE             15
09:30                 NNE             18
12:30                 NNE             18
15:30                 NNE             18
18:30                 NNE             14
21:30                 NNE             15

1. Percentage distribution of DECEMBER winds by wind velocities:

    Percent
  of winds at       Beaufort     Miles per
indicated force      Scale         hour
       6               2           4-7
      10               3           8-12
      22               4          13-18
      17               5          19-24
      20               6          25-31
      13               7          32-38
       9               8          39-46
       ?               9          47-54

2. December wind directions and velocities:

   Beaufort
     Scale
   Knots
Wind  % of time  2-3    4-5    6-7   8-12
     in quarter  4-10  11-21  22-33  33 and over
 N           9     1          5    3         +
 NE          4     1          2    0(?)      +
 E           4     1          2    1         +
 SE          7     2          3    2         +
 S          14     2          5    5          2
 SW         11     3          3    3          1
 W          24     4          8    8          4
 NW         20     2          9    7          2