A lesson on water boundaries: Part II.

Over considerable time the court has developed important doctrines to deal with water boundaries that provide guidance to the land surveyor and attorney when confronted with a water boundary determination. These doctrines have been categorized into those applicable in an ambulatory setting, wherein the boundary is affected by gradual movement of a river; and avulsive circumstances, which operate to supplant the ambulatory nature for a given event, and thus fix the boundary at its location just prior to the event.

Doctrines Affecting Boundaries


Accretion is the gradual process of attachment to riparian or littoral land through deposition of sand, gravel or sediment from the action of contiguous waters. The material deposited is alluvium. The term alluvion refers to the process of accretion, both terms sometimes being used synonymously.

Of key importance to any discussion of an ambulatory setting are the terms “gradual” and “slow and imperceptible.” It is from the definition of these terms that a distinction is made between an ambulatory boundary (one that moves with migration of the ordinary high water line, or OHWL), and an avulsive event (that does not cause a boundary shift for that event).

If one can stand, watch and observe, either the eroding away of material or deposition of same, visibly and moment by moment, then the court would likely rule for an avulsive event. However, if detection of change is only possible over time by comparison to some point of physical reference, then the court would likely rule in favor of accretion.

A riparian landowner bounded by a stream or river (navigable or not), the banks of which are moved or reconfigured by the gradual process of accretion or erosion, continues to hold to the watercourse as his boundary. If the land is increased, he benefits (a right) from the gain. If the land is diminished, he has no recourse (a risk) for the loss sustained.

A question that frequently arises in connection with accretion is whether the change was brought about through natural or artificial causes. The original notion for this type of change was predicated on natural causes. However, as man has infringed on his environment more heavily, this thinking under the law has necessarily evolved. The reader is cautioned that the answer to this question can vary from state to state.1

The doctrine of accretion has been applied to circumstances of both natural and artificial causes (see County of St. Clair V. Lovingston, supra, at 64-69; United States v. Claridge, 416 F. 2d 933 (CA9 1969)). Where accretions to riparian land are caused artificially by strangers to the land, the upland owner remains the beneficiary thereof (see Bonelli Cattle Co. v. Arizona, 414 U.S. 313 (1973)). The court has taken the position that one cannot artificially induce accretions to one’s own benefit. However, if in the course of exercising riparian rights one causes unknowingly and absent collusion, a deviation from “natural conditions” that results in accretive action to a downstream or upstream riparian parcel, the court has most usually allowed the benefit of the attached material to stand.


Erosion is the wearing or washing away of soil from the banks of a watercourse or shore of a water body through the water’s action. Sand, sediment and other materials are gradually dislodged, transported and deposited elsewhere. Erosion is the opposite process of accretion.

Evidence of erosion appears as cut banks and scarping in the case of a watercourse, or benching in a littoral setting (to be discussed in Part III). In river mechanics studies it has been determined that the meandering, or classic curvilinear configuration, of a watercourse begins with upstream bank erosion. As material is eroded away and moved downstream, the process alternates between scouring and deposition, increasingly enunciating the curvilinear configuration. Water velocity increases on the outside of a bend where channel depths and scouring are the greatest. Conversely, on the inside of a bend, water velocity is slowed, which enhances deposition of suspended material from upstream erosion.

In the case of a lake, wave action causes differing levels and extent of scarping (creating benching) as material is moved (eroded) from the shore. The process is one of alternating scouring and deposition.

Comments made previously with regard to slow and imperceptible change apply equally to erosion. Thus, a riparian boundary line may be considered “live” or ambulatory in terms of its ability to gradually move through erosional processes. Generally, issues of natural vs. artificial cause previously commented on with regard to accretion apply for erosion as well. The reader is cautioned that rulings on these matters can vary from state to state.


Reliction is where waters of a watercourse or water body have gradually receded, and previously submerged soil has been exposed as a result. Generally, land that was once submerged and is now left dry through the relictive process belongs to the riparian or littoral owner from whose shore the water has receded.

Reliction is a process also regarded as being opposite of accretion, wherein the result is identical (land has been added to the adjoining riparian or littoral owner), but the cause is through retreat of the water rather than deposition of material. Sometimes referred to as dereliction or emergence, the relictive process is operative regardless of whether the waters are navigable or non-navigable, riparian or littoral. Reliction does not apply under temporary drought conditions or seasonal fluctuations in water level; results must be more long-term than these.

The question of whether the reliction must be permanent and irreversible is an interesting one. Keep in mind we are discussing “living” boundary doctrines within the context of ambulatory settings. Boundaries are subject to move under these doctrines, reliction being just one cause. There are cases where the water has receded and subsequently risen, sometimes many years later. For example, in the matter of Devils Lake, North Dakota, a class action suit (423 NW 2d. 141 (1988)) was brought by a group of upland owners fronting on Devils Lake. The North Dakota Supreme Court ruled that relictive processes are subject to reversal, thus affirming the doctrines of accretion and reliction.

A method of determining if the relictive process has begun is often necessary. Under most circumstances, one can defer to the presence of various evidences for determining the OHWL (e.g., the appearance of upland vegetation) as a guide in ascertaining whether the relictive processes are underway. Remember these are evidences of slow and imperceptible change over long continued years. Historical records, photographs and testimony can also be included in the bundle of evidence for consideration.

As with accretion, certain jurisdictions will recognize reliction as being operative under both natural and artificial settings. Others, however, will not allow reliction to hold under man-caused circumstances. State rulings may not apply if the uplands are federal lands. It is extremely important to check both state and federal case law within your jurisdiction for guidance.

Apportionment of relicted lands is probably more commonly experienced in littoral settings. Once again, no single approach to apportionment can be applied in all settings and still satisfy the equity issues, which are paramount. Where government lotting adjoins a relicted area and no patents have been issued, the apportionment solution may well be a completion survey wherein the aliquot part lines are extended across the relicted area.

Partitioning of Accreted Lands

Where accretion occurs, it frequently attaches to more than one parcel. The question then arises as to where the boundary line is over the deposited soil between the two or more adjoining parcels having benefited from the accretive action. Identification of the extent of accreted material is an important first step, requiring field, and often historical, investigation.

Methods of equitably partitioning accreted lands between adjoiners are very similar to those applied in cases of reversion of streets and highways. In either instance, there is no one method that works under all circumstances. This is due to the infinite configuration possibilities inherent to water boundaries. Common methods of partitioning include, but are not limited to, the straight line extension method, the perpendicular method and the proportionate shoreline method. For an excellent discussion of these and other methods, with illustrations, refer to James A. Simpson’s River and Lake Boundaries, Plat Key Publishing, Kingman, Arizona. Keep in mind that this is an area where the court looks very closely at the equities of the matter.

An illustration of re-emergence.

The concept of re-emergence is perhaps best explained with a basic illustration. Referring to Figure 1 on page 47, in simplest form, Parcel A fronts on a navigable river and is therefore a riparian parcel. Over considerable time, the river moves southerly and eventually engulfs all of Parcel A, extinguishing same. There is no question that avulsion has not been involved. Parcel B, originally remote to the river, now adjoins the river. Subsequently, the river begins to slowly and imperceptibly move again, this time back to the north and, in the process, deposits material through accretion to Parcel B until the former position of the common boundary between Parcels A and B is exposed. This process continues, exposing a good portion of the former Parcel A.

The question that arises in this hypothetical example is whether Parcel B’s property has enlarged as a result of this northward movement of the river (as would be the case in a normal riparian setting), or whether Parcel A has now reappeared or re-emerged. Remember that Parcel B was initially a non-riparian, upland parcel. This question has been considered on a number of occasions, with no singular doctrine emerging. What has resulted is basically two opposing legal concepts.

One concept is known as the “once riparian, always riparian” rule. Here, Parcel B in our illustration would become a candidate to acquire any subsequent accretions. This once-upland parcel then becomes riparian and remains so (unless also wholly extinguished), while the former Parcel A remains permanently extinguished. Some states have even passed legislation statutorily adopting the “once riparian, always riparian” rule.

The second concept recognizes re-emergence and, in so doing, dictates that a remote, upland parcel owner can only accrete out to the limit of his former common boundary with the adjoining riparian owner. Any accretion beyond this point constitutes re-emergence of the previously extinguished, former riparian adjoiner. Some states, primarily through case law, have endorsed the re-emergence rule. The federal government has historically held to the re-emergence rule where federal lands are involved. Probably the most notable case in support of this is Towl v. Kelly and Blankenship, 54 I.D. 455 (1934). Federal lands that have eroded and subsequently been restored have been reclaimed by the government under the Towl ruling.

Needless to say, state courts have concurred with the federal government in some instances, and in others not. Interestingly, a case of relatively recent time counters the Towl ruling on what might be described as a technicality. In Ralph F. Rosenbaum, et al., 66 IBLA 374, 89 I.D. 415 (1982), an earlier case (Earl Hummel, 44 IBLA 110) was cited wherein an administrative law judge stated that re-emergence applied only in littoral settings because there the same soil was first inundated and then exposed, or re-emerged. In a watercourse, he reasoned, soil was first eroded away and then replaced through deposition by other (different) material in an accretive process. Thus, out of Rosebaum the long-standing principle of re-emergence was seemingly upset, leaving somewhat open to question the federal government’s position. Bear in mind, however, the particular question this raises applies where federal lands are involved. Recall the caveat where adjoining federal lands are at issue in a water boundary problem. Where private lands are involved, state law will normally govern.

Generally, case law across the country has been consistent that a parcel cannot re-emerge on the other side of the watercourse. This would be in alignment with matters of equity. Referring again to our illustration in Figure 1, while Parcel A began losing to erosion, Parcel C was gaining from accretion, thus maintaining its riparian status. To say that because the river began to retreat back northward and stopped, for example, such that former Parcel A was cut in half, and therefore had re-emerged both southerly and northerly of the present river’s position, would clearly sever Parcel C from riparian status. Accordingly, the courts have only recognized re-emergence as being applicable from the side of the river where re-emergence initiates.

Doctrine Where Boundary Not Affected

Avulsion is when an uncharacteristically sudden, often catastrophic and perceptible change occurs in a watercourse. Though usually associated with riparian settings, avulsive action has been recognized in littoral environments as well. Sometimes referred to as “revulsion,” avulsive action is not restricted to natural causes alone. The court has ruled for avulsion under artificial (man-caused) circumstances.

Naturally Occurring Avulsion

Distinguishment of the doctrine of avulsion from that of previously discussed doctrine lies in the “sudden and perceptible” vs. “gradual and imperceptible.” With the former, a boundary remains in place at its pre-avulsive location, whereas in the latter, a boundary will “run with” the gradual and imperceptible change in course.

In Nebraska v. Iowa, 143 U.S. 359 (1892), avulsion was defined as follows:

“…where a stream, which is a boundary, from any cause suddenly abandons its old and seeks a new bed, such change of channel, termed avulsion, works no change of boundary; the boundary remains as it was, in the center of the old channel, although no water may be flowing therein.”

From Bonelli Cattle Co. v. Arizona, 414 U.S. 313 (1973): “The rationale for the doctrine of avulsion is a need to mitigate the hardship that a shift in title caused by a sudden movement of the river would cause the abutting landowners were the accretion principle to be applied.” From Nebraska v. Iowa, supra: “(When in) deserting its original bed, the river forces a new channel in another direction, then the nation through whose territory the river thus breaks its way suffers injury by the loss of territory greater than the benefit of retaining the natural river boundary, and that boundary remains in the middle of the deserted river bed.”

There is a presumption at law that the gradual processes of accretion and erosion have been operative on riparian land, unless evidence clearly indicates otherwise (see, e.g., Wyckoff v. Mayfield, 130 Or. 687; 280 P 340, 342). In terms of a natural setting, there will invariably be evidence of unaltered improvements, undergrowth, vegetation or timber between the old (pre-avulsed) and new (post-avulsion) channels. This is a strong indicator for avulsion, which can further be substantiated through use of historical records and aerial or terrestrial photographs.

In McCafferty v. Young, 144 Mont. 385; 397 P 2d 96 (1964), the court observed, in part:

“While it is true, as counsel for defendant contends, that it is presumed that changes in river banks are due to accretion rather than avulsion (Wyckoff v. Mayfield, 130 Or. 687; 280 P 340), that rule does not apply where there is evidence of avulsive change. We think the evidence showing the age of trees lying between the former channel and the new channel precludes any conclusion that the lateral migration of the river was slow and imperceptible. The witness Hamre, who was the Helena National Forest Supervisor, testified that the trees lying on the land between the two channels were 70 to 80 years in age and still growing. Had the lateral migration of the river been gradual, the soil supporting the roots would have been washed out. Instead, this physical evidence demonstrates that those trees have remained strong since at least 1880 or 1890. The question is one of fact, and the trial judge found there had been an avulsive change. We feel there is ample and credible evidence to support that finding, and therefore, it will not be disturbed.” Ramsey v. Spratt, 79 Mont. 158; 255 P 5 (Emphasis added).

From the Manual of Surveying Instructions 1973, U.S. Department of Interior, Bureau of Land Management, at Section 7-73:

“An avulsive change cannot be assumed to have occurred without positive evidence. When no such showing can be made, it must be presumed that the changes have been caused by gradual erosion and accretion.”

Note that in McCafferty v. Young, supra, the court makes reference to questions of fact as key decisive elements in this case. Here we are reminded that what boundaries are is a question of law, but where boundaries are is a question of fact and evidence. While professional land surveyors tend to deal principally in the latter, we must also be more than casually knowledgeable of the laws governing water boundary determinations. For the client and public to be adequately served, the land surveyor must possess a measure of competence no less than this.

Where avulsion has in fact occurred, the pre-avulsive boundary is now fixed in place, having suspended for all time the processes of accretion and erosion over the extent of the former channel, for that event. Thus, the ambulatory characteristics formerly associated with the pre-avulsed boundary must now take on those of an upland boundary. It is important that a survey be performed to establish the pre-avulsive location of the former channel, which must be capable of re-establishment as a fixed boundary into the future.

Conditions necessary to satisfy the test of avulsion might include some or all of the following:

  1. Perceptible change.

  2. Sudden, unexpected or unusual event.

  3. Abandonment of old channel, or portion thereof.

  4. Pre-avulsive channel must be recognizable; it may be dry or may contain stagnant water, but must be identifiable as the former channel.

  5. The cut-off land mass (between old and new channels) must remain the same land mass and must not simply be a replacement land mass. It must not lose its pre-avulsive integrity and must still be identifiable.

  6. Under most circumstances, the length of a watercourse will be shortened as a result of an avulsion.

Artificially Caused Avulsion

As man has encroached further upon his environment, we have seen the extent of rivers and streams in their wildland state diminish. With growth and urbanization come the need for flood control, water conservancy, and the like, all related in some fashion to engineering works. These “man-caused” facilities take on many forms, often significantly and permanently altering the watercourse or water body environment. They control what previously was left to the forces of nature. The question that arises is: what impact does this have on the watercourse as a boundary, particularly when we see that water boundary doctrine in place has largely evolved on the concept of naturally occurring events?

Our system of common law, while historically slow to change, is dynamic in nature, being particularly sensitive to socio-economic trends. The point is raised here simply to bridge the definition of avulsion from naturally occurring causes to those induced by man, especially in terms of the tests for avulsion of being sudden and perceptible. Most engineering works for waterway projects are of considerable magnitude, requiring lengthy planning and construction. On the surface a contradiction would seem to exist between this fact and the condition of being sudden and perceptible for the court to rule on avulsion. How has the court rationalized this seeming disparity?

Over time the court has developed precedent in this area. Keep in mind, however, that each case is decided on its own merits and the evidence presented.

Avulsion is complete when water no longer flows in the abandoned bed. Standing water may remain for a period, but seems to have little bearing on ruling for avulsion. The precedent is this:

When the last section of bulkhead is poured in place, or the last load of material deposited, the watercourse is suddenly, at that moment, visibly and perceptibly altered, thus satisfying the key tests for avulsion.

In Arkansas v. Tennessee, 246 U.S. 158 (1918), the U.S. Supreme Court addressed, in part, the point in time at which an avulsion is completed:
“An avulsion has this effect whether it results in the drying up of the old channel or not. So long as that channel remains a running stream, the boundary marked by it is still subject to be changed by erosion and accretion; but when the water becomes stagnant, the effect of these processes is at an end; the boundary then becomes fixed in the middle of the channel as we have defined it, and the gradual filling up of the bed that ensues is not to be treated as an accretion to the shores, but as an ultimate effect of the avulsion.” (See also Arkansas v.Tennessee, 397 U.S 88 (1970)).

As in the case of accretion, the court has disallowed man’s influence to benefit himself, but has allowed the actions of one (a stranger to the land) to benefit another, so long as collusion is not present. In applying this tenet to a setting where accretion vs. avulsion is in debate, the court has ruled both ways on principles of equity, or to preclude a windfall on the part of one party over another.