|
This copy of report CNR-677CC08C9987 is published at https://app.curvenumber.com/p/6dJRYt_bZwus57VwoGpC7iXLks4ppxQs. Open that address to compare this printout with the published document. Reports are shown as they print, on a light page. CurveNumber
Runoff report
Report CNR-677CC08C9987 Inputs CNI-901197A54001 Computed Sep 26, 2026, 7:14 PM EDT Version CurveNumber 0.9.0 Status PROVISIONAL, read at Sep 26, 2026, 7:23 PM EDT
House addition and driveway, OlneyExample 4: Small lot, photograph check and quick check, client Example homeowner (fictional), jurisdiction Montgomery County, Maryland. Prepared by CurveNumber Examples, CurveNumber, as entered by the account holder. Located in Montgomery County, Maryland (boundary centroid at latitude, longitude 39.1472, -77.0772). Site boundary Site boundary 0.711 ac, centroid at latitude 39.147250, longitude -77.077200. North is up. Imagery: USGS The National Map, public domain, flown July 12, 2023. What this result rests on No waiver, no override of a derived or default value, no accepted estimate and no refusal still in force. The site has not changed since this run. 4 changes a person made to the proposed design are listed with their reason on summary page 3. Read this before the figures
This report was produced using CurveNumber.com, a computational aid for estimating NRCS curve numbers and direct runoff. It is not an engineering deliverable and it is not a certification of compliance with any regulation, ordinance, or design standard. The curve number method, as published in USDA NRCS Technical Release 55 and National Engineering Handbook Part 630 Chapter 10, is an empirical method with substantial uncertainty, and its applicability to any specific site is a professional judgment that this software has not made. Input data were retrieved from third party sources listed at https://curvenumber.com/data-sources and are provided by those sources without warranty. All results require review, verification, and adoption by a qualified professional before use in design, permitting, or construction. CurveNumber.com is not the engineer of record for any project. No Data Quality Grade is stated on this report. The grade is a defined measure of how good the inputs behind a result are, this version does not compute one, and an absent grade is not a good grade. What this report carries instead is the provenance of each input individually, in the dataset and assumption sections below. 728 cu ft Increase in runoff volume, existing against proposed, at the 4.87 in design storm: 0.017 ac-ft, 10.9 percent of the existing runoff volume. Runoff depth rises by 0.282 in (10.9 percent of the existing depth). This is an increase, not a reduction. The proposed condition produces more runoff than the existing condition at this storm. Existing depth 2.591 in Proposed depth 2.872 in Existing volume 6,690 cu ft (0.154 ac-ft) Proposed volume 7,419 cu ft (0.170 ac-ft) Design storm 4.87 in over 24.0 hr, NRCS Type II. Peak discharge by the NRCS standard dimensionless unit hydrograph (peak rate factor 484): existing 2.48 cfs; proposed 2.73 cfs. Summary, page 2: what was entered
Time of concentration used: existing 0.100 hr (6.0 min), raised from the 0.099 hr estimate to the minimum the method allows (CurveNumber's estimate); proposed 0.100 hr (6.0 min), raised from the 0.099 hr estimate to the minimum the method allows (CurveNumber's estimate for the proposed cover, on the same flow path as the existing estimate). No practice is on this run. Summary, page 3: what a person changed, and why
The full record follows this summary, unchanged: every section, with each figure's source. The full recordThe proposed condition totals 0.712 ac and the existing 0.711 ac; they differ by 4 sq ft (0.01 percent), which is within the 218 sq ft tolerance and is shown as information, not as a finding. Volumes and peaks are computed over the areas entered: 0.711 ac existing and 0.712 ac proposed. The drawn boundary is 0.711 ac. The areas were entered by hand; none of the derived areas is in use. Land cover folded into a larger class on its soil group (each below CurveNumber's minimum share for a class of its own; waiving the fold keeps each class at its own share): On soil group B, 0.112 ac of Developed, open space (15.8% of that group) was folded into Developed, low intensity; both are CN 69, so the site figure does not change. The state and county named by the datasets were compared with every geographic statement typed onto this file, and they agree. Provisional: assumptions not accepted 8 assumptions carried this answer and a person has accepted 0 of them. 8 carry no acceptance, so this report is provisional. Provisional means that the figures on this report were computed and are not withheld, and that the account holder has not yet submitted an acceptance of every assumption underneath them. The assumptions are listed further down with what each one is worth in the answer and what evidence would displace it. 6 of the 8 outstanding are an entry for which CurveNumber's list of default values states that no external source exists, or which it rates Low confidence. Those are accepted one at a time, cannot be accepted in bulk, and each one asks for a written reason. An acceptance on this report is an account holder submitting one, at the time printed with it. This product does not verify professional licensure: it holds no license number and no state of licensure, and it records no evidence that an entry was read as distinct from accepted. An acceptance is not a seal, a signature or a certification, and the absence of one is not a finding that anything is wrong. This count is as of , when this copy of the page was produced. It can change after that; the report identifier does not, because it names the computation. Two printouts with the same identifier are the same computation; if their counts differ, the later one is current. An entry is accepted on its own when it has no published source (a judgment, or a threshold CurveNumber chose) or when its confidence is low. Every other entry (a method constant, a published table, a local manual's figure or a research value rated moderate or better) may be accepted together with others in one act, and the record says so. Site mapThe boundary is the one saved on this site when it was computed, and the practice footprint is the one drawn for the practice, if any. The aerial photograph was taken from the public service when this report was made and is kept with the report, so it shows the ground as it was photographed, not as it is today. The map is for locating the site; the areas on this report are the figures, not measurements off this picture. Where this site is
This boundary was drawn or loaded on the map. It was NOT geocoded: this product does not turn an address into an area of interest, because the Census geocoder returns a point interpolated along a street centerline and runoff computed around such a point is runoff for a piece of road. The coordinates below are the polygon a person drew, and every dataset on this report was read over that polygon and over nothing else. The boundary every dataset on this report was read over
What is claimed about this ground, and on whose evidence
Annual NLCD Land Cover (CONUS) 2024: A national map is sampled at this boundary's own coordinates, so it agrees with the boundary whatever the boundary is and is not evidence about where the boundary is. The curve number method, chosen by the storm depth rule: Judged area by area on each area's own curve number: P of 4.87 in is above 2 Ia on every area that yields a material share of the runoff. The curve number governs, with a mandatory sensitivity band. Sensitivity: A five CN unit uncertainty either way on every area's curve number moves the site's area weighted runoff at this depth from 2.18 in to 2.90 in, against 2.59 in as computed, a factor of 1.3 from end to end. Method by area. Each area is judged on its own curve number and initial abstraction (TR-55 Chapter 2); impervious cover is computed at its TR-55 curve number. The site figure is the area weighted mean of the runoff depths below.
Storm
Peak dischargeNRCS unit hydrograph peak, lag = 0.6 Tc (NEH Part 630, Chapter 16). Unit hydrograph: the NRCS standard dimensionless unit hydrograph (peak rate factor 484), because the site is not on the Delmarva coastal plain and no local rule asks for another. The site with no practice on it, under the NRCS Type II storm distribution, which was chosen for this run; its fit to this region was not assessed. Time of concentration used: existing 0.100 hr (6.0 min), raised from the 0.099 hr estimate to the minimum the method allows (CurveNumber's estimate); proposed 0.100 hr (6.0 min), raised from the 0.099 hr estimate to the minimum the method allows (CurveNumber's estimate for the proposed cover, on the same flow path as the existing estimate). Time step 0.0125 hr. The peak is built from runoff computed for each area separately and added (distributed). A hand check at the composite curve number gives a lower peak where impervious and pervious areas are mixed, which TR-55 itself warns a composite understates. A practice's effect on the peak is shown with the practice, and the note on the distribution table below applies to its routed peaks too.
Assumption. Tc raised from 5.9 minutes to the TR-55 minimum of 6 minutes. Distribution: NRCS Type II, chosen for this run. TR-55 Figure B-2 assigns the NRCS type by county; this product does not hold that map, so the choice has not been checked against it. Not checked against the source The unit hydrograph shape table was transcribed and has not been checked against the source. The NRCS Type II rainfall distribution was read from the published NRCS table by script and has not yet had a second person's check. The four NRCS 24 hour distributions in this product are the published NRCS tables at a 0.1 hour step (as issued with WinTR-55 version 2.00), read from the published files by a script and compared by machine with a second copy of the same tables. They have NOT yet had a second person's check against the NRCS program files. This matters more here than a curve number would: a distribution ordinate changes the peak intensity, and the peak intensity drives the inflow hydrograph, the stage the practice reaches and therefore the volume split reported below. Where the jurisdiction publishes its own distribution, supply it. The datasets these areas were read fromThis section is the record for this report: the editions it read and the dates they were read. Every dataset CurveNumber reads, with the edition in use, is listed at https://curvenumber.com/data-sources. A dataset's retrieved date can be earlier than the day the areas were read. CurveNumber keeps a copy of each answer a publisher's service gives it, and a later request for the same map tile, soil survey area or storm point is answered from that copy rather than from the publisher again. The read date is when this site's areas were read; the retrieved date is when CurveNumber fetched that answer from the publisher, and the edition printed with it is the one that answer carries.
n is the boundary's area measured in land cover map squares: the area divided by the area of one square. Every square that touches the boundary is counted only for the part of it inside, so the derived areas add up to the boundary. n says how many independent readings of the ground the land cover rests on; a site of a few squares is coarse for this dataset whatever its areas add up to. What the datasets said about themselves. 1 land cover class was folded into a larger class on the same soil group, because each was smaller than one 30 m land cover pixel (0.222 ac) on that soil group, below what the map resolves: On soil group B, 0.112 ac of Developed, open space (15.8% of that group) was folded into Developed, low intensity; both are CN 69, so the site figure does not change. Waiving the fold keeps every class at its own share. This site is small for 30 m land cover: about 3 squares of 30 m (land cover pixels of 30 m), so one square read as the wrong class moves a cover share by about 31% of the site. Check the cover shares against the plan. The next step is the aerial photograph classification (NAIP), which reads the site at about 1 m and offers its classes (impervious, grass, trees, bare soil) as areas to confirm; or enter the surfaces by hand. Where each source can be read.
Refusals CurveNumber raised on this runCurveNumber raised no refusal on either condition or on a practice during this run. That is a statement about this run and not a warranty: a refusal is raised where a named condition is met, and the conditions CurveNumber tests are the ones in its list of refusals. Both aggregation paths
The two paths differ because runoff is a non-linear function of the curve number, so averaging the curve numbers first and averaging the runoff first do not give the same answer; the distributed path is reported because it applies the method to each surface as the method was fitted, and the composite path is printed beside it because it is what a reviewer recomputing by hand from a single area weighted curve number will obtain. What was estimated, from what, and what each estimate is worthCheck these 3 first: time of concentration, a factor of 19.62 on time of concentration; slope, a factor of 9.29 on slope across the area; temporal distribution, a factor of 8.37 on peak one hour rainfall fraction. The other 5 move their own figures by a factor of 4.00 or less and are reported rather than put in front of you. RANK 1 HEADLINE Time of concentration computed 0.099 hr Can change the time of concentration by up to 19.6 times. Rests on: TR-55 segmental: sheet flow over short grass prairie at n = 0.15 for 100 ft, then shallow concentrated flow. Slope 8.96 percent from USGS 3DEP, two year 24 hour depth 3.17 in from NOAA Atlas 14 Volume 2 Version 3, flow length 249 ft estimated as the longest chord of the drawn boundary, which is not a flow path. The surface is worth a factor of 19.62 on Tc and decides whether the peak discharge is drawn at the TR-55 minimum; the sheet flow cap 1.90; the estimated flow length 1.05. Displaced by: The flow path drawn on the grading plan, measured from the hydraulically most distant point to the design point, which is what the method actually asks for and what no dataset contains. A LiDAR flow accumulation routing over the graded surface displaces it too and is not implemented. Computed here from the figures named above. There is no external source because nothing external was read for it. RANK 2 HEADLINE Slope of the receiving area derived 8.96 percent Can change the slope across the area by up to 9.3 times. Rests on: USGS 3DEP (1 m DEM, acquired ) Slope Degrees raster function retrieved . Median of 2066 sampled pixels, in percent, which is the tangent of the angle and not the angle. The site spans P10 2.33 to P90 21.64 percent, a factor of 9.29, around a median of 8.96. A maximum slope condition is decided conservatively by the P90 and a minimum slope condition by the P10, and no single number is conservative for both, so the median is reported with both beside it. Displaced by: A graded surface from the plan, which is what the credit is actually checked against: the slope that matters is the one the receiving area will have after construction, and a DEM measures the one it has now. USGS 3DEP (1 m DEM, acquired ) Slope Degrees raster function retrieved RANK 3 HEADLINE Temporal distribution derived 0.277 fraction of total depth Can change the peak one hour rainfall fraction by up to 8.4 times. Rests on: NOAA Atlas 14 Volume 2 Version 2.1 temporal distribution file for Volume 2 (Ohio River Basin and Surrounding States), area general, 24h duration, fetched and hashed rather than transcribed. WHICH of its 45 curves is a judgment and is CurveNumber's default value “Which of the NOAA Atlas 14 quartile cases and probability of occurrence lines a design storm shape is taken from”. The selection spans a factor of 5.10 on the peak one hour fraction inside this file, and a factor of 8.37 against the NRCS Type II curve the product uses without it. The volume is untouched. Displaced by: The quartile case and probability line the governing jurisdiction names, which supersedes outright. Failing that, NOAA's own guidance on which case a design should use, which is published alongside the files and which this product has not read. A local gauge record of the storms that actually produced the design flows displaces the whole question. NOAA Atlas 14 (Volume 2 Version 2.1) first quartile case, 10 percent cumulative probability of occurrence line, 24h duration, Volume 2 (Ohio River Basin and Surrounding States), area general retrieved RANK 4 Irrigation water demand defaulted 253 cu ft/day Can change the irrigation demand in the governing month by up to 4.0 times. CurveNumber default value (“Plant factor in a landscape water budget”), which is an assumption and not a measurement. RANK 5 Antecedent dry period derived 12 days Can change the antecedent dry period by up to 3.0 times. NOAA NCEI GHCN-Daily (station record through ) station USC00180700, 90th percentile of interevent dry period length retrieved RANK 6 Green roof evapotranspiration rate defaulted 0.125 in/day Can change the green roof recovery time by up to 2.5 times. CurveNumber default value (“Ratio of an extensive green roof's evapotranspiration to the grass reference”), which is an assumption and not a measurement. RANK 7 Contributing flow path length defaulted 249 ft Can change the contributing flow path length by up to 1.5 times. CurveNumber default value (“Which geometric bound is offered for a jurisdictional credit measurement”), which is an assumption and not a measurement. RANK 8 Design storm depth derived 4.87 in Can change the design storm depth at a named return period by up to 1.2 times. NOAA Atlas 14 (Volume 2 Version 3) 24-hr duration, 10 year ARI, mean estimate, partial duration series, Ohio River Basin retrieved Design storm return period SUPPLIED 10 yr Supplied by CurveNumber Examples on , and CurveNumber recomputed from it. Nothing was overwritten: this is an argument CurveNumber takes, not a value it produced. It is an override all the same, because a person put it into this report, and it is in the overrides section below with every other one. Reason: Illustrative rule for this example only: the county is taken to ask a small lot permit to show the change at the 10-year, 24-hour storm. Check the governing standard for a real site. Receiving flow path length NOT OFFERED DELIBERATELY LEFT BLANK. A bound computable from the site boundary would make the condition easier to satisfy, so the field stays empty, the condition comes back unchecked and the credit is denied. Receiving flow path length: the length of pervious ground the flow crosses, which is a dimension of a practice nobody has drawn yet and not a dimension of the site. The narrowest crossing of the site boundary is a bounding dimension and using it here would credit a disconnection across ground that may hold no disconnection at all. The field is left empty, so a condition reading it comes back unchecked and the credit is not granted, which is the correct outcome. Filter strip width NOT OFFERED DELIBERATELY LEFT BLANK. A bound computable from the site boundary would make the condition easier to satisfy, so the field stays empty, the condition comes back unchecked and the credit is denied. Width: the width of a filter strip measured across the face it treats, which needs the flow direction and the strip, and CurveNumber has neither. A bounding dimension of the site is not a strip width. The field is left empty and a condition reading it denies the credit. Indoor water demand NO DATASET NOT IN ANY DATASET THIS PRODUCT READS. There is nothing to estimate from and a plausible figure would be an invention. An indoor water demand cannot be estimated from anything this product reads: it depends on occupancy, which no dataset holds. The irrigation part of a cistern's demand is estimated; the indoor part is not, because a guessed continuous demand would give a cistern its whole credit on a number nobody supplied. To credit it, enter a metered monthly demand schedule with a written commitment that the use continues. Overrides of the derived and defaulted valuesImpervious fraction accepted from an aerial photograph accepted adopted at 0.388696 dimensionless, and NOT used in this run's figures, because the site has not been derived since it was accepted; the areas keep the impervious fraction each was given, inside the band 0.3887 to 0.3887 that this photograph supports once the shadowed area is allowed to be anything A person examined a classification of a NAIP photograph acquired and adopted the figure read off it, unchanged. Nothing was overridden and nothing recomputed: the magnitude is the machine’s own, and it is a guess about an image rather than a measurement of this ground. The default value named below is the entry that states what such a figure is worth, not one this figure departs from. What the photograph is, what the classification did and what an independent published measurement of the same ground says are in the aerial photograph section of this report. Not recorded against any one area: this override is against the site. Accepted by CurveNumber Examples on Sep 26, 2026, 7:13 PM EDT, CurveNumber's default value “That an impervious fraction read off aerial imagery is a classification and not a measurement” Reason given: Checked against the owner's (fictional) 2024 house location survey: roof, driveway, walks and patio total about 0.27 acres, which matches the 0.277 acres read off the July 2023 photograph. Nothing has been built or paved since. Area, Impervious (aerial photograph) overridden set to 0.277 ac On the existing condition, area Impervious (aerial photograph). Changed by CurveNumber Examples on Sep 26, 2026, 7:13 PM EDT Reason given: From the aerial photograph accepted as revision 3, confirmed by the engineer. Checked against the same (fictional) house location survey: the planted part of the lot is mowed lawn with a few shade trees, in good condition, so Open space, good condition on group B. Area, Grass or trees (aerial photograph) overridden set to 0.435 ac On the existing condition, area Grass or trees (aerial photograph). Changed by CurveNumber Examples on Sep 26, 2026, 7:13 PM EDT Reason given: From the aerial photograph accepted as revision 3, confirmed by the engineer. Checked against the same (fictional) house location survey: the planted part of the lot is mowed lawn with a few shade trees, in good condition, so Open space, good condition on group B. Design storm return period supplied set to 10 yr A person supplied this input and CurveNumber recomputed the estimate from it. Nothing was overwritten: the figure it displaced is CurveNumber’s own, and it is printed with the estimates above. Not recorded against any one area: this override is against the site. Supplied by CurveNumber Examples on Sep 26, 2026, 7:14 PM EDT Reason given: Illustrative rule for this example only: the county is taken to ask a small lot permit to show the change at the 10-year, 24-hour storm. Check the governing standard for a real site. Records that are not overrides. The product wrote these when a condition was copied, or they are the design of a practice. Proposed condition seeded from existing by CurveNumber Examples on Sep 26, 2026, 7:14 PM EDT: 0.711 ac copied. Nothing copied has been edited since, so the surfaces are the same in both conditions and any change is the practice. The aerial photograph, and what was read off itA person accepted a classification of an aerial photograph. accepted The photograph. NAIP, acquired , ground resolution 0.30 m. Published data, cited as USGS NAIP Imagery (The National Map) (NAIP, acquired , 0.3 m native, 0.30 m delivered) 1 scene, bands red/green/blue/nir retrieved . The citation belongs to the image and does not extend to anything concluded from it. The classification. An unsupervised clustering of the pixels of that photograph, applying rules nobody published. It is a guess about an image and not a measurement of the ground, and it is recorded against default value “That an impervious fraction read off aerial imagery is a classification and not a measurement”, which states what it is worth. 38.9% of the boundary is hard surface, and this photograph supports no other figure once the shadowed area is allowed to be anything Checked against an independent published measurement. The photograph gives 38.9% hard surface; the published map (Annual NLCD Land Cover (CONUS) 2024) gives 29.5%. They differ by 9.4 percentage points, within the 10.0 allowed, so they agree. The published figure is outside the range this photograph supports. The published figure was read over 4 whole map squares for this check. Annual NLCD Land Cover (CONUS) itself reports this result as resting on too few pixels to be a site measurement: 4 against the minimum of 30 CurveNumber sets. It is used here as an independent reading and not as ground truth. Different editions of the ground: imagery flown against Annual NLCD Land Cover (CONUS) for 2024. A real change on the site between the two dates appears here as a disagreement. The most recent NAIP flight over this site was , about 3 years ago. NAIP is flown on a cycle of a few years, so anything graded, paved or built since then is not in this photograph, and a classification of it describes the site as it was. This is a classification of a photograph flown . It describes the site as it was on that day. Anything graded, paved, built or cleared since is not in the image and is not in these numbers, and on a site being designed that gap is the point rather than a detail. Every class here is a guess about a photograph rather than a measurement of the ground. The impervious fraction is DEFAULTED against CurveNumber's default value “That an impervious fraction read off aerial imagery is a classification and not a measurement”, not derived, and it must read differently from the NLCD fractional impervious figure everywhere the two appear together. No cluster fell below the shadow brightness break, so no area was set aside as undetermined. On a site with buildings or mature trees that is more likely to mean the clustering merged the shade into a surface class than that the site has no shade, and an unhandled shadow inflates impervious area. NAIP is flown leaf on. A driveway or a parking bay under a mature canopy is not in this photograph, so the impervious fraction is understated by exactly the area the canopy covers and no work on these four bands recovers it. Bounding it rather than correcting it: vegetation is 61.1% of the site and 33.5% of the site is vegetation within 3 m of classified impervious surface, which is where an overhanging canopy would be. Displacing evidence is a first return minus bare earth height model, a published canopy layer, or a leaf off winter flight. The vegetated area is not split into lawn and tree canopy, because a photograph from overhead has no height in it. TR-55 puts woods in fair condition below open space in fair condition by 13 curve number units on group A, 9 curve number units on group B, 6 curve number units on group C, 5 curve number units on group D; over the 61% of this site read as vegetation that is 3.06 to 7.95 units of composite, depending on the soil group. The CurveNumber's default value “Which classes an imagery classification resolves the site into” carries the arithmetic. The pixel values are raw brightness with no calibration, so a threshold tuned on another photograph does not transfer. The line between vegetation and hard surface was set from this photograph's own pixels, at a vegetation index of +0.24; the usual fixed value of 0.2 would have classified a different site. The bare soil and water lines are FIXED and do not adapt; both say so on their default values. The reported classes are one of several groupings the clustering can settle on, and which one it finds depends on where it starts, so the same pixels were also grouped from two other fixed starts: the other end of every tied quantile gives 38.9% impervious; furthest point gives 39.5% impervious. They disagree with it by 0.0068 of impervious area, which is 0.20 curve number units on the 98 against 69 pairing and is beyond the 0.0035 CurveNumber treats as settled. Nothing in the photograph chooses between them, so the figure above is one of several answers these pixels support and not the answer they give. 1 of the 2 alternatives fit these pixels more closely than the reported grouping does; the reported one is kept anyway, because substituting it would move a figure on reports already issued with no new evidence about the ground. Pixel centers inside the boundary cover 2879.4 sq m against the boundary's own geodesic area of 2879.1 sq m, a closure of +0.010 percent. Shares above are of the sampled area. From the imagery: The most recent NAIP flight over this site was , about 3 years ago. NAIP is flown on a cycle of a few years, so anything graded, paved or built since then is not in this photograph, and a classification of it describes the site as it was. Not applied yet: the site has not been derived since this was accepted, so no figure on this report uses it. CurveNumber Examples accepted a classification of a NAIP photograph acquired on , on . The photograph is published data and is cited. The figure accepted from it is not a measurement of this site: it is the output of clustering the pixels of that photograph and applying rules nobody published, and it is recorded as a default against CurveNumber's default value “That an impervious fraction read off aerial imagery is a classification and not a measurement” rather than derived from the imagery it was computed from. Accepted by CurveNumber Examples on (classification 1) Reason given: Checked against the owner's (fictional) 2024 house location survey: roof, driveway, walks and patio total about 0.27 acres, which matches the 0.277 acres read off the July 2023 photograph. Nothing has been built or paved since. This acceptance is also in the overrides section, because it is one: a person put a figure into this record under their own name and nothing CurveNumber derived was displaced by it. It is marked ACCEPTED there and not OVERRIDDEN or SUPPLIED, because the magnitude adopted is the machine's own and no value moved. Classified by CurveNumber Examples on (classification 1) The curve number fitted to an observed record, and what it is aboutNo curve number was fitted to an observed record for this site. Nothing on this report rests on a gauged rainfall and discharge record. This is the ordinary case and it is stated rather than left to be inferred from the absence of a section. It is also the expected case: a curve number fitted at a stream gauge describes the whole watershed above that gauge, and this product will not print such a figure against a site that is not inside that watershed and a large enough fraction of it, which most sites are not. Refusals a supplied input resolvedNo refusal was resolved by a supplied input. Nothing on this report rests on a figure a user handed CurveNumber to answer a refusal with. Where a refusal fired, it fired and it is printed as a refusal. Refusals a person waivedNo refusal was waived. Every figure on this report is one CurveNumber produced. Nobody waived a refusal, and no figure here rests on a signature in place of a computation. The assumptions that carry this answerHydrologic condition defaulted Fair (a judgment with no published source; low confidence) PENDINGno acceptance recorded Worth in the answer: 5 to 12 CN units. One of the two largest silent levers in the method. Displaced by: A site visit, current aerial imagery, or a measured ground cover fraction against the NEH Chapter 9 mapping. Curve number suppression band, lower defaulted 2 multiples of Ia (a threshold CurveNumber chose; low confidence) PENDINGno acceptance recorded Worth in the answer: A five CN unit uncertainty is worth a factor of 18.1 at 1.5 Ia, 4.4 at 2 Ia, 2.3 at 3 Ia and 1.4 at 9 Ia. Displaced by: Expert panel calibration. Curve number caution band, upper defaulted 5 multiples of Ia (a threshold CurveNumber chose; low confidence) PENDINGno acceptance recorded Worth in the answer: Above this the method is in the regime it was fitted for. Displaced by: Expert panel calibration. Impervious connectivity defaulted 0 (a judgment with no published source; low confidence) PENDINGno acceptance recorded Worth in the answer: Bounded at 10.2 CN units. Figure 2-4 adjusts the composite as CN = CNp + (Pimp/100)(98 - CNp)(1 - R/2), so the whole span from fully connected to fully unconnected is 0.5 (Pimp/100)(98 - CNp), largest at the 30 percent total impervious ceiling the figure is valid to and at the lowest pervious curve number TR-55 publishes, which is 30: 0.15 x 68 = 10.2 units. At a pervious CN of 61 the same bound is 5.55 units, and at 80 it is 2.7. Displaced by: A delineated drainage path showing impervious area discharging to pervious ground. Automated detection from LiDAR routing is an open gap. That an impervious fraction read off aerial imagery is a classification and not a measurement defaulted (a judgment with no published source; low confidence) PENDINGno acceptance recorded A named person adopted a figure produced under this entry, which is recorded in the overrides section with their name, the date and the reason they gave. Nothing was displaced: the figure adopted is the machine's own, and this entry is what states what such a figure is worth. Adopting the figure is not the same act as accepting the assumption, so this entry is still counted above and is still waiting for somebody to put their name to it. Adopted by CurveNumber Examples on ; see the overrides section. Worth in the answer: The whole of the difference between a number and a guess about a photograph. On the recorded Fairfax site the classification returns an impervious fraction of 0.6693 where the NLCD fractional impervious map, which is a published measurement, returns 0.6553: 1.40 points apart, 0.41 curve number units on group B with the pervious remainder as open space in fair condition, and 1.81 percent of the runoff depth at P = 3 in. That agreement is the best case. The band this same photograph supports once the shadowed area is allowed to be anything runs 0.5746 to 0.7161, which is CN 85.7 to 89.77 and 1.6372 in against 1.9642 in at P = 3 in, 19.97 percent more on the smaller, and 0.2949 in against 0.4473 in at P = 1.2 in, 51.70 percent more. The classified figure is therefore one choice inside a band four curve number units wide, and it is DEFAULTED against this entry rather than DERIVED from the imagery it was computed from. Displaced by: A planimetered impervious area off a site survey or a stamped as-built, which supersedes outright. Failing that, the engineer relabelling areas against the photograph, which is recorded as an override with who and why. The published 30 m fractional impervious map does not displace it and is not displaced by it: the two are independent readings and the product reports both with the difference between them. Cover type, NLCD 22 developed low intensity defaulted Open space, Fair, on the pervious part (a judgment with no published source; moderate confidence) PENDINGno acceptance recorded Worth in the answer: 9.2 CN units on group B against the double counted alternative. The residential 1/4 acre row is exactly 0.38 x 98 + 0.62 x 61 = 75.06, so it already holds its impervious share; using it as the pervious cover and adding the measured 34.6 percent impervious as its own area gives 83.0 where the correct composite is 73.8, and 1.444 in against 0.898 in, 61 percent high. Both sides are on TR-55's own good condition assumption for urban pervious area, so the figure isolates the double counting from the condition choice; under the defaulted fair condition it is 3.9 CN units and 21.5 percent. Preferring the class midpoint of 38 percent to the measured 34.6 is worth a further 1.3 CN units and 7.3 percent. Displaced by: A high resolution land cover product that resolves roofs and driveways, or a site plan giving the actual lot coverage. 2 further default values applied to this computation and are left out here, because a reader shown thirty defaults checks none of them. Those above are ordered by the strength of the evidence behind them, weakest first. The complete list, with what each is worth and what would replace it, is on the Reference page of CurveNumber. Across all 8 of them, 0 have been acknowledged and 8 have not; the machine readable companion to this report lists every one with its state. Assumptions with no external sourceThese values carry the answer and nothing external supports them: some are judgments for which no source exists at all (a judgment with no published source), and others are product thresholds that were reasoned rather than measured (a threshold CurveNumber chose). They are assumptions, recorded as assumptions. The text below is CurveNumber's own statement of each value, verbatim. Hydrologic condition defaulted Fair (a judgment with no published source) NO EXTERNAL SOURCE EXISTS. Fair is the mid-range assignment and avoids systematic bias. Recorded as a judgment, not a citation, because a user told this is 'the published default' will not think to change it, and a user told it is a mid-range assumption worth 5 to 12 CN units will. Worth in the answer: 5 to 12 CN units. One of the two largest silent levers in the method. Displaced by: A site visit, current aerial imagery, or a measured ground cover fraction against the NEH Chapter 9 mapping. Curve number suppression band, lower defaulted 2 multiples of Ia (a threshold CurveNumber chose) REASONED, NOT MEASURED. Proposed default awaiting calibration. Worth in the answer: A five CN unit uncertainty is worth a factor of 18.1 at 1.5 Ia, 4.4 at 2 Ia, 2.3 at 3 Ia and 1.4 at 9 Ia. Displaced by: Expert panel calibration. Curve number caution band, upper defaulted 5 multiples of Ia (a threshold CurveNumber chose) REASONED, NOT MEASURED. Worth in the answer: Above this the method is in the regime it was fitted for. Displaced by: Expert panel calibration. Impervious connectivity defaulted 0 (a judgment with no published source) NO EXTERNAL SOURCE EXISTS. The conservative assumption. Worth in the answer: Bounded at 10.2 CN units. Figure 2-4 adjusts the composite as CN = CNp + (Pimp/100)(98 - CNp)(1 - R/2), so the whole span from fully connected to fully unconnected is 0.5 (Pimp/100)(98 - CNp), largest at the 30 percent total impervious ceiling the figure is valid to and at the lowest pervious curve number TR-55 publishes, which is 30: 0.15 x 68 = 10.2 units. At a pervious CN of 61 the same bound is 5.55 units, and at 80 it is 2.7. Displaced by: A delineated drainage path showing impervious area discharging to pervious ground. Automated detection from LiDAR routing is an open gap. That an impervious fraction read off aerial imagery is a classification and not a measurement defaulted (a judgment with no published source) NO EXTERNAL SOURCE EXISTS. Nobody publishes the classification CurveNumber performs, and the imagery it reads is a photograph rather than a map: it records how much light of four wavelengths left the ground on one day, not what the ground is. The provenance rule follows from that and is the point of the entry. An impervious fraction of 0.43 read out of the NLCD fractional impervious map is DERIVED and cites MRLC; an impervious fraction of 0.43 produced here is DEFAULTED and cites this entry; and the second must read differently on the face of a report, because a 0.6 m number carries an authority a 30 m number does not and that authority is exactly what must not be extended to a guess by accident. One more thing belongs on the face of this entry: the classification has been checked against an independent measurement on exactly ONE site, the recorded Fairfax square, which is a suburban office park flown in October. Every figure quoted anywhere in the classify.* entries is from that site. A method validated at n = 1 is not validated; it is one agreement, and it is reported as one. Worth in the answer: The whole of the difference between a number and a guess about a photograph. On the recorded Fairfax site the classification returns an impervious fraction of 0.6693 where the NLCD fractional impervious map, which is a published measurement, returns 0.6553: 1.40 points apart, 0.41 curve number units on group B with the pervious remainder as open space in fair condition, and 1.81 percent of the runoff depth at P = 3 in. That agreement is the best case. The band this same photograph supports once the shadowed area is allowed to be anything runs 0.5746 to 0.7161, which is CN 85.7 to 89.77 and 1.6372 in against 1.9642 in at P = 3 in, 19.97 percent more on the smaller, and 0.2949 in against 0.4473 in at P = 1.2 in, 51.70 percent more. The classified figure is therefore one choice inside a band four curve number units wide, and it is DEFAULTED against this entry rather than DERIVED from the imagery it was computed from. Displaced by: A planimetered impervious area off a site survey or a stamped as-built, which supersedes outright. Failing that, the engineer relabelling areas against the photograph, which is recorded as an override with who and why. The published 30 m fractional impervious map does not displace it and is not displaced by it: the two are independent readings and the product reports both with the difference between them. Cover type, NLCD 22 developed low intensity defaulted Open space, Fair, on the pervious part (a judgment with no published source) NO EXTERNAL SOURCE EXISTS. Maps to open space for the PERVIOUS remainder only, with the measured impervious fraction carried as its own area. The composite residential rows are not used at all, because they are a class midpoint for imperviousness with a cover assumption stapled to it. Worth in the answer: 9.2 CN units on group B against the double counted alternative. The residential 1/4 acre row is exactly 0.38 x 98 + 0.62 x 61 = 75.06, so it already holds its impervious share; using it as the pervious cover and adding the measured 34.6 percent impervious as its own area gives 83.0 where the correct composite is 73.8, and 1.444 in against 0.898 in, 61 percent high. Both sides are on TR-55's own good condition assumption for urban pervious area, so the figure isolates the double counting from the condition choice; under the defaulted fair condition it is 3.9 CN units and 21.5 percent. Preferring the class midpoint of 38 percent to the measured 34.6 is worth a further 1.3 CN units and 7.3 percent. Displaced by: A high resolution land cover product that resolves roofs and driveways, or a site plan giving the actual lot coverage. Inputs, for recomputationExisting condition
Proposed condition
How to read the provenance marksentereddouble borderA person typed this area: its area, and its cover and soil group where named. Who and when are printed on the row. Where the curve number is the TR-55 table value for that cover and soil group, the row also says the curve number was looked up; the area and the choice of cover are still the person's.editeddouble border, underlinedA person changed the area, cover, soil group or curve number of this area after it was derived, entered or copied. The change, who, when and the reason are printed on the row and in the overrides section.copiedheavy solid borderCopied from the other condition. The row says whether the area it was copied from was read from the datasets or typed by a person.derivedsolid borderRetrieved from a published source. The citation and its edition are printed with the value, because TR-55 1986 and TR-55 1975 disagree and a citation without an edition cannot be checked.defaulteddashed borderAn assumption, not a measurement. CurveNumber's list of default values entry is named, and what the assumption is worth in the answer is stated with it. Where no external source exists for the assumption, this report says so in those words.overriddendouble border, hatched, boldA person changed the value. Who, when, the prior value and the reason given are recorded in the overrides section and cannot be edited afterwards, because the parameter record is append only.supplieddouble border, boldA person gave CurveNumber an argument and it recomputed from it: a measured flow path length in place of the geometric bound, the flow surface the ground actually is, the month the design storm is taken in. Nothing the product had stored was overwritten, which is why this mark is not the override mark; what the argument displaced is CurveNumber's own figure, and the estimates section prints it. Who, when and the reason given are in the overrides section and the record is append only.accepteddouble border, dotted inner rule, boldA person examined a figure a machine produced and adopted it unchanged. It carries a person's mark because a person put it into this record under their own name, and it is not the supplied mark because nothing was handed to anything and nothing recomputed: the magnitude is the machine's own. Two acts carry it. The acceptance of a classification of an aerial photograph, which is a guess about an image and not a measurement of the ground, and which the imagery section explains; and the acceptance of a curve number fitted to a gauged rainfall and discharge record, which is a property of the whole watershed above that gauge and not a measurement of this site, and which the calibration section explains. The mark is also on an area of ground the soil survey does not rate, computed on the most likely soil group a person accepted (an estimate, not a survey rating), and on an impervious area sized by an accepted photograph. Who, when and the reason given are in the overrides section and the record is append only.fitteddashed border, dotted inner ruleCurveNumber fitted this magnitude to an observed record. It carries no person's mark, because no person put it here, and it is not the computed mark, because a computed value is a function of the inputs printed beside it and a fitted one is a function of the inputs AND of conventions: change the baseflow separation and not one input moves while the magnitude moves by curve number units. It is never printed without its interval, and a fitted figure becomes an input to a design only when a named person accepts it, which is a separate act carrying the accepted mark above. The marks are separated by the word, by the border style and by the fill, in that order of reliability, so this report can be read correctly from a black and white photocopy and from a browser printing with background graphics switched off. Reproducibility
Decision support. Every figure in this report requires independent verification, and a licensed professional is responsible for the design. Values marked DERIVED carry a citation with an edition. Values marked DEFAULTED carry an assumption rather than a measurement, and for some of them no external source exists, which CurveNumber's list of default values states on its face and which this report repeats rather than hides. Values marked OVERRIDDEN were changed by a named person, and the prior value and the reason given are printed in the overrides section. | |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||