{"id":13145,"date":"2021-07-02T12:15:07","date_gmt":"2021-07-02T11:15:07","guid":{"rendered":"https:\/\/www.vtei.cz\/?p=13145"},"modified":"2024-07-17T11:52:11","modified_gmt":"2024-07-17T10:52:11","slug":"quantification-of-rill-erosion-using-contactless-methods","status":"publish","type":"post","link":"https:\/\/www.vtei.cz\/en\/2021\/07\/quantification-of-rill-erosion-using-contactless-methods\/","title":{"rendered":"Quantification of rill erosion using contactless methods"},"content":{"rendered":"

<\/i> This article is available in Czech only. For translation or more information on this topic, please contact author.<\/h4>\n

 <\/p>\n

Souhrn<\/h2>\n

S\u00a0p\u0159ekotn\u00fdm\u00a0v\u00fdvojem v\u00a0oblasti geoinforma\u010dn\u00edch a\u00a0komunika\u010dn\u00edch technologi\u00ed se objevuj\u00ed nov\u00e9 mo\u017enosti mapov\u00e1n\u00ed a\u00a0kvantifikace erozn\u00edch proces\u016f zp\u016fsoben\u00fdch p\u0159\u00edvalov\u00fdmi sr\u00e1\u017ekami. Pro z\u00edsk\u00e1n\u00ed relevantn\u00edch v\u00fdsledk\u016f je d\u016fle\u017eit\u00e1 kvalita, hustota a\u00a0rozsah vstupn\u00edch dat. Rychl\u00fd a\u00a0pom\u011brn\u011b levn\u00fd sb\u011br p\u0159esn\u00fdch geodat umo\u017e\u0148uj\u00ed bezpilotn\u00ed leteck\u00e9 prost\u0159edky (UAV\u00a0\u2013 z\u00a0anglick\u00e9ho Unmanned Aerial Vehicle). S\u00a0vyu\u017eit\u00edm metod digit\u00e1ln\u00ed fotogrammetrie je mo\u017en\u00e9 z\u00a0po\u0159\u00edzen\u00fdch leteck\u00fdch sn\u00edmk\u016f rekonstruovat morfologii ter\u00e9nu odpov\u00eddaj\u00edc\u00ed \u010dasu m\u011b\u0159en\u00ed. Vybran\u00e1 lokalita zasa\u017een\u00e1 v\u00fdraznou r\u00fdhovou eroz\u00ed byla v\u00a0monitorovac\u00ed kampani nasn\u00edm\u00e1na bezpilotn\u00edm leteck\u00fdm syst\u00e9mem slo\u017een\u00fdm z\u00a0hexakopt\u00e9ry TAROT 690 a\u00a0fotoapar\u00e1tu Sony Alpha 6000. Pro n\u00e1sledn\u00e9 polohopisn\u00e9 a\u00a0v\u00fd\u0161kopisn\u00e9 ur\u010den\u00ed digit\u00e1ln\u00edho modelu povrchu (DMP) byly p\u0159\u00edstrojem GNSS rover Trimble R2 zam\u011b\u0159eny pozemn\u00ed kontroln\u00ed body (GCP\u00a0\u2013 z\u00a0anglick\u00e9ho Ground Control Points). Celkov\u00fd objem erodovan\u00e9 p\u016fdy byl stanoven na 4 274 m3<\/sup>. Odchylky kontroln\u00edch bod\u016f se pohybovaly v\u00a0prvn\u00edch jednotk\u00e1ch centimetr\u016f. Studie popisuje p\u0159\u00edpravu a\u00a0pr\u00e1ci v\u00a0ter\u00e9nu, bezkontaktn\u00ed mapov\u00e1n\u00ed z\u00e1jmov\u00e9ho \u00fazem\u00ed, zpracov\u00e1n\u00ed z\u00edskan\u00fdch dat a\u00a0v\u00fdsledn\u00fd v\u00fdpo\u010det a\u00a0kvantifikaci objemu odnesen\u00e9ho materi\u00e1lu ze zem\u011bd\u011blsk\u00e9ho pozemku po\u0161kozen\u00e9ho eroz\u00ed.<\/p>\n

\u00daVOD<\/h2>\n

Orn\u00e1 p\u016fda je v\u00a0posledn\u00edch letech \u010dast\u011bji vystavov\u00e1na pov\u011btrnostn\u00edm extr\u00e9m\u016fm, jako jsou dlouh\u00e1 such\u00e1 obdob\u00ed a\u00a0p\u0159\u00edvalov\u00e9 sr\u00e1\u017eky. A\u00a0pr\u00e1v\u011b p\u0159\u00edvalov\u00e9 sr\u00e1\u017eky jsou velmi \u010dasto spou\u0161t\u011b\u010dem plo\u0161n\u00e9 a\u00a0r\u00fdhov\u00e9 vodn\u00ed eroze [1]. M\u00edru rizika vzniku eroze d\u00e1le umoc\u0148uje vysok\u00fd sklon svahu, faktor nep\u0159eru\u0161en\u00e9 d\u00e9lky svahu, chyb\u011bj\u00edc\u00ed vegeta\u010dn\u00ed kryt nebo nevhodn\u011b zvolen\u00fd osevn\u00ed postup [2]. D\u016fsledkem je odnos nej\u00farodn\u011bj\u0161\u00edch p\u016fdn\u00edch horizont\u016f ze zem\u011bd\u011blsk\u00e9ho pozemku, sni\u017eov\u00e1n\u00ed obsahu organick\u00e9 hmoty, miner\u00e1ln\u00edch \u017eivin a\u00a0humusu v\u00a0p\u016fd\u011b, po\u0161kozen\u00ed p\u011bstovan\u00fdch rostlin nebo smyv osiva a\u00a0sadby [3, 4]. Sekund\u00e1rn\u00edm d\u016fsledkem erozn\u00edho smyvu je transport splavenin a\u00a0zan\u00e1\u0161en\u00ed vodn\u00edch tok\u016f a\u00a0vodn\u00edch n\u00e1dr\u017e\u00ed erodovan\u00fdm materi\u00e1lem [5]. Erozn\u00ed procesy tak maj\u00ed negativn\u00ed vliv na produk\u010dn\u00ed schopnosti orn\u00e9 p\u016fdy a\u00a0p\u0159in\u00e1\u0161ej\u00ed celou \u0159adu probl\u00e9m\u016f tak\u00e9 vodn\u00edmu hospod\u00e1\u0159stv\u00ed. Vlivem eroze na zem\u011bd\u011blskou p\u016fdu a\u00a0jej\u00ed ochranou se ve sv\u00e9 pr\u00e1ci podrobn\u011b zab\u00fdv\u00e1 nap\u0159\u00edklad Jane\u010dek a\u00a0kol. [6].<\/p>\n

Pro stanoven\u00ed erozn\u00ed ohro\u017eenosti p\u016fdy byly v\u00a0minulosti vyvinuty modely zalo\u017een\u00e9 na empirick\u00fdch metod\u00e1ch. Mezi nejpou\u017e\u00edvan\u011bj\u0161\u00ed pat\u0159\u00ed univerz\u00e1ln\u00ed rovnice ztr\u00e1ty p\u016fdy (USLE) [7] nebo distributivn\u00ed model erozn\u00edch a\u00a0transportn\u00edch proces\u016f WaTEM\/SEDEM [8]. Existuj\u00ed tak\u00e9 modely zalo\u017een\u00e9 na fyzik\u00e1ln\u00edch metod\u00e1ch, kter\u00e9 charakterizuj\u00ed pr\u016fb\u011bh erozn\u00edch a\u00a0transportn\u00edch proces\u016f na z\u00e1klad\u011b fyzik\u00e1ln\u00edch vztah\u016f a\u00a0z\u00e1vislost\u00ed. Tyto modely jsou velmi n\u00e1ro\u010dn\u00e9 na vstupn\u00ed data a\u00a0v\u00fdpo\u010detn\u00ed techniku. Rozd\u011blen\u00ed a\u00a0p\u0159ehled metod a\u00a0model\u016f erozn\u00ed ohro\u017eenosti popisuje nap\u0159\u00edklad Merritt a\u00a0kol. [9]. Na z\u00e1klad\u011b zm\u00edn\u011bn\u00fdch model\u016f vymezil Novotn\u00fd a\u00a0kol. pro \u00fazem\u00ed \u010cesk\u00e9 republiky t\u00e9m\u011b\u0159 50 % plochy orn\u00e9 p\u016fdy jako ohro\u017een\u00e9 vodn\u00ed eroz\u00ed [2]. Z\u00a0pohledu cel\u00e9 kontinent\u00e1ln\u00ed Evropy pak Panagos a\u00a0kol. ur\u010dil pr\u016fm\u011brnou ztr\u00e1tu orn\u00e9 p\u016fdy na 2,46 t\/ha\/rok [10]. Existuj\u00edc\u00ed modely tohoto typu dob\u0159e slou\u017e\u00ed pro odhad pr\u016fm\u011brn\u00e9 ztr\u00e1ty p\u016fdy a\u00a0dok\u00e1\u017eou lokalizovat zem\u011bd\u011blskou p\u016fdu ohro\u017eenou eroz\u00ed.<\/p>\n

V\u00a0\u010dl\u00e1nku popisovan\u00e1 p\u0159\u00edpadov\u00e1 studie se naproti tomu zab\u00fdv\u00e1 monitoringem ji\u017e prob\u011bhl\u00e9 erozn\u00ed ud\u00e1losti a\u00a0kvantifikac\u00ed objemov\u00fdch ztr\u00e1t orn\u00e9 p\u016fdy na konkr\u00e9tn\u00edm zem\u011bd\u011blsk\u00e9m pozemku.<\/p>\n\"\"<\/a>\n

Obr.\u00a01. Lokalizace eroz\u00ed po\u0161kozen\u00e9ho svahu
\nFig. 1. Localisation of the slope damaged by erosion<\/h6>\n

Pro mapov\u00e1n\u00ed r\u00fdhov\u00e9 eroze se v minulosti vyu\u017e\u00edvaly kontaktn\u00ed metody [11]. V posledn\u00edch letech pak do pop\u0159ed\u00ed vstupuj\u00ed metody bezkontaktn\u00ed, mezi kter\u00e9 pat\u0159\u00ed nap\u0159\u00edklad dru\u017eicov\u00e9 sn\u00edmkov\u00e1n\u00ed, leteck\u00e9 sn\u00edmkov\u00e1n\u00ed a sn\u00edmkov\u00e1n\u00ed pomoc\u00ed UAV. Dru\u017eicov\u00e9 a leteck\u00e9 sn\u00edmky jsou vhodn\u00e9 sp\u00ed\u0161e pro lokalizaci erozn\u00ed ud\u00e1losti, nedosahuj\u00ed toti\u017e rozli\u0161en\u00ed pot\u0159ebn\u00e9ho pro detailn\u00ed vyhodnocen\u00ed a kvantifikaci eroze [12]. Naproti tomu syst\u00e9my UAV umo\u017e\u0148uj\u00ed mapovat zemsk\u00fd povrch z velmi n\u00edzk\u00fdch letov\u00fdch hladin, a z\u00edskat tak spojitou informaci o aktu\u00e1ln\u00edm stavu monitorovan\u00e9ho \u00fazem\u00ed ve velmi vysok\u00e9m rozli\u0161en\u00ed. V ter\u00e9nu z\u00edskan\u00e1 data lze pomoc\u00ed fotogrammetrick\u00fdch metod p\u0159ev\u00e9st do podoby podrobn\u00fdch ortofotomap nebo detailn\u00edch digit\u00e1ln\u00edch model\u016f povrchu, nad nimi\u017e lze n\u00e1sledn\u011b po\u010d\u00edtat mapov\u00e1 algebra. Hojn\u011b pou\u017e\u00edvanou fotogrammetrickou metodou, vy-\u2028u\u017eitou i\u00a0pro prezentovanou p\u0159\u00edpadovou studii, je Structure from Motion (SfM), kter\u00e1 na z\u00e1klad\u011b dvojrozm\u011brn\u00fdch obrazov\u00fdch sekvenc\u00ed dok\u00e1\u017ee v\u00a0digit\u00e1ln\u00edm prost\u0159ed\u00ed rekonstruovat trojrozm\u011brn\u00e9 prostorov\u00e9 struktury [13]. P\u0159esnost\u00ed metody SfM v\u00a0souvislosti s\u00a0v\u00fd\u0161kou, z\u00a0n\u00ed\u017e byly sn\u00edmky po\u0159\u00edzeny, se zab\u00fdv\u00e1 Smith a\u00a0kol. Dokl\u00e1d\u00e1 line\u00e1rn\u00ed z\u00e1vislost mezi vzr\u016fstaj\u00edc\u00ed letovou hladinou a\u00a0chybovost\u00ed kone\u010dn\u00e9ho v\u00fdsledku [14]. Z\u00e1znamem a\u00a0m\u011b\u0159en\u00edm eroze s\u00a0vyu\u017eit\u00edm UAV se ve sv\u00fdch prac\u00edch zab\u00fdvali nap\u0159\u00edklad Carollo a\u00a0kol. nebo Peter a\u00a0kol. [15, 16]. Postupy p\u0159\u00edm\u00e9ho monitoringu a\u00a0kvantifikace erozn\u00edch jev\u016f pomoc\u00ed digit\u00e1ln\u00ed fotogrammetrie p\u0159edstavuje tak\u00e9 \u017d\u00ed\u017eala a\u00a0kol. [12].<\/p>\n

\u010cl\u00e1nek p\u0159in\u00e1\u0161\u00ed v\u00fdsledky p\u0159\u00edpadov\u00e9 studie usiluj\u00edc\u00ed o\u00a0co mo\u017en\u00e1 nejp\u0159esn\u011bj\u0161\u00ed ur\u010den\u00ed objemu odnesen\u00e9ho materi\u00e1lu ze svahu zasa\u017een\u00e9ho eroz\u00ed s\u00a0vy-\u2028u\u017eit\u00edm bezkontaktn\u00edch metod. Z\u00e1znam r\u00fdhov\u00e9 eroze byl po\u0159\u00edzen pomoc\u00ed UAV syst\u00e9mu a\u00a0k\u00a0vyhodnocen\u00ed erozn\u00edho smyvu slou\u017eil software Agisoft PhotoScan Professional a\u00a0ESRI ArcGIS. D\u00edky pou\u017eit\u00fdm metod\u00e1m a\u00a0technologi\u00edm bylo mo\u017en\u00e9 sestrojit podrobn\u00fd digit\u00e1ln\u00ed model povrchu (DMP), kter\u00fd reflektuje stav monitorovan\u00e9ho svahu v\u00a0dob\u011b m\u011b\u0159en\u00ed. DMP nese po\u017eadovanou informaci o\u00a0morfologii ter\u00e9nu a\u00a0je mo\u017en\u00e9 nad n\u00edm odvodit metriku erozn\u00edch r\u00fdh. P\u0159i aplikaci optim\u00e1ln\u00edho postupu pak lze kvantifikovat mno\u017estv\u00ed erodovan\u00e9ho materi\u00e1lu.<\/p>\n\"\"<\/a>\n

Obr.\u00a02. Denn\u00ed sr\u00e1\u017ekov\u00e9 \u00fahrny a\u00a0lokalizace monitorovan\u00e9ho svahu
\nFig. 2. Daily rainfall totals and localisation of the monitored slope<\/h6>\n

METODIKA SB\u011aRU DAT, P\u0158\u00cdPRAVA A PR\u016eB\u011aH M\u011a\u0158EN\u00cd<\/h2>\n

Pilotn\u00ed lokalita<\/h3>\n

Monitorovan\u00fd svah se nach\u00e1z\u00ed v\u00a0katastr\u00e1ln\u00edm \u00fazem\u00ed obce Vratisl\u00e1vka v\u00a0severn\u00ed \u010d\u00e1sti Jihomoravsk\u00e9ho kraje (obr.\u00a01<\/em>). Jedn\u00e1 se o\u00a0zem\u011bd\u011blsky vyu\u017e\u00edvan\u00fd pozemek obd\u00e9ln\u00edkov\u00e9ho tvaru shora ohrani\u010den\u00fd m\u00edstn\u00ed asfaltovou komunikac\u00ed a\u00a0ve spodn\u00ed \u010d\u00e1sti rem\u00edzkem s\u00a0korytem a\u00a0b\u0159ehy Kozl\u00edho potoka. Dominantn\u00edm p\u016fdn\u00edm typem jsou kambizem\u011b [17]. Celkov\u00e1 plocha svahu zasa\u017een\u00e9ho eroz\u00ed je 9,5\u00a0ha a\u00a0sklon se pohybuje v\u00a0rozmez\u00ed 3\u20137\u00b0, tj. kategorie m\u00edrn\u00fd sklon. M\u00edru eroze umoc\u0148uje faktor nep\u0159eru\u0161en\u00e9 d\u00e9lky svahu, na zem\u011bd\u011blsk\u00e9m pozemku nen\u00ed aplikov\u00e1no \u017e\u00e1dn\u00e9 protierozn\u00ed opat\u0159en\u00ed. Nadmo\u0159sk\u00e1 v\u00fd\u0161ka pozemku se pohybuje v\u00a0intervalu 495\u2013523 m n. m. P\u0159ibli\u017en\u011b polovina svahu je dle monitoringu V\u00daMOP za\u0159azena mezi m\u00edrn\u011b erozn\u011b ohro\u017een\u00e9 p\u016fdy [17].<\/p>\n

Spou\u0161t\u011b\u010dem hodnocen\u00e9 r\u00fdhov\u00e9 eroze byly opakovan\u00e9 p\u0159\u00edvalov\u00e9 sr\u00e1\u017eky ze 7., 13. a\u00a014. \u010dervna 2020 (obr.\u00a02<\/em>), kdy dle dat \u010cesk\u00e9ho hydrometeorologick\u00e9ho \u00fastavu na monitorovan\u00fd zem\u011bd\u011blsk\u00fd pozemek spadlo v\u00a0pr\u016fb\u011bhu 24 hodin 42 mm sr\u00e1\u017eek (7. \u010dervna), 9 mm sr\u00e1\u017eek (13. \u010dervna), respektive 57 mm sr\u00e1\u017eek (14.\u00a0\u010dervna). Prvn\u00ed a\u00a0druh\u00e1 sr\u00e1\u017ekov\u00e1 epizoda nasytila p\u016fdu vodou a\u00a0naru\u0161ila soudr\u017enost a\u00a0integritu svrchn\u00ed vrstvy p\u016fdn\u00edho horizontu. P\u0159\u00edvalov\u00e1 sr\u00e1\u017eka ze dne 14.\u00a0\u010dervna pak spadla na \u010d\u00e1ste\u010dn\u011b rozru\u0161enou p\u016fdn\u00ed strukturu s\u00a0v\u00fdrazn\u011b sn\u00ed\u017eenou infiltra\u010dn\u00ed schopnost\u00ed. P\u016fvodn\u011b plo\u0161n\u00fd odtok se postupn\u011b soust\u0159edil do p\u016fdn\u00edch z\u00e1\u0159ez\u016f, kter\u00e9 se ve st\u0159edn\u00ed \u010d\u00e1sti svahu spojily a\u00a0vytvo\u0159ily znateln\u00e9 erozn\u00ed r\u00fdhy (obr.\u00a03<\/em>). Orn\u00e1 p\u016fda na monitorovan\u00e9m svahu byla v\u00a0dob\u011b n\u00e1stupu p\u0159\u00ed\u010dinn\u00fdch sr\u00e1\u017eek bez vegeta\u010dn\u00edho krytu, co\u017e vedlo k\u00a0akceleraci eroze. V\u00a0dob\u011b m\u011b\u0159en\u00ed zem\u011bd\u011blsk\u00fd pozemek pokr\u00fdval v\u00fddrol. Okoln\u00ed svahy kryt\u00e9 vegetac\u00ed byly bez zn\u00e1mek erozn\u00edho po\u0161kozen\u00ed, a\u00a0to i\u00a0p\u0159es jejich vy\u0161\u0161\u00ed sklonitost.<\/p>\n

M\u011b\u0159ic\u00ed aparatura, softwarov\u00e9 vybaven\u00ed<\/h2>\n

UAV syst\u00e9m<\/h3>\n

Hexakopt\u00e9ra TAROT 690 byla navr\u017eena pro bezkontaktn\u00ed sb\u011br dat pomoc\u00ed leteck\u00e9ho sn\u00edmkov\u00e1n\u00ed (obr.\u00a04<\/em>). Stroj disponuje vysokov\u00fdkonn\u00fdm hardwarov\u00fdm vybaven\u00edm, \u0159\u00edzen m\u016f\u017ee b\u00fdt manu\u00e1ln\u011b i\u00a0automaticky. Kevlarov\u00e1 konstrukce bezpilotn\u00edho letadla je osazena kompletn\u00edm letov\u00fdm \u0159\u00eddic\u00edm syst\u00e9mem DJI A2, kter\u00fd je sestaven z\u00a0inerci\u00e1ln\u00ed jednotky IMU (Inertial Measurement Unit) vybaven\u00e9 barometrick\u00fdm senzorem pro m\u011b\u0159en\u00ed v\u00fd\u0161ky letu, GPS kompasem, signaliza\u010dn\u00edm bluetooth modulem, nap\u00e1jec\u00ed jednotkou a\u00a0regul\u00e1tory pro ovl\u00e1d\u00e1n\u00ed motor\u016f hexakopt\u00e9ry. Na z\u00e1klad\u011b \u00fadaj\u016f z\u00a0IMU a\u00a0GPS \u0159\u00eddic\u00ed jednotka ovl\u00e1d\u00e1 bezpilotn\u00ed letadlo a\u00a0zaji\u0161tuje automatickou stabilizaci n\u00e1klonu a\u00a0v\u00fd\u0161ky letu. Pod hexakopt\u00e9ru je zav\u011b\u0161en gyroskopicky stabilizovan\u00fd v\u00fdkyvn\u00fd dr\u017e\u00e1k, kter\u00fd nese sn\u00edmac\u00ed za\u0159\u00edzen\u00ed\u00a0\u2013 v\u00a0tomto p\u0159\u00edpad\u011b digit\u00e1ln\u00ed fotoapar\u00e1t SONY Alpha 6000. Fotoapar\u00e1t disponuje 24 Mpx CMOS sn\u00edma\u010dem a\u00a0hybridn\u00edm syst\u00e9mem f\u00e1zov\u00e9ho ost\u0159en\u00ed. Je tak schopn\u00fd po\u0159izovat sn\u00edmky ve vysok\u00e9 kvalit\u011b a\u00a0rozli\u0161en\u00ed i\u00a0za slo\u017eit\u00fdch pov\u011btrnostn\u00edch podm\u00ednek [18].<\/p>\n

Trimble R2<\/h3>\n

GNSS p\u0159ij\u00edma\u010d Trimble R2 slou\u017e\u00ed ke kvalitn\u00edmu, rychl\u00e9mu a\u00a0pohodln\u00e9mu sb\u011bru p\u0159esn\u00fdch geoprostorov\u00fdch dat (obr.\u00a04<\/em>). Dvoufrekven\u010dn\u00ed GNSS p\u0159ij\u00edma\u010d vyu\u017e\u00edv\u00e1 nejen sign\u00e1ly ze v\u0161ech st\u00e1vaj\u00edc\u00edch i budovan\u00fdch dru\u017eicov\u00fdch syst\u00e9m\u016f (GPS, GLONASS, Galileo, BeiDou atd.), ale tak\u00e9 extern\u00ed korekce r\u016fzn\u00fdch zdroj\u016f. P\u0159\u00edstroj je vybaven \u010dipem Trimble Maxwell 6 s 220 kan\u00e1ly pro maxim\u00e1ln\u00ed v\u00fdkon, spolehlivost a p\u0159esnost m\u011b\u0159en\u00ed v re\u00e1ln\u00e9m \u010dase. M\u011b\u0159ic\u00ed souprava vyu\u017e\u00edv\u00e1 korekce z\u00a0tradi\u010dn\u00edch RTK a\u00a0VRS s\u00edt\u00ed a\u00a0umo\u017e\u0148uje sb\u011br geoprostorov\u00fdch dat s\u00a0centimetrovou p\u0159esnost\u00ed v\u00a0horizont\u00e1ln\u00edm i\u00a0vertik\u00e1ln\u00edm sm\u011bru. Odoln\u00e1 konstrukce p\u0159ij\u00edma\u010de dovoluje pr\u00e1ci v\u00a0nejn\u00e1ro\u010dn\u011bj\u0161\u00edch podm\u00ednk\u00e1ch a\u00a0d\u00edky technologii Floodlight i\u00a0v\u00a0\u010d\u00e1ste\u010dn\u011b zast\u00edn\u011bn\u00e9m prost\u0159ed\u00ed. Kompaktn\u00ed hardwarov\u00e9 vybaven\u00ed je dopln\u011bno sofistikovan\u00fdm softwarem Trimble ACCESS [19].<\/p>\n\"\"<\/a>\n

Obr.\u00a03. Leteck\u00fd sn\u00edmek monitorovan\u00e9ho svahu
\nFig. 3. Aerial image of the monitored slope<\/h6>\n

Agisoft PhotoScan Professional<\/h3>\n

Pro zpracov\u00e1n\u00ed z\u00edskan\u00fdch leteck\u00fdch sn\u00edmk\u016f byl vyu\u017eit profesion\u00e1ln\u00ed software Agisoft PhotoScan Professional. Jedn\u00e1 se o\u00a0samostatn\u00fd fotogrammetrick\u00fd software ur\u010den\u00fd k\u00a0automatick\u00e9mu generov\u00e1n\u00ed bodov\u00fdch mra\u010den, polygon\u00e1ln\u00edch model\u016f, georeferencovan\u00fdch ortofotomap a\u00a0digit\u00e1ln\u00edch a\u00a0v\u00fd\u0161kov\u00fdch model\u016f ve vysok\u00e9m rozli\u0161en\u00ed. Software vyu\u017e\u00edv\u00e1 modern\u00ed technologie a\u00a0metody digit\u00e1ln\u00ed fotogrammetrie k\u00a0rekonstrukci sn\u00edman\u00fdch 3D povrch\u016f na podklad\u011b z\u00edskan\u00fdch fotografi\u00ed. D\u00edl\u010d\u00ed i\u00a0kone\u010dn\u00fd v\u00fdsledek anal\u00fdz a\u00a0v\u00fdpo\u010dt\u016f lze v\u00a0r\u016fzn\u00fdch form\u00e1tech vyexportovat pro dal\u0161\u00ed zpracov\u00e1n\u00ed v\u00a0geografick\u00fdch informa\u010dn\u00edch syst\u00e9mech (GIS) [13].<\/p>\n

ESRI ArcGIS<\/h3>\n

Pro detailn\u00ed anal\u00fdzu sn\u00edman\u00e9ho \u00fazem\u00ed a\u00a0vytvo\u0159en\u00e9ho 3D modelu povrchu byl pou\u017eit geografick\u00fd informa\u010dn\u00ed syst\u00e9m ESRI ArcGIS ur\u010den\u00fd pro pr\u00e1ci s\u00a0prostorov\u00fdmi daty. V\u00a0rozhran\u00ed ArcMap je mo\u017en\u00e9 geodata podrobn\u011b analyzovat, editovat nebo nad nimi po\u010d\u00edtat matematick\u00e9 funkce [20]. Software slou\u017eil zejm\u00e9na ke stanoven\u00ed celkov\u00e9ho objemu erozn\u00edch r\u00fdh a\u00a0vizualizaci erozn\u00edho po\u0161kozen\u00ed monitorovan\u00e9ho svahu.<\/p>\n\"\"<\/a>\n

Obr.\u00a04. M\u011b\u0159ic\u00ed aparatura pou\u017eit\u00e1 pro sn\u00edmkov\u00e1n\u00ed monitorovan\u00e9ho svahu\u00a0\u2013 UAV a\u00a0GNSS p\u0159ij\u00edma\u010d Trimble R2
\nFig. 4. Measuring devices used for capturing the monitored slope\u00a0\u2013 UAV and Trimble R2 GNSS receiver<\/h6>\n

Ter\u00e9nn\u00ed m\u011b\u0159en\u00ed a\u00a0pou\u017eit\u00e9 metody<\/h2>\n

Vlastn\u00ed zam\u011b\u0159en\u00ed erozn\u00ed situace prob\u011bhlo 23. \u010dervna 2020 v\u00a0p\u0159\u00edzniv\u00fdch pov\u011btrnostn\u00edch podm\u00ednk\u00e1ch. V\u00a0lokalit\u011b p\u0159evl\u00e1dala polojasn\u00e1 obloha, bezv\u011bt\u0159\u00ed a\u00a0teploty mezi 23\u201326 \u00b0C. Tyto podm\u00ednky byly pro pr\u00e1ci s\u00a0bezpilotn\u00edm letadlem ide\u00e1l-\u2028n\u00ed, polojasn\u00e1 obloha p\u0159isp\u011bla ke kvalit\u011b z\u00edskan\u00fdch fotografi\u00ed a\u00a0d\u00edky bezv\u011bt\u0159\u00ed v\u00a0kombinaci s\u00a0teplotou byla maximalizov\u00e1na doba letu stroje.<\/p>\n

Po prvotn\u00edm ter\u00e9nn\u00edm pr\u016fzkumu monitorovan\u00e9ho svahu byla ve vztahu k\u00a0p\u0159edpokl\u00e1dan\u00e9mu letov\u00e9mu \u010dasu (\u010dty\u0159i pln\u011b nabit\u00e9 LiPo akumul\u00e1tory) stanovena v\u00fd\u0161ka letu na 25 m nad ter\u00e9nem. Takto zvolen\u00e1 letov\u00e1 hladina umo\u017enila zam\u011b\u0159it p\u0159ibli\u017en\u011b t\u0159etinu zasa\u017een\u00e9ho \u00fazem\u00ed p\u0159i vysok\u00e9 kvalit\u011b a\u00a0p\u0159esnosti z\u00edskan\u00fdch dat. Pro zam\u011b\u0159en\u00ed cel\u00e9ho svahu p\u0159i dostate\u010dn\u00e9m p\u0159ekryvu jednotliv\u00fdch sn\u00edmk\u016f by letov\u00e1 v\u00fd\u0161ka musela b\u00fdt v\u00edce ne\u017e dvojn\u00e1sobn\u00e1. Po\u0159\u00edzen\u00e9 sn\u00edmky by m\u011bly ni\u017e\u0161\u00ed rozli\u0161en\u00ed, a\u00a0to by vedlo k\u00a0shlazen\u00ed ter\u00e9nu, probl\u00e9mov\u00e9 identifikaci ter\u00e9nn\u00edch hran a\u00a0zlom\u016f\u00a0\u2013 do\u0161lo by ke sn\u00ed\u017een\u00ed korektnosti v\u00fdsledn\u00e9ho digit\u00e1ln\u00edho modelu povrchu.<\/p>\n

P\u0159ed samotn\u00fdm vzletem bezpilotn\u00edho letadla bylo nutn\u00e9 vybavit monitorovan\u00e9 \u00fazem\u00ed pozemn\u00edmi kontroln\u00edmi body (GCP\u00a0\u2013 Ground Control Points). Pomoc\u00ed za\u0159\u00edzen\u00ed Trimble R2 bylo zam\u011b\u0159eno 32 GCP, pro kter\u00e9 byly z\u00edsk\u00e1ny p\u0159esn\u00e9 X, Y, Z\u00a0 sou\u0159adnice, kde Z\u00a0je nadmo\u0159sk\u00e1 v\u00fd\u0161ka ve v\u00fd\u0161kov\u00e9m referen\u010dn\u00edm syst\u00e9mu Balt po vyrovn\u00e1n\u00ed (Bpv). Polohopisn\u00e1 a\u00a0v\u00fd\u0161kopisn\u00e1 datov\u00e1 sada byla v\u00a0dal\u0161\u00edch kroc\u00edch vyu\u017eita k\u00a0ur\u010den\u00ed geometrie a\u00a0georeferencov\u00e1n\u00ed digit\u00e1ln\u00edho modelu povrchu a\u00a0efektivn\u011bj\u0161\u00edmu spojen\u00ed po\u0159\u00edzen\u00fdch sn\u00edmk\u016f.<\/p>\n

V\u00a0prostoru zasa\u017een\u00e9ho svahu bylo mimo n\u00e1hodn\u011b rozm\u00edst\u011bn\u00fdch GCP zam\u011b\u0159eno dal\u0161\u00edch 90 pozemn\u00edch bod\u016f nevstupuj\u00edc\u00edch do v\u00fdpo\u010dtu digit\u00e1ln\u00edho modelu povrchu. X, Y, Z\u00a0sou\u0159adnice pro tyto body byly z\u00edsk\u00e1ny ve t\u0159ech lini\u00edch (v\u00a0horn\u00ed, st\u0159edn\u00ed a\u00a0doln\u00ed \u010d\u00e1sti svahu), kolm\u00fdch na erozn\u00ed r\u00fdhy. C\u00edlem bylo vytvo\u0159it kontroln\u00ed \u0159ez ter\u00e9nem pro n\u00e1sledn\u00e9 porovn\u00e1n\u00ed charakteristik erozn\u00edch r\u00fdh a\u00a0DMP a\u00a0ur\u010den\u00ed odchylek nadmo\u0159sk\u00e9 v\u00fd\u0161ky pro jednotliv\u00e9 body. Zam\u011b\u0159en\u00ed bod\u016f prob\u011bhlo s\u00a0vyu\u017eit\u00edm za\u0159\u00edzen\u00ed Trimble R2 a\u00a0m\u011b\u0159ic\u00edho p\u00e1sma v\u00a0cca dvaceticentimetrov\u00e9m kroku s\u00a0v\u011bt\u0161\u00ed hustotou uvnit\u0159 erozn\u00edch z\u00e1\u0159ez\u016f. Celkov\u00e1 d\u00e9lka ka\u017ed\u00e9ho jednotliv\u00e9ho \u0159ezu ter\u00e9nem se rovnala p\u011bti metr\u016fm.<\/p>\n

Bezkontaktn\u00ed sb\u011br dat pomoc\u00ed bezpilotn\u00edho syst\u00e9mu prob\u011bhl na principu sn\u00edmkov\u00e1n\u00ed eroz\u00ed po\u0161kozen\u00e9ho svahu z\u00a0p\u0159edem ur\u010den\u00e9 letov\u00e9 hladiny. Aby bylo mo\u017en\u00e9 jednotliv\u00e9 sn\u00edmky bez probl\u00e9m\u016f spojit a\u00a0n\u00e1sledn\u011b vytvo\u0159it kompletn\u00ed DMP, p\u0159i po\u0159izov\u00e1n\u00ed fotografi\u00ed bylo nutn\u00e9 dos\u00e1hnout 60\u201380% p\u0159ekryvu. Na z\u00e1klad\u011b v\u00fd\u0161ky letu, pot\u0159ebn\u00e9ho p\u0159ekryvu fotografi\u00ed a\u00a0po\u017eadovan\u00e9 kvality sn\u00edmk\u016f byla ur\u010dena rychlost pohybu hexakopt\u00e9ry TAROT 690 a\u00a0frekvence z\u00e1znamu povrchu pomoc\u00ed digit\u00e1ln\u00edho fotoapar\u00e1tu SONY Alpha 6000 p\u0159ipevn\u011bn\u00e9ho k\u00a0bezpilotn\u00edmu letadlu. Oper\u00e1tor pomoc\u00ed pozemn\u00ed ovl\u00e1dac\u00ed stanice udr\u017eoval rychlost letu hexakopt\u00e9ry v\u00a0rozmez\u00ed 2,5\u20133,0 m\/s\u00a0a\u00a0frekvenci fotografov\u00e1n\u00ed 20 sn\u00edmk\u016f\/min. Pro z\u00e1znam cca 2,6 ha po\u0161kozen\u00e9ho pozemku bylo po\u0159\u00edzeno 1756 fotografi\u00ed za vyu\u017eit\u00ed \u010dty\u0159 pln\u011b nabit\u00fdch LiPo akumul\u00e1tor\u016f, celkov\u00fd letov\u00fd \u010das byl p\u0159ibli\u017en\u011b 47 minut.<\/p>\n

Zpracov\u00e1n\u00ed dat<\/h3>\n

V\u00a0ter\u00e9nu z\u00edskan\u00e1 data v\u00a0podob\u011b leteck\u00fdch sn\u00edmk\u016f a\u00a0zam\u011b\u0159en\u00fdch GCP byla v\u00a0n\u011bkolika kroc\u00edch zpracov\u00e1na v\u00a0softwarov\u00e9m prost\u0159ed\u00ed Agisoft PhotoScan Professional. Specializovan\u00fd software dok\u00e1\u017ee metodou Structure from Motion spojit jednotliv\u00e9 fotografie na z\u00e1klad\u011b identifikace spole\u010dn\u00fdch bod\u016f na p\u0159ekr\u00fdvaj\u00edc\u00edch se sn\u00edmc\u00edch [13]. V\u00fdsledkem je \u0159\u00eddk\u00e9 bodov\u00e9 mra\u010dno a\u00a0informace o\u00a0poloze, sm\u011bru a\u00a0n\u00e1klonu fotoapar\u00e1tu v\u00a0moment\u011b vzniku fotografie. V\u00a0dal\u0161\u00edm kroku jsou pro n\u00e1sledn\u00e9 georeferencov\u00e1n\u00ed modelu lokalizov\u00e1ny GCP se zam\u011b\u0159en\u00fdmi geografick\u00fdmi sou\u0159adnicemi. Na z\u00e1klad\u011b vypo\u010dten\u00fdch a\u00a0vlo\u017een\u00fdch \u00fadaj\u016f je vytvo\u0159eno hust\u00e9 bodov\u00e9 mra\u010dno, kter\u00e9 po zvolen\u00ed sou\u0159adnicov\u00e9ho syst\u00e9mu, v\u00a0tomto p\u0159\u00edpad\u011b S-JTSK K\u0159ov\u00e1k East North, umo\u017e\u0148uje tvorbu digit\u00e1ln\u00edho modelu povrchu a\u00a0georeferencovan\u00e9 ortofotomapy monitorovan\u00e9ho \u00fazem\u00ed. Z\u00edskan\u00e1 data je mo\u017en\u00e9 v\u00a0r\u016fzn\u00fdch form\u00e1tech exportovat a\u00a0vyu\u017e\u00edt pro dal\u0161\u00ed zpracov\u00e1n\u00ed v\u00a0geografick\u00fdch informa\u010dn\u00edch softwarech (obr.\u00a05<\/em>).<\/p>\n\"\"<\/a>\n

Obr.\u00a05. Uk\u00e1zka detailu erozn\u00ed r\u00fdhy v\u00a0softwaru Agisoft PhotoScan Professional
\nFig. 5. Sample of the erosion rill detail in Agisoft PhotoScan Professional software<\/h6>\n

Z\u00e1kladn\u00edm vstupem pro stanoven\u00ed objemu v\u00fdmolov\u00e9 eroze je digit\u00e1ln\u00ed model ter\u00e9nu (DMT). Rozli\u0161en\u00ed modelu p\u0159\u00edmo ovliv\u0148uje v\u00fdsledn\u00fd objem, proto je nutn\u00e9 pou\u017e\u00edt co nejpodrobn\u011bj\u0161\u00ed data. Pro ur\u010den\u00ed m\u00edry eroze byla z\u00a0programu Agisoft PhotoScan Professional vyexportov\u00e1na geodatab\u00e1ze s\u00a0bodovou t\u0159\u00eddou prvk\u016f popisuj\u00edc\u00ed polohopisn\u00e9 a\u00a0v\u00fd\u0161kopisn\u00e9 ur\u010den\u00ed hust\u00e9ho bodov\u00e9ho mra\u010dna v\u00a0sou\u0159adnicov\u00e9m syst\u00e9mu S-JTSK K\u0159ov\u00e1k East North a\u00a0v\u00fd\u0161kov\u00e9m syst\u00e9mu Balt po vyrovn\u00e1n\u00ed (Bpv). Geodata reprezentuj\u00edc\u00ed monitorovan\u00e9 \u00fazem\u00ed byla nahr\u00e1na do prost\u0159ed\u00ed ESRI ArcGIS, kde byla na jejich z\u00e1klad\u011b vytvo\u0159ena nepravideln\u00e1 troj\u00faheln\u00edkov\u00e1 s\u00ed\u0165 (TIN\u00a0\u2013 Trianguled Irregular Network). Vygenerovan\u00fd TIN byl n\u00e1sledn\u011b p\u0159eveden na rastrov\u00fd form\u00e1t o\u00a0velikosti pixelu 0,5 cm. T\u00edm vznikl velmi podrobn\u00fd digit\u00e1ln\u00ed model povrchu se z\u00e1znamem r\u00fdhov\u00e9 eroze odpov\u00eddaj\u00edc\u00ed \u010dasu m\u011b\u0159en\u00ed.<\/p>\n

Stanoven\u00ed objemu erozn\u00edch r\u00fdh<\/h3>\n

Pro samotn\u00fd v\u00fdpo\u010det objemu erozn\u00edch r\u00fdh bylo nutn\u00e9 vymodelovat co nejp\u0159esn\u011bj\u0161\u00ed teoretick\u00fd pr\u016fb\u011bh ter\u00e9nu p\u0159ed n\u00e1stupem eroze. Ke konstrukci p\u016fvodn\u00edho ter\u00e9nu byl vyu\u017eit vytvo\u0159en\u00fd detailn\u00ed 3D model po\u0161kozen\u00e9ho svahu v\u00a0kombinaci s\u00a0georeferencovanou ortofotomapou ve vysok\u00e9m rozli\u0161en\u00ed. Nad zm\u00edn\u011bn\u00fdmi datov\u00fdmi zdroji byla manu\u00e1ln\u011b identifikov\u00e1na m\u00edsta nezasa\u017een\u00e1 eroz\u00ed, sedimentac\u00ed a\u00a0nekryt\u00e1 vegetac\u00ed. Vzniklo mra\u010dno diskr\u00e9tn\u00edch bod\u016f, kter\u00e9mu byly z\u00a0podkladov\u00e9ho digit\u00e1ln\u00edho modelu povrchu definov\u00e1ny \u00fadaje o\u00a0nadmo\u0159sk\u00e9 v\u00fd\u0161ce s\u00a0vyu\u017eit\u00edm funkce Extract Values to Points. Ze 4\u00a0762 nov\u011b vznikl\u00fdch diskr\u00e9tn\u00edch bod\u016f byla vytvo\u0159ena nepravideln\u00e1 troj\u00faheln\u00edkov\u00e1 s\u00ed\u0165 reprezentuj\u00edc\u00ed p\u016fvodn\u00ed povrch p\u016fdy p\u0159ed n\u00e1stupem eroze. Aby ho bylo mo\u017en\u00e9 l\u00e9pe porovn\u00e1vat a\u00a0prov\u00e1d\u011bt matematick\u00e9 operace, byl podklad TIN na rastr o\u00a0stejn\u00e9 velikosti pixelu jako DMP zaznamenan\u00e9 eroze. Odfiltrov\u00e1n\u00edm nadprahov\u00fdch hodnot byl p\u016fvodn\u00ed DMP o\u010di\u0161t\u011bn o\u00a0vegeta\u010dn\u00ed kryt\u00a0\u2013 vznikl DMT. Rozd\u00edlovou anal\u00fdzou rastru DMT eroze a\u00a0DMT teoretick\u00e9ho p\u016fvodn\u00edho povrchu vznikla rastrov\u00e1 vrst-\u2028va zn\u00e1zor\u0148uj\u00edc\u00ed rozsah a\u00a0hloubku erozn\u00edch z\u00e1\u0159ez\u016f v\u00a0absolutn\u00edch hodnot\u00e1ch. Objem erodovan\u00e9ho materi\u00e1lu na monitorovan\u00e9m pozemku byl pak odvozen jako n\u00e1sobek plochy a\u00a0pr\u016fm\u011brn\u00e9 hodnoty rozd\u00edlov\u00e9ho rastru vypo\u010dten\u00e9 pomoc\u00ed zon\u00e1ln\u00ed statistiky (obr.\u00a06<\/em>).<\/p>\n\"\"<\/a>\n

Obr.\u00a06. Sch\u00e9ma v\u00fdpo\u010dtu objemu erozn\u00ed r\u00fdhy
\nFig. 6. Scheme of the calculation of the erosion rill volume<\/h6>\n

V\u00ddSLEDKY<\/h2>\n

Na z\u00e1klad\u011b rozd\u00edlov\u00e9 anal\u00fdzy zam\u011b\u0159en\u00e9ho DMT a\u00a0vytvo\u0159en\u00e9ho teoretick\u00e9ho p\u016fvodn\u00edho povrchu byl pro monitorovan\u00fd svah stanoven objem eroze na 4\u00a0274\u00a0m3<\/sup>. V\u00a0erozn\u00edch r\u00fdh\u00e1ch a\u00a0v\u00fdrazn\u00fdch z\u00e1\u0159ezech do\u0161lo p\u0159i p\u0159epo\u010dtu na celou plochu pozemku ke smyvu p\u016fdn\u00edho horizontu o\u00a0pr\u016fm\u011brn\u00e9 v\u00fd\u0161ce 4,5 cm. Hloubka nejv\u00fdrazn\u011bj\u0161\u00edch erozn\u00edch r\u00fdh p\u0159esahovala 85 cm p\u0159i \u0161\u00ed\u0159ce t\u00e9m\u011b\u0159 110 cm. Pro v\u00e1hov\u00e9 vyj\u00e1d\u0159en\u00ed odnesen\u00e9ho materi\u00e1lu byla pou\u017eita objemov\u00e1 hmotnost such\u00e9 p\u016fdy, kterou pro svrchn\u00ed vrstvu p\u016fdy Honz\u00edk a\u00a0kol. uv\u00e1d\u00ed v\u00a0rozmez\u00ed 1,2\u20131,5\u00a0t\/m3<\/sup> [21]. P\u0159i zvolen\u00e9 objemov\u00e9 hmotnosti na horn\u00ed hranici zm\u00edn\u011bn\u00e9ho intervalu se odnos materi\u00e1lu ze zem\u011bd\u011blsk\u00e9ho pozemku rovn\u00e1 675 t\/ha. V \u010cesk\u00e9 republice je p\u0159\u00edpustn\u00e1 ztr\u00e1ta p\u016fdy pro st\u0159edn\u011b hlubok\u00e9 (30\u201360\u2008cm) a hlubok\u00e9 p\u016fdy (nad 60\u2008cm) stanovena na 4 t\/ha\/rok [6]. V pr\u016fb\u011bhu jednoho t\u00fddne tak p\u0159\u00edvalov\u00e9 sr\u00e1\u017eky zp\u016fsobily na monitorovan\u00e9m svahu 169kr\u00e1t vy\u0161\u0161\u00ed odnos materi\u00e1lu, ne\u017e je stanoven\u00e1 p\u0159\u00edpustn\u00e1 ztr\u00e1ta. Ortofotosn\u00edmek a vizualizace pr\u016fb\u011bhu erozn\u00edch r\u00fdh ve vybran\u00e9 \u010d\u00e1sti svahu jsou zn\u00e1zorn\u011bny na obr.\u00a07<\/em>.<\/p>\n\"\"<\/a>\n

 <\/p>\n

Obr.\u00a07. Vizualizace hloubek erozn\u00edch r\u00fdh ve vybran\u00e9 \u010d\u00e1sti po\u0161kozen\u00e9ho svahu (vlevo), ortofotomapa stejn\u00e9 \u010d\u00e1sti svahu (vpravo)
\nFig. 7. Visualization of the depths of the erosion rills in a selected part of the damaged slope (on the left); ortophotomap of the same part of the slope (on the right)<\/h6>\n

Z\u00a0v\u00fd\u0161e uveden\u00e9ho je patrn\u00e9, \u017ee zem\u011bd\u011blsk\u00fd pozemek byl vystaven devastuj\u00edc\u00edm erozn\u00edm \u00fa\u010dink\u016fm a\u00a0byla trvale po\u0161kozena \u00farodnost zasa\u017een\u00e9ho p\u016fdn\u00edho horizontu. Vlastn\u00ed eroze byla nav\u00edc zes\u00edlena chyb\u011bj\u00edc\u00edm vegeta\u010dn\u00edm krytem v\u00a0dob\u011b p\u0159\u00edchodu p\u0159\u00edvalov\u00fdch sr\u00e1\u017eek.<\/p>\n

Pro verifikaci vytvo\u0159en\u00e9ho digit\u00e1ln\u00edho modelu povrchu bylo v\u00a0ter\u00e9nu nez\u00e1visle zam\u011b\u0159eno 90 pozemn\u00edch bod\u016f. Tyto body um\u00edst\u011bn\u00e9 ve form\u011b t\u0159\u00ed p\u0159\u00ed\u010dn\u00fdch profil\u016f slou\u017eily k\u00a0anal\u00fdze v\u00fd\u0161kov\u00e9 p\u0159esnosti 3D modelu a\u00a0nevstupovaly do v\u00fdpo\u010dtu DMP. Pr\u016fb\u011bh ter\u00e9nu vymodelovan\u00fd metodami digit\u00e1ln\u00ed fotogrammetrie s\u00a0vyzna\u010den\u00fdmi, nez\u00e1visle zam\u011b\u0159en\u00fdmi body a\u00a0ortofotosn\u00edmkem vybran\u00e9ho p\u0159\u00ed\u010dn\u00e9ho profilu ukazuje obr.\u00a08<\/em>.<\/p>\n\"\"<\/a>\n

Obr.\u00a08. Pr\u016fb\u011bh ter\u00e9nu vymodelovan\u00fd metodami digit\u00e1ln\u00ed fotogrammetrie (UAV) s\u00a0vyzna\u010den\u00fdmi, nez\u00e1visle zam\u011b\u0159en\u00fdmi body (GNSS Trimble) a\u00a0ortofotosn\u00edmkem vybran\u00e9ho p\u0159\u00ed\u010dn\u00e9ho profilu
\nFig. 8. Terrain course created by digital photogrammetry methods (UAV) with marked independently captured points (Trimble GNSS) and ortophotograph of a\u00a0selected cross section<\/h6>\n

Z\u00edskan\u00e1 data byla importov\u00e1na do pracovn\u00edho\u00a0prost\u0159ed\u00ed ESRI ArcGIS, v\u00a0n\u011bm\u017e byly vypo\u010dteny odchylky hodnot nadmo\u0159sk\u00fdch v\u00fd\u0161ek pro vymodelovan\u00fd ter\u00e9n a\u00a0zam\u011b\u0159en\u00e9 body. V\u00fdsledky jsou pro p\u0159ehlednost vyobrazeny ve form\u011b krabicov\u00e9ho grafu na obr.\u00a09<\/em>. V\u00a0n\u011bm jsou zam\u011b\u0159en\u00e9 body rozd\u011bleny do jednotliv\u00fdch \u0159ez\u016f ter\u00e9nem a\u00a0d\u00e1le pak dle um\u00edst\u011bn\u00ed bodu uvnit\u0159\/vn\u011b erozn\u00ed r\u00fdhy.<\/p>\n

Medi\u00e1n, horn\u00ed a\u00a0doln\u00ed kvartil odchylek nadmo\u0159sk\u00fdch v\u00fd\u0161ek se v\u00a0zam\u011b\u0159en\u00fdch \u0159ezech pohyboval v\u00a0t\u011bsn\u00e9 bl\u00edzkosti nuly. Minim\u00e1ln\u00ed a\u00a0maxim\u00e1ln\u00ed hodnoty odchylek pak v\u00a0\u0159\u00e1dech desetin centimetru. V\u00fdjimkou byl jeden zam\u011b\u0159en\u00fd bod v\u00a0bl\u00edzkosti kolm\u00e9 st\u011bny velmi hlubok\u00e9ho erozn\u00edho z\u00e1\u0159ezu. Zde m\u016f\u017ee m\u011b\u0159en\u00ed ovliv\u0148ovat zast\u00edn\u011bn\u00ed p\u0159i sn\u00edmkov\u00e1n\u00ed a\u00a0mo\u017en\u00e1 tendence shlazovat vymodelovan\u00fd ter\u00e9n v\u00a0m\u00edstech s\u00a0ni\u017e\u0161\u00ed hustotou pokryt\u00ed daty.<\/p>\n

Zam\u011b\u0159en\u00e9 body uvnit\u0159 erozn\u00edch z\u00e1\u0159ez\u016f ukazuj\u00ed sp\u00ed\u0161e na nadhodnocov\u00e1n\u00ed nadmo\u0159sk\u00e9 v\u00fd\u0161ky vymodelovan\u00e9ho ter\u00e9nu. Body z\u00edskan\u00e9 GNSS m\u011b\u0159en\u00edm uvnit\u0159 erozn\u00edch r\u00fdh maj\u00ed odchylku hodnot medi\u00e1nu, horn\u00edho a\u00a0doln\u00edho kvartilu v\u00a0bl\u00edzkosti \u00farovn\u011b -1 cm. Odchylky nadmo\u0159sk\u00e9 v\u00fd\u0161ky zam\u011b\u0159en\u00fdch bod\u016f v\u00a0eroz\u00ed nezasa\u017een\u00e9 plo\u0161e jsou oproti vymodelovan\u00e9mu ter\u00e9nu minim\u00e1ln\u00ed a\u00a0pohybuj\u00ed se v\u00a0m\u00edrn\u011b\u00a0kladn\u00fdch \u010d\u00edslech.<\/p>\n

P\u0159\u00ed\u010dn\u00e9 profily i\u00a0krabicov\u00fd graf ukazuj\u00ed silnou korelaci mezi ter\u00e9nem vymodelovan\u00fdm metodami digit\u00e1ln\u00ed fotogrammetrie a\u00a0body zam\u011b\u0159en\u00fdmi pomoc\u00ed GNSS p\u0159ij\u00edma\u010de Trimble R2. Zji\u0161t\u011bn\u00e9 odchylky nadmo\u0159sk\u00fdch v\u00fd\u0161ek se u\u00a0v\u011bt\u0161iny bod\u016f pohybuj\u00ed v\u00a0prvn\u00edch desetin\u00e1ch centimetru. V\u00fdrazn\u011bj\u0161\u00ed rozd\u00edly v\u00a0nadmo\u0159sk\u00fdch v\u00fd\u0161k\u00e1ch jsou patrn\u00e9 pouze u\u00a0bod\u016f lokalizovan\u00fdch do erozn\u00edch z\u00e1\u0159ez\u016f. Zde vymodelovan\u00fd ter\u00e9n ukazuje vy\u0161\u0161\u00ed nadmo\u0159sk\u00e9 v\u00fd\u0161ky, a\u00a0m\u016f\u017ee tak podhodnocovat celkovou m\u00edru erodovan\u00e9ho materi\u00e1lu. Nejvy\u0161\u0161\u00ed hodnoty odchylek byly zaznamen\u00e1ny v\u00a0bl\u00edzkosti kolm\u00fdch st\u011bn erozn\u00edch z\u00e1\u0159ez\u016f\u00a0\u2013 3D model ter\u00e9nu m\u00e1 tendenci shlazovat kolm\u00e9 st\u011bny erozn\u00edch r\u00fdh. V\u00a0p\u0159\u00edpad\u011b hlubok\u00fdch erozn\u00edch z\u00e1\u0159ez\u016f se z\u00e1porn\u00e9 odchylky hodnot zam\u011b\u0159en\u00fdch bod\u016f pohybovaly okolo -2 cm, v\u00a0jednom p\u0159\u00edpad\u011b z\u00e1porn\u00e1 odchylka p\u0159ekro\u010dila -5 cm. Na podhodnocov\u00e1n\u00ed objemu eroze a\u00a0shlazov\u00e1n\u00ed kolm\u00fdch st\u011bn erozn\u00edch r\u00fdh u\u00a0fotogrammetrick\u00fdch metod ve sv\u00e9 pr\u00e1ci upozor\u0148uje i\u00a0Glendell a\u00a0kol. [22]. T\u011bsnost zam\u011b\u0159en\u00fdch bod\u016f a\u00a0vymodelovan\u00e9ho ter\u00e9nu je vzhledem k\u00a0velikosti zaznamenan\u00e9 plochy velmi vysok\u00e1. 3D model ter\u00e9nu lze ozna\u010dit za reprezentativn\u00ed a\u00a0stanoven\u00fd objem odnesen\u00e9ho materi\u00e1lu se d\u00e1 pova\u017eovat za odpov\u00eddaj\u00edc\u00ed realit\u011b, p\u0159\u00edpadn\u011b lehce podhodnocen\u00fd.<\/p>\n

P\u0159\u00edmo v\u00a0ter\u00e9nu bylo vizu\u00e1ln\u011b patrn\u00e9 ukl\u00e1d\u00e1n\u00ed erodovan\u00e9ho materi\u00e1lu p\u0159i pat\u011b svahu a\u00a0v\u00a0\u00fadolnici a\u00a0koryt\u011b Kozl\u00edho potoka, odkud bude postupn\u011b un\u00e1\u0161en n\u00ed\u017ee po toku. Ve st\u0159edn\u011bdob\u00e9m \u010dasov\u00e9m horizontu bude erodovan\u00e1 zemina sedimentovat v\u00a0soustav\u011b mal\u00fdch vodn\u00edch n\u00e1dr\u017e\u00ed (MVN) na Kozl\u00edm potoku v\u00a0katastru obce \u017dd\u00e1rec (soustava MVN je od monitorovan\u00e9ho svahu vzd\u00e1lena necel\u00fd jeden kilometr). Zan\u00e1\u0161en\u00ed vodn\u00edch n\u00e1dr\u017e\u00ed je mo\u017en\u00e9 ozna\u010dit za dal\u0161\u00ed z\u00a0negativn\u00edch d\u016fsledk\u016f eroze zp\u016fsoben\u00e9 kombinac\u00ed nespr\u00e1vn\u00e9ho vyu\u017e\u00edv\u00e1n\u00ed \u00fazem\u00ed a\u00a0p\u0159\u00ed\u010dinn\u00fdch p\u0159\u00edvalov\u00fdch sr\u00e1\u017eek.<\/p>\n\"\"<\/a>\n

Obr.\u00a09. Krabicov\u00fd graf Z\u00a0odchylek GNSS m\u011b\u0159en\u00ed a\u00a03D modelu ter\u00e9nu z\u00edskan\u00e9ho metodami digit\u00e1ln\u00ed fotogrammetrie
\nFig. 9. Box plot from the deviations of GNSS measurement and 3D terrain model obtained by digital photogrammetry methods<\/h6>\n

DISKUZE<\/h2>\n

Pro rekonstrukci re\u00e1ln\u00e9ho digit\u00e1ln\u00edho modelu ter\u00e9nu slou\u017e\u00edc\u00edho k\u00a0dal\u0161\u00edm anal\u00fdz\u00e1m je zcela z\u00e1sadn\u00ed z\u00edskat co nejp\u0159esn\u011bj\u0161\u00ed geodata v\u00a0maxim\u00e1ln\u00edm rozsahu a\u00a0hustot\u011b pokr\u00fdvaj\u00edc\u00ed monitorovan\u00e9 \u00fazem\u00ed. Chyby mohou vznikat p\u0159i pr\u00e1ci v\u00a0ter\u00e9nu i\u00a0p\u0159i n\u00e1sledn\u00e9m zpracov\u00e1n\u00ed dat (postprocessing). P\u0159i sb\u011bru dat s\u00a0vyu\u017eit\u00edm bezpilotn\u00edho letadla je nutn\u00e9 db\u00e1t na pe\u010dliv\u00e9 zam\u011b\u0159en\u00ed GCP, dostate\u010dn\u00fd p\u0159ekryv po\u0159\u00edzen\u00fdch sn\u00edmk\u016f a\u00a0vhodn\u011b zvolenou letovou hladinu. Vy\u0161\u0161\u00ed letov\u00e1 v\u00fd\u0161ka zpravidla sni\u017euje rozli\u0161en\u00ed a\u00a0korektnost kone\u010dn\u00e9ho v\u00fdsledku [23]. S\u00a0ni\u017e\u0161\u00ed letovou v\u00fd\u0161kou je zachyceno v\u00edce ter\u00e9nn\u00edch detail\u016f, naproti tomu kles\u00e1 rozloha zaznamenan\u00e9ho \u00fazem\u00ed a\u00a0zvy\u0161uje se \u010dasov\u00e1 n\u00e1ro\u010dnost. Stanoven\u00e1 hodnota objemu odnesen\u00e9ho materi\u00e1lu je ovlivn\u011bna tak\u00e9 kvalitou rekonstrukce stavu p\u016fvodn\u00edho povrchu p\u0159ed erozn\u00ed ud\u00e1lost\u00ed. Zde mohou vznikat chyby v\u00a0nespr\u00e1vn\u00e9 lokalizaci bod\u016f do erozn\u00edch prohlubn\u00ed, sedimenta\u010dn\u00edch lavic nebo jinak zm\u011bn\u011bn\u00fdch p\u016fvodn\u00edch \u010d\u00e1st\u00ed svahu. V\u00a0potaz je nutn\u00e9 vz\u00edt i\u00a0zanedb\u00e1n\u00ed plo\u0161n\u00e9 eroze. Stanoven\u00fd objem erodovan\u00e9ho materi\u00e1lu je tedy sp\u00ed\u0161e m\u00edrn\u011b podhodnocen\u00fdm odhadem ne\u017e definitivn\u00edm ur\u010den\u00edm skute\u010dnosti.<\/p>\n

Skokov\u00fd v\u00fdvoj v\u00a0technologi\u00edch bezpilotn\u00edch leteck\u00fdch syst\u00e9m\u016f v\u00a0kombinaci se sni\u017eov\u00e1n\u00edm jejich finan\u010dn\u00ed n\u00e1ro\u010dnosti vede k\u00a0rychlej\u0161\u00edmu, p\u0159esn\u011bj\u0161\u00edmu a\u00a0bezpe\u010dn\u011bj\u0161\u00edmu sb\u011bru dat. V\u00a0sou\u010dasn\u00e9 dob\u011b existuj\u00ed UAV vybaven\u00e9 RTK (z\u00a0anglick\u00e9ho Real-Time Kinematic) syst\u00e9mem a\u00a0vysoce precizn\u00edmi GNSS moduly pro satelitn\u00ed pozicov\u00e1n\u00ed. Takto vybaven\u00fd stroj dok\u00e1\u017ee sn\u00edmat velmi p\u0159esn\u00e1 RTK data a\u00a0sledovat a\u00a0mapovat trajektorii letu dronu s\u00a0centimetrovou p\u0159esnost\u00ed. Ke ka\u017ed\u00e9mu po\u0159\u00edzen\u00e9mu sn\u00edmku jsou k\u00a0dispozici p\u0159esn\u00e1 metadata, kter\u00e1 usnad\u0148uj\u00ed n\u00e1sledn\u00e9 zpracov\u00e1n\u00ed a\u00a0zvy\u0161uj\u00ed korektnost kone\u010dn\u00e9ho v\u00fdsledku. Odpad\u00e1 nutnost pou\u017eit\u00ed GCP a\u00a0sni\u017euje se \u010dasov\u00e1 n\u00e1ro\u010dnost pr\u00e1ce v\u00a0ter\u00e9nu i\u00a0mo\u017en\u00e1 chybovost v\u00fdsledn\u00e9ho digit\u00e1ln\u00edho modelu povrchu. Po\u0159izovac\u00ed cena UAV stroj\u016f vybaven\u00fdch RTK modulem a\u00a0redundantn\u00edm GNSS syst\u00e9mem se pohybuje okolo 200 tis\u00edc K\u010d a\u00a0v\u00a0kone\u010dn\u00e9m d\u016fsledku takto vybaven\u00e9 drony zefektiv\u0148uj\u00ed pr\u00e1ci v\u00a0ter\u00e9nu a\u00a0zp\u0159es\u0148uj\u00ed a\u00a0zrychluj\u00ed i\u00a0n\u00e1sledn\u00fd postprocessing nasb\u00edran\u00fdch dat [24].<\/p>\n

Z\u00c1V\u011aR<\/h2>\n

Proveden\u00e1 monitorovac\u00ed mise je sou\u010d\u00e1st\u00ed dlouhodob\u00e9ho sledov\u00e1n\u00ed erozn\u00edch ud\u00e1lost\u00ed a\u00a0dal\u0161\u00edch n\u00e1sledk\u016f p\u0159\u00edvalov\u00fdch sr\u00e1\u017eek brn\u011bnskou pobo\u010dkou V\u00daV TGM s\u00a0c\u00edlem z\u00edskat datovou z\u00e1kladnu pro detailn\u00ed anal\u00fdzy p\u0159\u00ed\u010din a\u00a0d\u016fsledk\u016f spojen\u00fdch s\u00a0p\u0159\u00edvalov\u00fdmi sr\u00e1\u017ekami. Prim\u00e1rn\u00edm c\u00edlem mapov\u00e1n\u00ed erozn\u00edho po\u0161kozen\u00ed bylo stanoven\u00ed co nejp\u0159esn\u011bj\u0161\u00edho odhadu objemu eroze s\u00a0vyu\u017eit\u00edm bezkontaktn\u00edch metod m\u011b\u0159en\u00ed. Dal\u0161\u00edm c\u00edlem bylo ov\u011b\u0159en\u00ed korektnosti 3D modelu konstruovan\u00e9ho pomoc\u00ed leteck\u00e9ho sn\u00edmkov\u00e1n\u00ed a\u00a0metod Structure from Motion v\u00a0porovn\u00e1n\u00ed s\u00a0nez\u00e1visle zam\u011b\u0159en\u00fdmi kontroln\u00edmi body.<\/p>\n

Z\u00edskan\u00e1 data ukazuj\u00ed v\u00fdraznou korelaci, odchylka nam\u011b\u0159en\u00fdch nadmo\u0159sk\u00fdch v\u00fd\u0161ek se pohybuje v\u00a0\u0159\u00e1dech desetin centimetru. Potvrzen\u00e1 vysok\u00e1 p\u0159esnost bezkontaktn\u00edch metod p\u0159i modelov\u00e1n\u00ed ter\u00e9nu je p\u0159edur\u010duje k\u00a0\u0161ir\u0161\u00edmu vyu\u017eit\u00ed nejen v\u00a0oblasti hodnocen\u00ed erozn\u00edch situac\u00ed. P\u0159i vhodn\u011b zvolen\u00e9 letov\u00e9 hladin\u011b je dosa\u017eeno vysok\u00e9 hustoty pokryt\u00ed daty, a\u00a0t\u00edm je minimalizov\u00e1n n\u00e1sledn\u00fd vliv interpolace do spojit\u00e9ho digit\u00e1ln\u00edho modelu ter\u00e9nu. Dal\u0161\u00ed v\u00fdhodou je vysok\u00e1 operabilita bezpilotn\u00edch letadel. Za nev\u00fdhody lze ozna\u010dit silnou z\u00e1vislost UAV na pov\u011btrnostn\u00edch podm\u00ednk\u00e1ch, subjektivitu p\u0159i stanoven\u00ed p\u016fvodn\u00edho ter\u00e9nu p\u0159ed p\u0159\u00edvalovou sr\u00e1\u017ekou a\u00a0mo\u017en\u00e9 shlazov\u00e1n\u00ed kolm\u00fdch st\u011bn erozn\u00edch z\u00e1\u0159ez\u016f, \u010d\u00edm\u017e doch\u00e1z\u00ed k\u00a0podhodnocov\u00e1n\u00ed skute\u010dn\u00e9ho objemu erozn\u00edho odnosu.<\/p>\n

Data nasb\u00edran\u00e1 UAV syst\u00e9mem a\u00a0zpracovan\u00e1 metodami digit\u00e1ln\u00ed fotogrammetrie poskytuj\u00ed vysoce kvalitn\u00ed zdroj informac\u00ed vyu\u017eiteln\u00fd nejen v\u00a0geografick\u00fdch informa\u010dn\u00edch syst\u00e9mech. Popsan\u00e9 postupy odhadu m\u00edry eroze jsou jednou z\u00a0mo\u017enost\u00ed vyu\u017eit\u00ed takto z\u00edskan\u00fdch geodat. P\u0159edstaven\u00e9 postupy lze ozna\u010dit za efektivn\u00ed a\u00a0v\u00fdsledn\u00e9 hodnoty objemu eroze za re\u00e1ln\u00e9.<\/p>\n

Pod\u011bkov\u00e1n\u00ed<\/h3>\n

P\u0159\u00edsp\u011bvek vznikl za podpory projektu CZ.07.1.02\/0.0\/0.0\/16_040\/0000382 \u201eRekrea\u010dn\u00ed potenci\u00e1l vody v\u00a0Praze\u00a0\u2013 stav a\u00a0v\u00fdhledy\u201c \u0159e\u0161en\u00e9ho v\u00a0r\u00e1mci opera\u010dn\u00edho programu Praha\u00a0\u2013 p\u00f3l r\u016fstu.<\/em><\/p>\n

P\u0159\u00edsp\u011bvek pro\u0161el lektorsk\u00fdm \u0159\u00edzen\u00edm.<\/strong><\/p>\n","protected":false},"excerpt":{"rendered":"

S\u00a0p\u0159ekotn\u00fdm\u00a0v\u00fdvojem v\u00a0oblasti geoinforma\u010dn\u00edch a\u00a0komunika\u010dn\u00edch technologi\u00ed se objevuj\u00ed nov\u00e9 mo\u017enosti mapov\u00e1n\u00ed a\u00a0kvantifikace erozn\u00edch proces\u016f zp\u016fsoben\u00fdch p\u0159\u00edvalov\u00fdmi sr\u00e1\u017ekami. Pro z\u00edsk\u00e1n\u00ed relevantn\u00edch v\u00fdsledk\u016f je d\u016fle\u017eit\u00e1 kvalita, hustota a\u00a0rozsah vstupn\u00edch dat. Rychl\u00fd a\u00a0pom\u011brn\u011b levn\u00fd sb\u011br p\u0159esn\u00fdch geodat umo\u017e\u0148uj\u00ed bezpilotn\u00ed leteck\u00e9 prost\u0159edky (UAV\u00a0\u2013 z\u00a0anglick\u00e9ho Unmanned Aerial Vehicle). S\u00a0vyu\u017eit\u00edm metod digit\u00e1ln\u00ed fotogrammetrie je mo\u017en\u00e9 z\u00a0po\u0159\u00edzen\u00fdch leteck\u00fdch sn\u00edmk\u016f rekonstruovat morfologii ter\u00e9nu odpov\u00eddaj\u00edc\u00ed \u010dasu m\u011b\u0159en\u00ed.<\/p>\n","protected":false},"author":8,"featured_media":12256,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[2,86,93],"tags":[2558,2555,2557,2560,2556,2559],"coauthors":[1423],"class_list":["post-13145","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-from-the-world-of-water-management","category-hydraulics-hydrology-and-hydrogeology","category-two-articles","tag-aerial-photogrammetry","tag-contactless-mapping","tag-erosion-quantification","tag-torrential-rainfall","tag-unmanned-aerial-system","tag-water-erosion"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/posts\/13145","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/comments?post=13145"}],"version-history":[{"count":3,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/posts\/13145\/revisions"}],"predecessor-version":[{"id":30626,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/posts\/13145\/revisions\/30626"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/media\/12256"}],"wp:attachment":[{"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/media?parent=13145"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/categories?post=13145"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/tags?post=13145"},{"taxonomy":"author","embeddable":true,"href":"https:\/\/www.vtei.cz\/en\/wp-json\/wp\/v2\/coauthors?post=13145"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}